Microgrids: Overview and guidelines for practical implementations and operation

Applied Energy 258 (2020) 114039

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Microgrids: Overview and guidelines for practical implementations and operation

T

A. Cagnanoa, E. De Tugliea, , P. Mancarellab ⁎

a b

Department of Electrical and Information Engineering, Polytechnic University of Bari, Via Re David, 200, 70125 Bari, Italy Department of Electrical and Electronic Engineering, the University of Melbourne, Parkville, Victoria 3010, Australia

HIGHLIGHTS

the main design features of different microgrids around the world. • Identify paper explores the main issues arising from the development of a microgrid. • This attempt to define potential solutions to overcome the main technical issues. • An • It defines guidelines for practical implementation and operation of microgrids. ARTICLE INFO

ABSTRACT

Keywords: Microgrids Islanded operation Non-bumpless islanding Bumpless islanding Resynchronization Blackout On-grid black-start Off-grid black-start

A microgrid is a small portion of a power distribution system with distributed generators along with energy storage devices and controllable loads which can give rise to a self-sufficient energy system. From the utility grid side, a microgrid is seen as an equivalent generator that is able to seamlessly disconnect and operate autonomously once a fault affects the main grid. The design, installation and operation of such systems lead to dealing with a number of technical and operational challenges including control, protection and infrastructure requirements. To help designers and researchers address these challenges and draw potential recommendations for practical microgrid implementations, in this paper a review of the main design features of existing microgrids is undertaken, also in light of the experience gained during the realization of the Prince Lab microgrid at Polytechnic University of Bari, Italy. The main control functions required to guarantee an economic, reliable and secure operation of a microgrid are also reviewed. Finally, key practical guidelines for monitoring, operation and implementation of microgrids are provided.

1. Introduction Microgrids offer a viable solution for integrating Distributed Energy Resources (DERs), including in particular variable and unpredictable renewable energy sources, low-voltage and medium-voltage into distribution networks. Basically, a microgrid can be defined as an electrically bounded area of the distribution network that aggregates local distributed generation sources along with energy storage devices and controllable loads so as to form a self-sufficient energy system [1,2]. Therefore, if properly managed, it can act as a single controllable entity operated in parallel with the utility grid or in islanded mode. Although the benefits that microgrids can bring to end users are numerous, their integration into current distribution networks is still hindered by several issues mainly related to their operation, protection and control [3,4]. As consequence, intensive economic supporting programs have ⁎

been undertaken during the last years by several countries to finance research projects aimed at conducting studies on these topics [5]. As a result, several microgrids demonstration projects have been built and investigated all over the world [6,7]. Most of the existing microgrids are related to isolated or grid-connected systems. In particular, isolated microgrids can offer a reliable energy supply in small remote areas where the development or the expansion of power grids turns out to be technically and/or economically unfeasible. Since these systems cannot take advantage of the main grid support, they provide a useful test bed for the development of proper control functions able to ensure reliable supply of electricity [8]. Grid-connected microgrids are largely adopted to support the integration of DG units and, in particular, of renewable energy sources (RES) in distribution networks [9]. Although the research results obtained within various test cases have addressed many of the technical challenges that obstacle the deployment of microgrids,

Corresponding author at: via Re David, 200, 70125 Bari, Italy. E-mail address: [email protected] (E. De Tuglie).

https://doi.org/10.1016/j.apenergy.2019.114039 Received 16 May 2019; Received in revised form 14 October 2019; Accepted 16 October 2019 0306-2619/ © 2019 Elsevier Ltd. All rights reserved.

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

their practical application is still at the initial stage. This is mainly due to the lack of the necessary expertise required to practitioners for designing a microgrid, as well as the lack of sufficient knowledge of the main technical challenges encountered during its integration into the existing distribution networks. To cover this gap of knowledge and draw potential recommendations for modern microgrid implementations, in this paper a review of the main design factors of current microgrids is performed, also based on the experience gained during the realization of the Prince Lab experimental microgrid located at the Polytechnic University of Bari [10]. This study focuses on the design and implementation issues that have been faced in the course of this project and the adopted solutions, with particular emphasis on control functions, monitoring and optimal management of the microgrid. Moreover, on the basis of this experience, a comprehensive literature review aimed at outlining the main control functions required to guarantee an economic, reliable and secure operation of a microgrid is undertaken. Finally, key practical guidelines for monitoring, operation and implementation of microgrids are provided.

to regulate the frequency and the voltages of an isolated system. The Bornholm microgrid can be managed as an island as well as in parallel with the utility grid. Nonetheless, it is classified as an isolated microgrid because it is operated in the off-grid mode for most of the time. Thanks to a synchrocheck relay, it provides a powerful test bed for developing resynchronization control strategies. Moreover, it is also adopted to set up off-grid black start procedures. Also the CERTS testbed [11–13] and the CESI RICERCA DER Test Facility [5] are included into the class of droop controlled microgrids. These have been built with the aim to investigate on voltage and frequency stability as well as on new protection schemes and design requirements for storage devices of autonomous microgrids. The CERTS microgrid also represents a test-bed facility for developing and testing new control strategies for ensuring a seamless transition from on-grid to off-grid operation and vice-versa. The CESI RICERCA DER microgrid is equipped with a centralized control system that allows changing the system configuration so that several grid topologies can be studied [14]. Also the DeMoTec microgrid at the Kassel’s Institute for Electrical Energy Technology in Germany [15] can be operated in both off- and ongrid modes, but needs to be de-energized during the transition phase. Differently from the above cited microgrids, it is based on a masterslave control scheme where the master can be chosen among three different generators. Most of the reviewed microgrids have the ability to switch from grid-connected operation to islanded operation following a nonplanned event or by means of a planned transition. The formation of an emergency island can be considered as the worst contingency for the survival of a microgrid. Therefore, new control strategies for ensuring the stability of the island were developed at the Bronsbergen Holiday Park in the Netherlands [16]. This system is a low voltage radial distribution network which is connected to the main grid through a central energy storage device that is operated as an Uninterruptible Power Supply (UPS) acting as the master for the isolated microgrid, while all other dispatchable microsources behave as slaves. Under the same scheme, but adopting a CHP at the point of connection, the residential microgrid of Am Steinweg in Stutensee is operated in the master-slave control mode [17]. If technical or economic reasons suggest operating the microgrid in off-grid mode, a planned islanding can be considered as in the case of the NTUA, the Hydro Quebec and the BC hydro master-slave controlled

2. Review of existing microgrids and their applications The aim of this section is to perform a review of the main design features of existing microgrids in order to provide useful designing and managing insights. In particular, in Table 1 microgrids have been classified according to their ability to be operated in all possible states and transitions. The survey of major demonstration projects points out that there is no structured knowledge in designing of such systems. In fact, depending on research objectives, microgrids have been built with several architectures and control structures, including microgrids that can be operated in on-grid mode only and in both on- and off-grid modes. In the latter case, the system voltages and frequency are controlled by adopting master-slave controllers or droop controlled converters. Among droop-controlled microgrids, the Kythnos Island microgrid [5] is well known, which was built with the aim of developing centralized and decentralized control strategies for autonomous systems. On the other hand, the reliability and economic management of an isolated microgrid is the main aim of the Huatacondo microgrid, whereas the Continuon’s MV/LV microgrid is used to develop control strategies able Table 1 Summary of the applications of selected major microgrid projects.* Microgrids

Operating modes On-Grid

Off-Grid

Alert State

Transitions

Emergency State

Blackout

Bumpless Islanding

Emergecy Islanding

Synchr

Blackstart On-grid

KERI Microgrid [67] Korea- KEPRI Microgrid [8–10,67,68] Huatacondo [69] Continuon’s MV/LV facility [68] Am Steinweg [17] Eigg Island Project [68,70] Kythnos [5,68,70] Smart Polygeneration Microgrid [9] DeMoTec [15] Bornholm Island [68,70] NTUA [15] Hydro Quebec [18,71,72] BC Hydro [18,71,72] University of Manchester Microgrid [5] Bronsbergen Holiday Park [5,16] CESI Ricerca DER [5,16] CERTS [11–13] Sendai Project [68,70] Prince Lab [10]

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

Off-grid

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

✓

✓

✓ ✓ ✓ ✓ ✓

✓

* From processing data contained in [68,70]. 2

✓ ✓ ✓

✓

✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

✓

✓

✓ ✓

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

microgrids. In those systems new control strategies have been developed in order to reduce the tie-line power flow by managing internal production before the islanding. For this purpose, the Hydro Quebec microgrid regulates a single thermal power plant acting as an isochronous controller when a planned island is formed [18]. The BC Hydro microgrid has a bit more complicated structure, with two locally controlled hydro power generators that take care of the voltages and frequency at their connection points. The required information is exchanged among the two generators through a simple telephone line. Most of the above mentioned microgrids were custom-designed for laboratory development projects. For this reason, they often involve customized components to give high level of flexibility to test new control ideas. However, microgrids are particularly attractive for their ability to promote the integration of DER into distribution networks. Hence, the Prince Lab microgrid at the Polytechnic University of Bari (Italy) was developed to provide technical and operational recommendations for ensuring interoperability, reliability, resilience and security in more realistic cases and based on commercially available distributed energy sources. Relevant innovations include adjustments to the electrical connections of its internal DER so as to ensure their integration into a microgrid structure and the development of islanded and interconnected operating procedures allowing flexibility to seamlessly transition from grid-connected to isolated operation and vice-versa. Moreover, the open architecture of its SCADA system allow implementation of different and new control functions and testing of various devices.

with the connection standards, these devices are usually equipped with protective relays enabling them to isolate the microgrid following mains outages. Depending on the microgrid application and its control structure (Master/Slave or Droop Control) these connection devices can also be combined with additional devices. In droop-controlled microgrids these additional devices are mainly characterized by power converters, whereas in master-slave controlled microgrids they could be CHP systems [17] or Energy Storage systems [5,16], that are operated as an Uninterruptible Power Supply (UPS) acting as the master for the isolated microgrid. The major drawback of this latter architecture is that in case of failure of the master unit there is no chance for the microgrid to survive. Besides the radial structure, AC microgrids can be looped, meshed or mixed. However, these topologies are usually not implemented since they may impact on grid operation and commonly adopted protection strategies. Although the integration of AC microgrids into distribution networks is relatively easy and not much expensive because it only involves a partial expansion or upgrade of the existing electricity infrastructure, the intensive use of AC/DC converters pose several challenges on the protection, communication and operation of such microgrids. Therefore, DC microgrids are recently emerging as a possible solution in the case of only few isolated DC devices that need to be connected into ex-novo networks. In this configuration, most of the DER are connected through DC/DC or AC/DC power electronic converters to one or more DC buses with a regulated voltage. These microgrids are usually connected to the AC utility grid through an AC/DC converter that is programmed to allow microgrid islanding and resynchronization [8,19,20]. Although, DC microgrids offer several advantages with regard to AC ones, such microgrids are not fully exploited because the vast majority of devices currently in use are fed in AC. In fact, the need to integrate in a same microgrid both AC and DC devices leads to AC/DC hybrid microgrids where the advantages of both AC and DC microgrids are combined together. However, serious stability problems may derive from perturbations coming from the DC subsystem. For this reason, these microgrids have not received great attention and very few hybrid microgrids have been developed. Based on the above considerations, below the focus will be on analysing the main technical and operational challenges of AC microgrids.

3. Review of microgrid’s architecture, protection, communication, management and control features The aim of this section is to provide a comprehensive literature review related to microgrids by outlining the main issues and challenges being encountered during their deployment. In line with this objective, the different structure and topology of microgrids were firstly examined. After that, a review of the main studies recently carried out for microgrid protection has been undertaken by outlining the main challenges that must be tackled to reliably protect microgrids. Then, an overview of the current communication technologies has been performed so as to outline the impact that it may have on the microgrid operation. Finally, it has been performed a comprehensive literature review of different control strategies capable to manage the microgrid in both grid-connected and in islanded mode, under normal as well as fault conditions.

3.2. Protection schemes The design of an adequate protection scheme is another important challenge that must be tackled when developing a microgrid. In fact, differently from traditional distribution networks, fault currents in microgrids may drastically change depending upon the location of the fault. This is mainly due to the presence of inverter-based DER and to the fact that microgrids can dynamically change their architecture as well as their operating mode. Fixed relay settings usually adopted in conventional protection schemes of distribution systems seem to be inappropriate for microgrid protections, especially for those microgrids that can be operated in both grid-connected and islanded modes. In fact, in these microgrids, minimum values of the short circuit currents in the islanded mode may not be revealed since they could differ too much from those when connected to the main distribution grid. Therefore, new protection schemes should be designed. There is not yet any well-defined general solution for microgrid protection due to the large variety of factors affecting the design of a microgrid, such as microgrid type and topology, voltage operating level, geographical extension, DER technology and location, DER interface relays and their coordination, neutral grounding, operation mode, and reliability requirements [21–23]. To find the best protection scheme to meet multiple design requirements several options can be adopted as reported in [21]. A summary with remarks of the main protection schemes suggested in the

3.1. Microgrid topology and structure The survey of existing microgrids performed in the previous section points out that there is no settled structure for microgrids. Therefore, one of the main challenges that needs to be faced in designing a microgrid is to select the appropriate structure and topology. Since this choice is strongly dependent on microgrid applications and scale, many topologies and structures can be adopted. In order to guide the designers in choosing inherent safe and robust design solutions, a comprehensive literature review was undertaken with the aim to investigate the main topologies and architectural structures of existing microgrids. The main results are summarized in Table 2. The review shows that AC microgrids are the most used configuration due to their ability to directly integrate renewable energy sources already connected to current distribution networks with minimum modifications of the electric infrastructure. These microgrids are typically characterized by a radial topology as this configuration has the minimum impact on grid’s operation as well as of on the protection schemes usually adopted in distribution networks. In this configuration, DER are usually connected to a common busbar that is connected to a MV distribution network through circuit breakers [18], contactors [5,15] or static switches [10,12]. Furthermore, in agreement 3

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 2 Summary of the main microgrid architectures and topologies.** Microgrids

Type AC

Huatacondo [69] NTUA [15] BC Hydro [18,71,72] University of Manchester Microgrid [5] Eigg Island Project [68,70] Sendai Project [68,70] Prince Lab [10] Am Steinweg [17] Kythnos [5,68,70] Smart Polygeneration Microgrid [9] DeMoTec [15] Bronsbergen Holiday Park [5,16] CERTS [11–13] Bornholm Island [68,70] CESI Ricerca DER [5,16]

NUAA [8] INER [8]

Structure

Comparisons

DC

Advantages

✓ ✓ ✓ ✓

Radial Radial Radial Radial

✓ ✓ ✓ ✓ ✓ ✓

Radial Radial Radial Mesh Mesh Mesh

✓ ✓

Mesh Mesh

✓ ✓

✓

Mesh Mesh Radial

✓ ✓

Radial Radial

✓ ✓

Remarks

– Require a protection scheme that is capable to be self-adaptive at any change in microgrid architecture and operating mode

Disadvantages

– Easy integration into existing distribution networks – Suitable for large-scale microgrids – Reduced impact on existing protection schemes for distribution networks

– High power quality problems – High penetration of power electronic converters

– Few power quality problems; – Useful for smart homes and buildings – Absence of electromagnetic interferences;

– Requires voltage stabilization; – High number of DC/AC or DC/DC inverters – Difficulty in fualt detection due to the absence of zero crossing of current – High implementation costs – Complete revision of the existing protection schemes – Lower reliability – Complex control schemes

– Impact Protective Devices coordination; – Affect fault current magnitudes and directions – Need a complete redesigning or upgrade of the existing protection techniques

– Easy integration – Reduced number of electronic power converters – No need for synchronization of DG and storage

** From processing data contained in [8,73–76].

technical literature is reported in Table 3. As can be noted several options were suggested in the technical literature for designing adequate protection schemes for microgrids. Adaptive protection schemes seem to be more effective with regard to any change occurring in the microgrid architecture and its operating mode. Adaptive protection schemes can be classified into two categories: centralized and decentralized. The first ones are based on a centralized architecture, structured around a radial communication network. They usually adopt the standard IEC 61850, characterized by high reliability and low latencies. However, if communication is lost due to equipment failure or sabotage, the protection scheme will fail due to the absence of a back-up communication network [23]. Moreover, depending on the microgrid complexity, these schemes may reveal inefficiencies related to computational efforts and communication delays. To overcome these issues, decentralized adaptive protection schemes have been suggested. However, while these schemes are able to provide high performance, their implementation requires a complete upgrade of the existing protection schemes with high integration costs. Among others, differential protection schemes are usually preferred since they only involve partial upgrade or an extension of protection schemes usually adopted in distribution networks.

microgrid resilience and reliability requirements. The performed review shows that several types of communication networks are usually adopted, for example, wired, optical fibers, wireless, Global System for Mobile Communications (GSM), GPS, XLM and combination of them [24]. The review of the main design features for communication networks in existing microgrids is synthesized in Table 4. With regards to communication protocols, the trend is to adopt the same protocol of the main controller. Several proprietary and open standards communication protocols exist. Among the open standards, the most used are the IEC 61850, the Distributed Network Protocol 3.0 (DNP 3.0), the Modbus, the Profibus, Wi-Fi, and the TCP/IP [25–28]. Among them, the IEC 61850 can be considered as the most attractive for microgrid applications due to its high speed, high reliability, and high security levels, especially against cyberattacks. Nonetheless, its practical application is still hindered by the lack of commercially available microcontrollers supporting the IEC 61850 protocol. A possible solution could be the adoption of protocol converters, but it would likely introduce several bottlenecks. As a result, the development of this protocol should pass through the commercialization of microcontrollers and SCADA systems adopting the IEC 61850. At this stage, most of the existing microgrids adopt protocols widely used for the industrial sector. Among them, the Modbus is the most used due to its simplicity [29]. It is a master/slave protocol which can be transmitted over different physical networks such as the Ethernet TCP/IP, the RS 485 and RS 232 [30]. However, this communication protocol is not so effective when transmitting a huge amount of data from/to network. In fact, in this case high latencies may be experienced that make it unsuitable for some mission-critical applications such as corrective or emergency control. If long delays in the communication system cannot be tolerated, a hard-wired network could be superposed to other networks, such as in the case of the PrInCE LAB microgrid [10].

3.3. Communication networks The design of the communication network can be considered a crucial topic for the development of microgrids, aimed at establishing communication among several microgrid components in order to monitor and control in the real-time the overall microgrid. Achieving reliable communication among the microgrid devices is not trivial due to the great variety of factors affecting its design such as microgrid topology, operation mode, geographical extension, component communication interfaces, technology of inverter-based DER (master/slave or droop-control), protection schemes, control requirements, and 4

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 3 Summary of the more relevant studies on the protection schemes for microgrids. Operation modes On-Grid

Communication link

Ref.

– Non-directional overcurrent protection with high performances in terms of selectivity, speed and reliability. Bidirectional current may lead to a loss of selectivity – Conventional Overcurrent relays with definite time grading requiring low installation costs since partial upgrade of existing protection schemes is required – Intelligent Agent Based protections based on high-speed communications – Voltage-based protection scheme that is capable to achieve high selectivity for different faults, except for the symmetrical one. It shows high sensitivity related to switching transients – Centralized adaptive protection based on high speed and secure communicative protocols such as IEC 61850 is able to ensure fast, selective and reliable protection for microgrids in both grid-connected and isolated modes. However, their implementation requires a complete revamping of the existing protection schemes with consequent increase of the installation costs. Furthermore, since usually there is not a back-up protection, a fault in the communication link may lead to a failure in the protection trip mechanism with a consequent reduction of its reliability – Decentralized adaptive protection based on high speed and secure communicative protocols such as IEC 61850 are higher reliability than that based on a centralized philosophy – Differential protections are easier to be integrated into existing protection schemes because they involve a partial upgrade or extension of existing protection schemes. They are not capable to discriminate short-circuit levels in islanded mode – Distance protections are able to discriminate different fault currents even if their magnitude is small as the case of isolated microgrids. However, malfunctions of these protection schemes can arise due to harmonic or transient behaviour of currents

[77]

Island

✓

NO

✓

NO

✓

Remarks

✓

YES YES

✓

✓

YES

✓

✓

YES

✓

✓

YES

✓

✓

NO

3.4. Microgrid operation

[21]- [23] [78] [79,80] [22,23,81]

[22,23] [82–84] [85–86]

summary of the more relevant economic dispatch techniques with respect to validation methods and remarks. As can be noted, only few economic dispatch methodologies have been implemented and validated into an actual microgrid.

This subsection conducts a comprehensive literature review of the main control strategies proposed for microgrid operation with the aim to outline the minimum core-control functions to be implemented in the SCADA/EMS so as to achieve good levels of robustness, resilience and security in all operating states and transitions. From a systematic point of view, all possible operating states and transitions of a microgrid can be schematically represented as shown in Fig. 1.

3.4.1.3. Voltage regulation. If the microgrid is large enough, voltage regulation may be required in order to avoid the nuisance of voltage relays tripping and cascade events. In Table 7 a set of candidate control strategies for the voltage control is summarized. As can be noted, depending on the microgrid size, one can choose to use decentralized controllers rather than centralized ones, and to implement control methods aimed at improving the microgrid power quality rather than that aimed at flattening the voltage profile.

3.4.1. On-Grid operation In the grid-connected mode, a microgrid lies in a normal state for most of the time. In this operating state, the controllable energy sources are scheduled at the lowest operating cost by taking into account storages, nonprogrammable energy sources, and the forecasted load. Like large power systems, typical control functions are the unit commitment, the economic dispatch, the voltage regulation and the reserve management.

3.4.1.4. Reserve management. For those microgrids that can be operated in islanded/isolated mode, the Energy Management System (EMS) will also take care of optimizing the required amount of spinning reserve. To comply with this requirement, several control strategies have been developed, ranging from deterministic [32–34] to probabilistic [35] approaches. These strategies include the search strategy based on MILP [35] and other less “classical”, such as Orthogonal Array (OA) method [32].

3.4.1.1. Unit commitment problem. The Unit Commitment problem is aimed at determining the schedule of available generating units at every time interval, by taking into account control signals coming from the distribution network for ancillary service provision, market prices, forecasts of loads, and renewable power production. For this purpose, several control strategies have been developed that can be broadly divided into two categories: deterministic (D) and stochastic (S) methods. Both classes of control strategies are based on an optimization problem aimed at minimizing a cost function subject to a set of functional and technical constraints. The objectives of these optimization problems are different and a collection of them is presented in [31]. According to this paper, they may be production costs, fuel costs, maintenance costs, startup and shut-down costs, degradation costs and costs for the energy purchased from the utility grid. To find the optimal solution of these problems several optimization techniques can be adopted. In Table 5 a summary of these techniques is listed.

3.4.2. Transition from on-grid to off-grid mode The on-grid to off-grid operation transition of a microgrid can be performed following a contingency (Emergency Islanding) or by a planned operation. In this case, the EMS must be capable to manage the microgrid in order to ensure a seamless islanding transition. To comply with this need, a suitable control mechanism needs to be activated. 3.4.2.1. Non-bumpless transition – Emergency islanding. The sudden transition to emergency islanding represents a severe perturbation that could be critical for the survival of a microgrid. In fact, the abrupt absence of the active and reactive power support coming from the distribution network must be promptly replaced by internal devices. An adequate control mechanism is needed to guarantee the survival of the microgrid. Table 8 reports a summary of selected control strategies that can increase the microgrid resilience according to the specified technical and operational criteria. Recommended criteria to select the most suitable control strategies are based on the microgrid electrical structure and its control architecture.

3.4.1.2. Economic dispatch. The output of the Unit Commitment (UC) is passed to a Real-time Control Layer where it is processed by an economic dispatch algorithm to create the new schedules of the energy production of all microgrid’s sources for the next time step. Table 6 presents a 5

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 4 Summary of the main existing microgrid communication networks. Microgrids

Communications

Advantages

Media support

Protocol

Optical fiber

N.A.

Huatacondo [69]

N.A.

Modbus TCP/IP

Am Steinweg [17]

N.A.

Mobus TCP/IP

Kythnos [5,68,70]

Power line

N.A.

Smart Polygeneration Microgrid [9]

N.A.

IEC 61850

DeMoTec [15]

N.A. Optical fiber

Combination of ethernet and XLM-RPC protocol N.A.

N.A.

XLM

BC Hydro [18,71–72]

Telephone

N.A.

University of Manchester Microgrid [5]

Power line

N.A.

Bronsbergen Holiday Park [5,16] CESI Ricerca DER [5,16]

GSM

N.A.

Combination of LAN, wireless and power line

Ethernet

CERTS [11–13]

N.A.

Modbus TCP/IP

Sendai Project [68,70]

GPS

N.A.

Combination of LAN, fiber optic, copper

Modbus TCP/IP

Korea- KEPRI Microgrid [8–10,67,68]

Bornholm Island [68,70] NTUA [15]

Prince Lab [10]

– High-speed communication – Low latency time – High reliability – Easy to implement – Low installation costs – Supported by different communication links – Easy to implement – Low installation costs – Supported by different communication links – High speed communication – Adopt existing electical network – Low installation costs – High data transfer rate – High reliability – High-speed communication – High security levels especially against cyber attacks – High interoperability – Improved reliability – Enhanced security level – High-speed communication – Low latency time – High reliability – High operational flexibility – High readability – Easy to be integrated – Adopt existing electical network – Low installation costs – High data transfer rate – High reliability – Cost effective – Involve partial upgrade and expansion of existing IT networks – High reliability – Easy to implement – Low installation costs – Supported by different communication links – Easy to integrate – Low installation costs – Global accessibility – Easy to implement – Low installation costs – Supported by different communication links

Disadvantages

– Cover short distances – High latency times

– High network delays; – Low security levels against cyber attacks – Minimum security levels – Data attenuation – Great amount of noise – Low redundancy level – High implementation costs – Need to change or upgrade of both communication interfaces of already installed components and the existing IT networks – High computational costs – High installation costs – Cover short distances – High load of communication challanges – – – – –

Low reliability High implementation costs Minimum security levels Data attenuation Great amount of noise

– Limited trasmission bandwidith – Low secuirty levels – High installation costs – High network delays – Low security levels against cyber attacks – Reduced accuracy – Low reliability due to battery life – Low security levels – Low privacy – High network delays – Low security levels against cyber attacks – Involve partial upgrade and expansion of existing IT networks

(**) From information in [8,73–76,87–88].

3.4.2.2. Bumpless transition. At times economic reasons can suggest changing the microgrid operation from grid-connected to island mode. In this case, a pre-planned bumpless transition can be activated to minimize perturbations deriving from the opening of the tie-line switch. The basic idea of this control philosophy is to obtain a smooth transition by moving the system operating point in such a way that the power flowing through the tie-line is close to zero. To achieve this condition, several control strategies have been developed. Based on its simplicity, security and speed, the emphasis here is often posed on strategies that exploit the droop concept. However, their implementation would require the adoption of ad-hoc devices since most of the inverters available on the market are not equipped with droop controllers. For this reason, master-slave-based control strategies may be more appropriate. Among others, a very effective one appears to be the one proposed in [36], as it does not involve droop-controlled devices. The major shortcoming of this

methodology, however, is that in case of failure of the master unit there is no chance for the microgrid to survive. However, if the maximum regulation capacity of the master unit is not sufficient to balance the system during the islanding transition, also this procedure can fail. Therefore, a preventive control able to restore adequate reserve margins of the master unit needs to be implemented, by sharing the control burden among all available sources. Table 9 shows a summary of the selected bumpless islanding procedures. 3.4.3. Island state When operating as an island the microgrid should be in a stable state as when on-grid, thus also sharing very similar operation objectives. 3.4.3.1. Economic dispatch. The absence of the distribution grid support 6

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Fig. 1. The EMS control strategies related with the microgrid operating states [10].

in providing ancillary services for system security is critical here. For this reason, several research efforts have come up with a wide range of control solutions. Table 10 presents the summary of the economic dispatch strategies that can be implemented in the real-time control layer in order to guarantee a safe operation of an isolated microgrid.

Among the possible options, droop-controlled microgrids typically mimic the secondary and tertiary control of bulk power systems. With this aim, a hierarchical control strategy able to perform frequency control in multiple timescales is developed in [37]. In [38] an enhanced hierarchical control methodology based on virtual impedances is adopted to compensate voltage unbalances arising within a microgrid. In order to reduce the communication burden due to the centralized architecture of these controllers, in [39,40] decentralized secondary controls are suggested. With these control schemes, control actions are locally evaluated by every DG unit local controller. A cooperative multi-

3.4.3.2. Voltage and frequency regulation. Frequency and voltage stability problems could be exacerbated for islanded microgrids due to the structural weakness of such systems. It is therefore required that the real-time controller of the microgrid takes care of any possible system unbalance. Table 5 Summary of the more relevant studies on unit commitment for microgrids. Problem types

Validation level

D

S

S

✓ ✓ ✓ ✓ ✓ ✓

(1)

✓ ✓ ✓ ✓ ✓

✓ ✓

✓

✓ ✓

✓ ✓ ✓ (1) (2)

✓ ✓ ✓ ✓

E

Remarks

(2)

✓ ✓

Model Predictive Control (MPC) control scheme based on the receding horizon technique [89] Security-constrained UC considering Lagrangian Relaxation method (robust MIP) [90] Optimal power flow control with a receding-horizon approach, considering reactive power [91] Focusing on linearized UC problem, adopting Monte Carlo method [92] A robust Mixed-Integer Linear Programming (MILP) method, considering charge/discharge feature of energy storage devices. Provide start-up and shut-down decision for each unit [93] Multi-objective optimization problem considering power quality and energy savings [94] A large-scale Mixed-Integer Programming (MIP) with coupling constraints. Sub-problem for each microgrid appliance with dual decomposition [95] Sub-hourly schedule of dispatchable microgrid sources considering uncertainty of wind power output (robust MILP). Modified Benders decomposition method. A Markovian approach is used [96–97] Bi-level programming optimization model (robust MILP). Upper level maximizes profits, Lower level simulates the day-ahead market clearing process [98] Dynamic Multistage Stochastic Unit Commitment, considering the minimum up/down times of DG units [99] Hybrid harmony search algorithm, considering spinning reserve constraints [100] A chance constrained scheduling problem [101,102] Two-Stage Stochastic UC Problem (robust MILP). First stage: UC costs, Second stage: worst-case economic dispatch [103–104]

S: Simulation level. E: Experimental level. 7

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 6 Summary of the more relevant studies on economic dispatch of a microgrid. Validation level (1)

S

E

Remarks

(2)

✓

✓

Focusing only on the conventional generators fuel cost, based on a linearized economic dispatch problem [105] Bi-level multi-objective evolutionary algorithm based on Stackelberg game approach, considering costs/incomes resulting from energy purchase/sale from the upstream network [106] Real-time scheduling adopting a quasi-newton method [107] Optimal microgrid scheduling with Incentive-Based Demand Response Program [108] Multi-objective optimization based on a robust MILP method considering charge/discharge feature of energy storage devices and battery degradation costs [109,110] Multi-objective optimization considering operating and maintenance costs of energy storage systems. Solutions obtained by Dynamic programming techniques and Mixed-Integer Non-Linear Programming (MINLP) [111,112] A Linear MIP technique for a less-time expensive technique [113,114] Two stages short-term scheduling based on robust MILP. First stage: a day-ahead scheduler, Second stage: intra-day scheduler (15 min ahead) [115].

✓ ✓ ✓ ✓ ✓ ✓ (1) (2)

S: Simulation level E: Experimental level

agent system for power sharing management is suggested in [41,42] in which each DG local controller shares the needed information (frequency, voltage magnitude, active and reactive powers) with only its neighbor. In order to enhance the microgrid reliability and to eliminate the need of central processing, in [43,44] fully distributed adaptive control methods are developed for both voltage and frequency restoration. However, although these controllers provide good performance, their implementation into actual microgrids could be impractical since commercially available inverter-based devices are not droop-controlled inverters. For microgrids adopting master/slave controls, the master unit will normally take care of frequency and voltage regulation. This unit needs to be designed with a rated power capable to cover any reasonable perturbation occurring on the system. More sophisticated microgrids adopt a cooperative control strategy, as proposed for example in [45] and [14]. A routine able to automatically share the control burden among all sources in accordance with their reserve margins and their ramp-up and ramp-down rate limits is designed in [46,47]. A summary of selected control strategies for voltage and frequency regulation is reported in Table 11.

3.4.4. Alert state Starting from the normal or secure operating states, particular dispatching policies or unexpected events can move the system operating point into the Alert State, characterized by dangerously reduced security margins. Therefore, if a particular disturbance occurs, the microgrid would enter in the Emergency State and then cascading failures and blackouts could occur. In order to minimize the risk of blackouts, preventive control actions need to be evaluated and activated, following Static Security Assessment (SSA) studies. Table 13 reports a summary of the selected preventive control strategies that could be implemented. Depending on the microgrid operating state, on-grid preventive controls may be distinguished from off-grid ones. In particular, the former focus on the tie-line power flow that must be less than the Total Transmission Capacity (TTC). This requires the ability of the local EMS to manage the tie-line power flow so that congestion can be prevented, or at least corrected. 3.4.5. Emergency state Unpredicted events may move the system operating point into the Emergency State. This state is characterized by the violation of some limits, leading to the trip of some devices or even to system collapse. In order to avoid this catastrophic event, corrective control actions need to be quickly activated. For this purpose, the load/generation/storage shedding strategies seems to be the most practical way. A summary of the most suitable emergency control strategies is given in Table 14. Note that all these strategies focus on steady-state operating points, ignoring the dynamic behaviour of the microgrid. This simplification may be too rough in some cases, especially during the islanding process. In fact, if only evaluated for steady-state conditions, the amount of power to be shed may not be sufficient for system stabilization since load/generation/storage shedding brings about additional perturbation

3.4.3.3. Reserve management. Generation outages or forecasting errors of loads and renewable energy sources may lead to security problems for isolated microgrids. In order to prevent such problems, reserve management strategies need to be implemented. It is worth noting that only few papers have focused on master-slave-controlled microgrids, as the trend is to adopt a master unit with regulation capacity greater than the expected maximum power unbalance. However, this approach generally leads to both high investment costs and high operation costs. Moreover, the failure of the master unit would imply the system collapse (see Table 12). Table 7 Summary of main Microgrid voltage control strategies. Control types Centralized

Decentralized

✓ ✓

✓

Objective Function System Losses

Voltage Profile

✓ ✓ ✓ ✓ ✓

✓

Optimization algorithms

✓ ✓ ✓ ✓ ✓

– Genetic Algorithms [116]; Particle Swarm Optimization (PSO) [117]; Sensitivity Theory applied to the Lyapunov function [118] – Sensitivity theory [119] – Distribution network partitioning; Sensitivity Theory applied to the Lyapunov function [120]. – Sensitivity theory [121]; Multi-agent system [122] – Voltage-based droop control strategy [123] – Secondary Voltage control based on PI controller which is able to regulate the voltage at the Point of Commonc Coupling (PCC) of the microgrid [124] – Virtual impedance method capable to minimize voltage and frequency fluctuations allows to enhance microgrid stability [125]

8

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 8 Control strategies candidates for non-bumpless islanding transition of a microgrid. Control types Droop

Master-Slave

Validation level (1)

S

✓

✓

✓ ✓ ✓

✓

✓ ✓

✓ ✓

✓ ✓ ✓

✓ ✓ ✓

✓

(1) (2)

Ref

– Fix unbalances between the internal power production and the consumption while holding the voltage variable in its allowable range – On-line virtual impedance adjustment approach – Secondary control level – Decentralized energy management scheme based on multi-agent scheme – Reduced equivalent model of the grid – Hierarchical control system – Multi-level control. The First layer based on a Multi-Agent System (MAS) is responsible for power balance and economic dispatch; the Second Layer actuates the power commands coming from the MAS – Multi-objective optimization problem based on a predictive control approach – An improved model predictive current control scheme is employed to enhance the microgrid reliability. – A model predictive control of dual mode Z-Source Inverters (ZSIs) – An Uninterruptible Power Supply (UPS) system to connect the microgrid to the main grid – Control technique to rapidly switch the master unit (UPS) from P-Q to V/f control mode – A simple circuit for integrating devices usually operated as UPS system into a microgrid – Multi-master microgrid – Token ring procedure to pass the master function to other available master units – Retrieve the regulation capacity of the master unit by means a cooperative control based on Lyapunov Function

[126]

E

✓ ✓

✓

✓ ✓

Remarks

(2)

✓ ✓

✓

[127,128] [129] [130] [131] [132] [133–134] [135] [136] [36,137] [57] [138]

S: Simulation level. E: Experimental level.

Table 9 Recommended bumpless islanding procedures to increase microgrid resilience. Control types Droop

Master-Slave

Validation level Mix

(1)

✓ ✓

(3)

– Multi-droop control strategy based on Economic operation criterion [139] – Two-layer power sharing is employed for microgrid islanding (observer-based) and communication failure (auxiliary tracking droop-control) [140] – Adaptive defence mechanism for power re-dispatch and under frequency load shedding [66] – Two-layer optimal management is applied for managing the active power (first layer) and the reactive power (second layer) outputs of the master unit [141] – DER able to act as master units are operated through droop control [142]

✓ ✓

✓ ✓

(2)

E

Remarks

(3)

✓ ✓

✓

(1)

S

(2)

✓

Mix: combination of droop and master-slave control. S: Simulation level. E: Experimental level.

Table 10 Summary of the more relevant studies on the economic dispatch of an isolated microgrid. Control types Droop ✓ ✓

(1) (2)

Master-Slave

✓ ✓ ✓ ✓ ✓

Validation level (1)

S

✓ ✓ ✓ ✓ ✓ ✓

E

Remarks

(2)

✓

– – – – – – –

Self-adjusting droop characteristics of DG units for optimal dispatch [143] Frequency-IC droop scheme considering uncertainty [125,144] Dynamic scheduling considering the possibility to pass the isochronous generator function among all available sources [145] Two-stage optimal dispatch is employed for day-ahead (first stage) and real-time (second stage) scheduling [146] Focusing on linearized economic dispatch problem [147] Rolling-horizon optimal dispatch strategy (robust MILP) [148] Two different time scales are applied for extended real-time (15 min ahead) and online scheduling [149].

S: Simulation level. E: Experimental level.

on the already perturbed system. This condition may be worsened by the low-inertia conditions that characterize (small) microgrids. Therefore, Dynamic Security Assessment (DSA) needs to be preliminarily performed to determine whether such corrective actions pose a dynamic threat to the network stability. However, very few papers focus on the DSA and control of microgrids. Among them, paper [48] suggests to adopt the rate of change of the frequency drop to evaluate the percentage of the load to be shed in real time. In [49], load shedding and generation emergency dispatch control actions are evaluated in the real time through an optimization problem involving dynamic

models of all microgrid components. However, dynamic security assessments need accurate mathematical models of the microgrid components. The technical literature is rich in references on modeling of these components. Papers [50–52] developed full-order models of all microgrid components including their controllers. 3.4.6. Synchronization transition If technical or economic circumstances suggest the reconnection of the microgrid to the main grid, the operator can start the reconnection procedure by following one of the re-synchronization procedures 9

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 11 Control strategy candidates for isolated microgrid voltage and frequency regulation. Control types Droop

Master-Slave

✓ ✓ ✓ ✓ ✓

(2)

(1)

S

✓ ✓ ✓ ✓ ✓ ✓

(1)

Validation level E

Remarks

(2)

✓

– Two different timescales are applied for dynamic stability (65 ms) and steady-state stability (5 sec) controls [37] – Selective virtual impedance loop for optimal power sharing to compensate voltage unbalance and harmonics [38–40] – Lyapunov function-based decentralized control scheme [150] – A distributed multi-agent framework for optimal power sharing [41–44] – A passivity-based coordinated control strategy is developed to manage an AC microgrid without using a phase-locked loop system [151] – A cooperative control strategy for voltage and frequency control [45] – Two level voltage control [46,47]

✓ ✓

S: Simulation level. E: Experimental level.

recommended in Table 15. As can be noted from the table, active methods may be distinguished from passive ones. With active methods, active and reactive power outputs of the master generator or of all controllable sources are rapidly controlled in order to adapt the overall system frequency and voltage magnitude to the reference signal. With passive synchronization methods, no control actions are implemented in order to adjust the set of the three voltage phasors of the microgrid to the one of the main grid. It is simply waited that the two sets of voltage phasors at both microgrid and utility grid sides are synchronized before enabling the master unit to start the resynchronization process. These methods are particularly suitable for microgrids with low or even null inertia. In fact, if the reference signal is sent to the master generator before achieving the synchronization condition, the generator will rapidly force the microgrid to be synchronized with the frequency of the main grid, causing a frequency jump. These deviations may bring about tripping of some components particularly sensitive to frequency changes, potentially leading to cascade events with all components shutting down progressively.

black-start schemes. Among them, the most relevant ones are those suggested in [55] and [56], whereby they propose adopting microgrids for the black-start service of large-scale distribution networks. 4. The Prince Lab microgrid description and analysis of the main challenges faced during its realization The aim of this section is to outline the main technical and operational challenges encountered during the development of the PrInCE Lab microgrid, and how these were addressed in practice. Along this line, the structure of the microgrid is firstly described. Afterwards, the control and monitoring system specifically designed to acquire data and actuate control actions in this microgrid is discussed. Finally, the innovative control functions implemented into the microgrid controller for ensuring the interoperability among its resources so as to achieve good levels of reliability and security are analyzed. 4.1. The electrical structure of the PrInCE Lab microgrid The PrInCE Lab microgrid is a low-voltage radial distribution network structured as a TN-S system. It encompasses four different generation types along with a Battery Energy Storage System (BESS) and two load banks. Generators can be differentiated on the basis of the primary energy source used into renewable and non-renewable energy sources. The first ones include a photovoltaic (PV) system with a rated power of 50 kW and a Wind Turbine Emulator (WTE) having a rated power of 60 kW. Non-renewable energy resources include instead a 30 kW microturbine (MT) and a Synchronous Generator (SG) having a rated power of 120 kW. Both these generators are fuelled by natural gas and they are operated in Combined Heat and Power (CHP) mode. The BESS integrated into the microgrid is composed by two Sodium-Nickel (So-Nick) battery banks having a total storage capacity of about 180 kWh and a rated power of 60 kW. There are also two programmable loads with a rated power of 120 kW each. All these components are connected to a common AC busbar that is in turn connected to the main grid through a circuit breaker which can be opened in order to isolate the microgrid. This device is also equipped with a Synchrocheck Relay enabling the microgrid to be reconnected to

3.4.7. Black start In the on-grid as well as off-grid states, if corrective control actions are activated on the system and they are not able to bring back the microgrid to its normal state, a general blackout could take place. At this stage, black-start procedures need to be activated to restore the microgrid. The relevant black-start procedures can be separated into on-grid and off-grid mode. 3.4.7.1. On-grid black start. Very few papers have focused on blackstart procedures for the on-grid mode restoration. Among them, Ref. [53] proposes a Micro Grid Central Controller (MGCC) that reconnects all devices to the main grid according to a predefined sequence. This sequence can be obtained by a multi-agent control, as in [54]. It must be noted that, during the blackout, some generators could feed local loads forming sub-microgrids. In this case the black-start procedure must invoke the re-synchronization procedure of such generators. 3.4.7.2. Off-grid black start. Table 16 summarizes selected off-grid

Table 12 Recommended reserve management strategies to improve islanded microgrid reliability. Control types Droop

Master-Slave

✓ ✓ ✓ (1) (2)

Validation level S

(1)

✓ ✓ ✓

E

Remarks (2)

– Multi-objective optimization considering SOC of Energy Storage Systems [152] – Reserve scheduling considering uncertainty, day-ahead electricity market and reserve costs [153–155] – Power exchange for frequency control (PXFC) market for reserve scheduling [156]

S: Simulation level. E: Experimental level. 10

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 13 Selected preventive control strategies to restore both grid-connected and isolated microgrids to the normal state. Operating modes On-Grid

Remarks

Ref

Off-Grid

✓

– – – – – – – – –

✓ ✓ ✓

Optimal management of the tie-line power flow Voltage-based control strategy based on tap changers of a Smart Transformer (ST) installed at the connection point of the microgrid Two PI regulators for active and reactive power flowing through the tie-line Manchester blackout model based on an optimal AC power flow Optimization problem for the optimal reconfiguration of the microgrid Self-adaptive Multi-Agent System operating in real-time N-1 and N-2 contingency analysis Multi-objective optimization Risk-based evaluation method

[123] [157] [158] [159] [160,161]

operation with other sources. To overcome this limit, it has been necessary to make design changes to their individual electrical configuration making thus, their integration in a microgrid more complex than the expected. Therefore, with the aim to simplify the integration process, in this project it was preferred to properly make design changes to their connection schemes as explained in [57].

Table 14 Summary of selected emergency control strategies for isolated microgrids. Optimization algorithms – A Non-linear programming method for minimizing the total load to be shed [162] – Constant impedance load approaches [163] – A multi-objective optimization method considering the total operation costs including the fuel costs and the penalty costs associated to the load shedding [164] – Scheduling a list of corrective actions which could be taken by the microgrid operator for extending the duration of the power supply in emergency situations [165]

4.1.1.1. The proposed protection scheme: Challenges and solutions. Another problem was raised by the need to assemble several devices in a common microgrid was the development of a protection scheme capable of ensuring the safe operation of the microgrid in all its operating modes and transitions. The first challenge to be faced was related to the proper coordination of anti-islanding protection relays of DER [58] of all devices embedded into the PrInCE Lab microgrid. In fact, due to the low inertia that characterize this microgrid large voltage and frequency fluctuations could occur in its normal operation which may eventually result in anti-islanding protection relay tripping and cascade events. To overcome this problem, changes to their settings were made to make them less sensitive to transient disturbances. Afterwards, in order to protect the connection line of each microgrid component against short-circuits and overloads, a magnetothermic switch combined with a differential circuit breaker (ANSI 50,51,87TD) was adopted to connect it to the common AC busbar. Then, in order to ensure the correct operation of the derived protection system, all these protection devices were coordinated with the upstream circuit breaker on the PCC. The coordination philosophy here is achieved by operating all protection devices with fixed settings. However, experimentations performed on the PrInCE Lab microgrid proved that breakers with fixed settings may be inadequate, especially when the system changes its operating state. This is mainly due to fact that short-circuit currents in isolated mode operation can be too much lower than that occurring in the on-grid mode. Therefore, in order to prevent miscoordination or mal-operation of the

the main grid, if adequate conditions occur. In order to give major flexibility to the PrInCE Lab microgrid, a 200 kVA four quadrant AC/AC inverter can constitute a by-pass for the tie-line switch. following severe contingencies occurring on the main distribution grid a busbar protection relay can activate the islanding of the microgrid. More details on the microgrid structure and its components can be found in [10] and [57]. In the following subsections the main problems encountered in realizing the PrInCE Lab microgrid will be discussed along with their related solutions. 4.1.1. Implementation issues and adopted technical solutions One of the most difficult challenging issues that was faced during the realization of the PrInCE Lab microgrid was the assembling of components and equipments that were designed have been specifically designed to be connected to the distribution grid. Most of these problems were mainly related to the electrical connections of the microturbine, the synchronous generator and the BESS. This was mainly due to the fact that these resources have been originally designed to be operated as UPS systems and, thus, they cannot be used in parallel Table 15 Summary of the recommended re-synchronization procedures. Method type Passive

Active

Control types Droop

✓ ✓

Master-Slave ✓

✓ ✓

✓ ✓ ✓ ✓ ✓

✓ ✓

✓ ✓ ✓ ✓ ✓

Remarks

– Synchrocheck relay waits the natural superposition of the voltage phasors of the two systems. When this condition is reached and if it persists for a sufficient time for which a stability condition can be reached, the tie–line switch will be closed [166–168] – An analogic voltage reference signal is sent to the master unit which will adapt its output according to the given reference signal [169] – Two different control loops have been implemented to resynchronize the microgrid to the main grid. The first one is based on an active method which forces the master unit to adjust its active and reactive power outputs to rapidly adapt the overall system frequency and voltage magnitude to the reference signal. The second one is passive and waits that the two sets of voltage phasors at both microgrid and utility grid sides are synchronized before enabling the master unit to start the resynchronization process [62] – Droop-free power sharing technique based on a high order current based virtual Phase Locked Loop (PLL) [170] – A modified voltage-based droop control scheme [171] – A cooperative control of multiple droop-controlled converters is employed to adjust the set of the three voltage phasors [172] – A seamless synchronization method for a single-phase microgrid is proposed [173] – A multilayer control structure is proposed for the mode transition of multiple inverter based microgrids [174]

11

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Table 16 Summary of the selected off-grid black Start procedures. Control types Droop ✓ ✓ ✓

Remarks Master-Slave

✓

– – – –

Start the restoration process by means an inverter based device operated as Voltage Source Inverter (VSI) [175–178] Set of rules and conditions to be checked during the restoration stage [179] Hierarchical control scheme based on a multi-agent system [55,56] Second-order integer programming for microgrid reconfiguration [180]

protection system, all relays were tuned with reference to the minimum short-circuit current contribution of each device. More specifically, all microgrid protection relays were set to trigger for a current equal to 1.5 of the nominal current (In), that is, the maximum contribution of power converters to the fault current. However, it should be noted that even if these settings are adequate to guarantee protection triggering following any short-circuit event in all operating states, they are not capable to ensure selectivity of operation among circuit breakers.

in the real-time. Therefore, a hardwired high-speed communication channel conveying variables that need rapid processing for real-time control was developed. More details on the developed communication network can be found in [59]. Layer 3: Real-time Control Layer Data coming from lower levels are processed in this layer giving rise to control actions to be sent to local devices. This layer is accomplished by two industrial PLCs in hot back-up configuration, ensuring thus high reliability and security levels. In fact, thanks to this redundancy feature, a failure in the master controller will not cause the unavailability of the overall control system. Each controller hosts a set of Energy Management routines providing the actuators command on the basis of the information coming from the physical devices. Energy Management routines can be classified into two main categories. The first one is related to control strategies responsible for monitoring, optimizing and dispatching all microgrid sources for economic and reliability purposes in all possible operating states and transitions. The second ones are related to the security of the microgrid, such as the high-speed load/storage/generation shedding and the emergency islanding/reconnection.

4.1.1.2. Supervisory control and data acquisition (SCADA) system. Another challenging aspects related to the practical implementation of the PrInCE Lab microgrid was the realization of a suitable control system able to interact with the control and protection systems of the main grid as well as to perform control functions and fault protection/service restoration for the microgrid. This is especially important in view of the possibility of an intentional islanding of the microgrid and then its autonomous operation. To comply with this exigency a control system specifically designed for the PrInCE Lab microgrid configuration was implemented. This control system has a hierarchical structure arranged in five layers as shown in Figure 2. Layer 0: Physical Layer of devices

Layer 4: The SCADA Layer

Layer 0 deals with all physical devices that are connected to the microgrid. More specifically, it comprises the synchronous generator, microturbine, PV system, WTE and by-pass inverter. It also includes the BESS and two programmable loads. Other devices such as power electronic converters usually adopted to connect these components to the host grid according to their configurations fall within this layer.

The safe and economic operation of the microgrid is managed in this layer. To meet this requirements, long-term energy management functions are usually adopted. The derived operation plan is then passed to the real-time control layer, where it is processed to derive optimal control actions for local controllers of all energy sources. This control layer comprises two servers in the hot standby configuration to enhance the overall system reliability. Each server is able to provide process data for supervisory and control actions through a Human Machine Interface (HMI) software package. The HMI was programmed with screens tailormade for the PrInCE Lab microgrid. In addition to classical SCADA functions such as the management of alarms and trends, these screens allow the operator to control the microgrid by changing its operating state (islanding and resynchronization), sending active and reactive power set-points to local controllers, and activating particular Energy Management routines.

Layer 1: Physical Layer of power equipment Layer 1 encompasses the protection and metering systems. The protection system contains circuit breakers, switches, contactors, and multifunction protective relays. The metering system is constituted by remote I/O modules and meters capable to carry out from devices belonging to the Layer 0 all information required to monitor the operating state of the microgrid (i.e. voltage, frequency, power factor, active and reactive powers for each microgrid components and their status). These information are then passed to the upper control Layer 3.

4.1.2. PrInCE Lab microgrid operation The development of the an Energy Management System (EMS) able to efficiently coordinate all devices for an economic, reliable and secure operation of the entire microgrid in all possible operating states and transitions was another challenge in the development of the PrInCE Lab microgrid. The real-time control requirements of the system require the fully automatic microgrid operation with minimal operator involvement. To achieve this, several control functions were developed in this project. The first control function was implemented for the optimal operation of the microgrid when it is operated in the grid-connected mode. To comply with this exigency, an economic dispatch has been implemented in the real-time controller. This procedure, operating in the continuous time domain, is able to redistribute in the best economic way power unbalances among all dispatchable resources. The second

Layer 2: Communication Layer The Layer 2 transfers data between remote I/O modules and meters of the Layer 1 and the Layer 3. Since all components are produced by several manufacturers, they adopt different communication protocols, ranging from proprietary protocols, DNP, Profibus, Modbus TCP, or Ethernet [59]. In order to overcome this communication problem, protocol converter devices able to adapt the given protocol into the Modbus TCP/IP are required. This protocol was adopted because it can be considered as the standard communication protocol for industrial controllers. However, the huge amount of system variables (about 1300) exchanged through a serial protocol can give rise to long communication delays that cannot be compatible with control requirements 12

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Fig. 2. The Prince Lab Microgrid control layers.

down rate limits [61]. Another challenge that must be faced during the operation of the microgrid is related to its resynchronization with the main grid. For this microgrid, the passive synchronization routine developed in [62] was implemented into the real-time controller. Upon receiving the resynchronization command, the main controller starts to monitor the two sets of voltage phasors at both microgrid and utility grid sides. As soon as it detects the synchronization of the two voltage phasors, a voltage reference signal is sent to the master generator. Upon receiving this signal, the master unit will adjust its power outputs until its voltage and frequency are aligned to that of the main grid. The synchrocheck relay will verify the correctness of the synchronization and a closing signal will be sent to the tie-line breaker.

control function operates a seamless unplanned islanding of the microgrid in case of an outage of the main grid. In this case, the antiislanding relay installed on the main circuit breaker at the PCC is triggered by an over- or under-voltage, or by an over- or under-frequency, resulting thus in the tripping of the tie-line breaker. The trip of this circuit breaker is suddenly detected by the real-time microgrid controller that promptly react by forcing the preselected master unit to switch from P/Q to V/f control mode. This has to occur prior the microgrid reaches its “point of no return” and the blackout is inevitable. Experimental tests performed on the PrInCE Lab microgrid demonstrated that the system takes about 470 ms to reach the point of no return were voltage and frequency drops are such as to result in the DER protection devices tripping [10]. Therefore, to reduce the risk of the blackout the islanding transition must be recovered in less than 470 ms. To comply with this exigency, a hardwired relay logic was used to provide the automatic sequencing of the microgrid islanding [57]. The basic idea of this control strategy consists in blinding the anti-islanding protection relay of the unit that was selected to act as the master for the forming island so as to force its local controller to convert its control mode from P/Q to V/f control. As experimentally proven, this control strategy takes up about 150 ms to recover the microgrid voltage to its nominal value [57]. In order to minimize the perturbations deriving from the opening of the tie-line breaker, a pre-planned bumpless islanding transition can be performed. To comply with this exigency, the real-time controller has been programmed in order to get to zero the tie-line power flow by managing all internal resources. [60]. When this condition is achieved, the opening of the tie-line breaker does not produce significant perturbations, reducing thus the risk of blackouts. The voltage and frequency stability during the system operation in the off-grid mode constitutes another difficult task to deal with. To mitigate the risk of microgrid instability, the electric energy balance needs to be ensured in the on-line environment. For a master/slave controlled microgrid as in this case, this task is accomplished by the master unit that must have enough regulation capacity to cover any possible unbalances which may occur on the microgrid. This could mean of having an oversized master unit which would lead to unnecessary high investment and operation costs since it would operate at the maximum power only for few hours per years. To overcome this limit, in the EMS of the PrInCE Lab microgrid a cooperative control strategy for the reserve management has been implemented. Its main objective is to keep the overall spinning reserve to a sufficient level by managing the active power outputs of all DER embedded into the microgrid according to their reserve margins and their ramp-up and ramp-

5. Key lessons learned and recommendations The PrInCE Lab microgrid project demonstrated that is possible to realize a microgrid by adopting components and equipment originally developed for classical distribution network applications. However, the adoption of these components made their integration into a microgrid structure more complex than the expected. An extensive analysis of the main technical and operational challenges that may be encountered during the deployment of a realistic microgrid is then required in order to define a set of recommendations and general guidelines that can be used by developers and researchers. Although it is difficult to define general criteria that may be used for any specific microgrid installation the experience gained during the development of the PrInCE Lab microgrid could be helpful for defining some directions to go. 5.1. General guidelines for the design, implementation and monitoring of microgrids 5.1.1. Microgrid topology and architecture Lessons drawn from the examination of the existing microgrid projects suggest that both the topology and structure of such systems strongly depend on their specific applications, thus making the generalization of the microgrid design more difficult. Nevertheless, according to the comprehensive literature review carried out in this paper, it was proven that AC radial microgrids are the most effective solutions for the economic integration of DER already connected to the distribution networks. 13

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

5.1.2. Sizing and selection of the master unit in master/slave controlled microgrids The size and selection of the master unit to be installed in a master/ slave microgrid must be carefully considered. In fact, in the event of main grid outages, the master unit could be called to provide the sudden unbalance caused by the loss of the tie line power flow. Although there are no general criteria that should be applied to all microgrids to accomplish with this requirement, the analysis of the literature review carried out in this paper suggests to adopt specific planning tools for the optimal sizing and selection of the master in a microgrid. These tools are mainly based on cost-benefit analysis aimed at minimizing the cost of investments as well as operating costs and interruptions of service while ensuring that resilience and security requirements are met.

microgrid that will be summarized in the following subsections. 5.3.1. Dynamic stability assessment of microgrids One of the most challenging issues that must be addressed in the operation of a microgrid is to assess its ability to survive to disturbances. With this aim, time domain simulators for the dynamic security assessment need to be developed. For this purpose, such simulators need accurate mathematical models of all microgrid components with their controllers and the main electrical equipment. Although, the technical literature is rich in references on modeling of these components, experimental tests performed in the PrInCE Lab microgrid demonstrated that the behavior simulated by these models is not always adherent to that exhibited by physical devices. Moreover, depending on the complexity of the specific microgrid, these detailed models may give rise to computational inefficiencies if real-time DSA are performed. Therefore, research efforts should focus on developing accurate and computationally efficient dynamic models. A first attempt in this direction can be found in [63–65].

5.1.3. Electrical connection of the master unit After carefully identifying the type and size of the master unit, particular attention must be paid to its installation. In fact, as experienced in the PrInCE Lab microgrid, most of the sources that can act as voltage forming are designed to be operated as UPS systems and thus they cannot be used in parallel operation with other DER. To overcome this issue, the technical literature suggests modifying their internal electrical configuration or make changes to their control algorithms. Lessons learned in the PrInCE Lab microgrid suggest making design changes to the connection schemes of these DER so as to overcome the issues arising from the modification of their internal control circuits [57].

5.3.2. Failure of the master unit Another challenge that must be taken into account during the development of a microgrid refers to its safety in the event that a failure of the master unit occurs when it is operated in isolated mode. To mitigate this risk, the PrInCE Lab microgrid was structured in such a way to integrate three energy sources that can take the master function. Anyway, the choice of the master unit is done according to the actual operating point of such unit and the adopted power reserve policy. Then, the microgrid master controller transfers the master function among to all available units following a token ring procedure [66].

5.1.4. Upgrade of the existing protection schemes Fixed relay settings usually adopted in traditional protection schemes of distribution networks could be inadequate for microgrid protections. To overcome this issue and achieve correct fault clearance operation, several papers in the technical literature suggest to adopt adaptive protection schemes in which setting of protections against short-circuits can be varied according to the varying system topology and the system operating point. However, although this technique could provide good performance, it could not be practical since it implies higher investment costs for the need of programmable switching devices along with an expansion of the electrical and communication networks. To simplify the implementation of the microgrid protection system and to minimize the investment costs, a line and a busbar differential protection systems could be considered the most technical/ cost effective solution for a microgrid.

6. Conclusions The aim of this paper was to provide a set of recommendations and general guidelines that can be useful for designers and researchers to address the challenges emerging when dealing with actual microgrids. For this purpose, a comprehensive literature review was undertaken to outline the main design features of existing microgrids as well as the main control functions that are required to ensure their economic, reliable and secure operation in different operating modes and transitions. The learnings from the literature review have then been augmented with the authors’ specific practical experience from the Prince Lab project, whose main components and control system architecture features were selected to enable a high level of planning and operational testing flexibility. In fact, unlike most of the existing microgrids or those that are still under development, this facility was not designed and built with customized components, but rather with the aim to evaluate the possibility to realize a microgrid starting from commercially available components that are already deployed in distribution systems. Additionally, each dispatchable microgrid source was specifically chosen with a rated power insufficient to comply with all possible unbalances which may occur in the system. These design choices thus lead to deal with new technical and operational challenges including advanced control, protection and infrastructure requirements. The key outcomes of the analysis carried out in the paper outline the main technical and control challenges and the main solutions that can be adopted to enable smooth operation of the microgrid in all operating modes and transitions. Based on these outcomes, a set of recommendations and insights that can be used as a reference for developing similar microgrids have been provided. However, it should be pointed out that these recommendations should only be considered as guidelines since the development of a microgrid involves several unknowns that are associated with its specific applications and location. Hence, the identified solutions should be further tailored to meet the specific requirements of the case under analysis.

5.1.5. Upgrade of the existing communication networks The interoperability among all microgrid components and devices strictly depend on a high-speed and secure communication network. To comply with this exigency, a communication network compliant to the standard IEC 61850 seems to be the optimal solution even if until now it is not fully exploited because it is not yet supported by commercially available microcontrollers. Communication interfaces cannot be seen as a possible solution, however, because they introduce a bottleneck in the communication system. A wired communication infrastructure can then be considered the remaining solution for transmitting variables requiring low latency time. 5.2. Development of the control system architecture The adoption of a hierarchical control structure arranged on the above mentioned five layers, ranging from the master controller to the microgrid components, seems to be the more opportune for the development of a microgrid. 5.3. General guidelines for the microgrid operation There are several issues that must be addressed in the operation of a 14

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

Declaration of Competing Interest

applications. Renew Sustain Energy Rev 2013;19:191–9. [29] Martin-Martínez F, Sánchez-Miralles A, Rivier M. A literature review of microgrids: a functional layer based classification. Renew Sustain Energy Rev 2016;62:1133–53. [30] Moghimi M, Bennett C, Leskarac D, Stegen S, Lu J. Communication architecture and data acquisition for experimental MicroGrid installations. In: 2015 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC), November 2015. p. 1–5. [31] Khan AA, Naeem M, Iqbal M, Qaisar S, Anpalagan A. A compendium of optimization objectives, constraints, tools and algorithms for energy management in microgrids. Renew Sustain Energy Rev 2016;58:1664–83. [32] Xiang Y, Liu J, Liu Y. Robust energy management of microgrid with uncertain renewable generation and load. In: IEEE Transactions on Smart Grid 7(2); 2016:1034–1043. [33] Nutkani IU, Loh PC, Wang P, Blaabjerg F. Decentralized economic dispatch scheme with online power reserve for microgrids. IEEE Trans Smart Grid 2017;8(1):139–48. [34] Ortega-Vazquez MA, Kirschen DS. Security provision in systems with large penetration of wind power generation. In: IEEE PES General Meeting. IEEE. 2010. p. 1–8. [35] Liu G, Starke M, Xiao B, Zhang X, Tomsovic K. Microgrid optimal scheduling with chance-constrained islanding capability. Electr Power Syst Res 2017;145:197–206. [36] Katiraei F, Abbey C, Tang S, Gauthier M. Planned islanding on rural feeders—utility perspective. In: Power and Energy Society General Meeting-Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE, IEEE; 2008. p. 1–6. [37] Zhao Z, Yang P, Guerrero JM, Xu Z, Green TC. Multiple-time-scales hierarchical frequency stability control strategy of medium-voltage isolated microgrid. IEEE Trans Power Electron 2016;31(8):5974–91. [38] Han Y, Shen P, Zhao X, Guerrero JM. An enhanced power sharing scheme for voltage unbalance and harmonics compensation in an islanded AC microgrid. IEEE Trans Energy Convers 2016;31(3):1037–50. [39] Shafiee Q, Guerrero JM, Vasquez JC. Distributed secondary control for islanded microgrids—a novel approach. IEEE Trans Power Electron 2014;29(2):1018–31. [40] Savaghebi M, Jalilian A, Vasquez JC, Guerrero JM. Autonomous voltage unbalance compensation in an islanded droop-controlled microgrid. IEEE Transactions on Industrial Electronics 60(4);2013:1390–1402. [41] Rahman MS, Oo AMT. Distributed multi-agent based coordinated power management and control strategy for microgrids with distributed energy resources. Energy Convers Manage 2017;139:20–32. [42] Kofinas P, Dounis AI, Vouros GA. Fuzzy Q-learning for multi-agent decentralized energy management in microgrids. Appl Energy 2018;219:53–67. [43] Dehkordi NM, Sadati N, Hamzeh M. Fully distributed cooperative secondary frequency and voltage control of islanded microgrids. IEEE Trans Energy Convers. 2017;32(2):675–85. [44] Li Z, Zang C, Zeng P, Yu H, Li S. Fully distributed hierarchical control of parallel grid-supporting inverters in islanded AC microgrids. IEEE Trans Ind Inf 2018;14(2):679–90. [45] Kim YS, Kim ES, Moon SI. Frequency and voltage control strategy of standalone microgrids with high penetration of intermittent renewable generation systems. In: IEEE Transactions on Power Systems, 2016, vol. 31(1). p. 718–28. [46] Mahmoud MS, Al-Buraiki O. Two-level control for improving the performance of MicroGrid in islanded mode. In: 2014 IEEE 23rd International Symposium on Industrial Electronics (ISIE). IEEE; June 2014. p. 254–9. [47] Mortezaei A, Simões MG, Marafão FP. Cooperative operation based master-slave in islanded microgrid with CPT current decomposition. In: 2015 IEEE Power & Energy Society General Meeting. IEEE; 2015. p. 1–5. [48] Ramos DS, Del Carpio-Huayllas TE, Vasquez-Arnez RL. Load shedding application within a microgrid to assure its dynamic performance during its transition to the islanded mode of operation. Energy Power Eng 2013;5(7):437–45. [49] Gouveia C, Rua D, Moreira CL, Lopes JP. Coordinating distributed energy resources during microgrid emergency operation. In: Renewable Energy Integration. Springer Singapore; 2014. p. 259–303 [50] Mousavizadeh S, Haghifam MR, Shariatkhah MH. A linear two-stage method for resiliency analysis in distribution systems considering renewable energy and demand response resources. Appl Energy 2018;211:443–60. [51] Vian A, Bignucolo F, Turri R, Cagnano A, De Tuglie E. Dynamic modeling of the PrInCE Lab experimental microgrid. In: 2018 AEIT International Annual Conference. IEEE; 2018. p. 1–6. [52] Mahmoud MS, Hussain SA, Abido MA. Modeling and control of microgrid: an overview. J Franklin Inst 2014;351(5):2822–59. [53] Feltes JW, Grande-Moran C. Black start studies for system restoration. In: Power and Energy Society General Meeting-Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE. IEEE; July 2008. p. 1–8. [54] Li P, Song B, Wang W, Wang T. Multi-agent approach for service restoration of microgrid. In: 2010 5th IEEE Conference on Indstrial Electronics and Applications, IEEE. IEEE. June 2010. p. 962–6. [55] Cai N, Xu X, Mitra J. A hierarchical multi-agent control scheme for a black startcapable microgrid. In: 2011 IEEE Power and Energy Society General Meeting, IEEE; 2011. p. 1–7. [56] El-Sharafy MZ, Farag HE. Back-feed power restoration using distributed constraint optimization in smart distribution grids clustered into microgrids. Appl Energy 2017;206:1102–17. [57] Cagnano A, De Tuglie E, Trovato M, Cicognani L, Vona V. A simple circuit model

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References [1] Lasseter R. Microgrids and distributed generation. J Energy Eng 2007;133:144–9. [2] Karugarama MK. Mitiiation of blackout in kigali using a microgrid with advanced energy storage and solar photovoltaics. Doctoral dissertation. Virginia Tech; 2016. [3] Haritza C, Octavian C, Aitor E, Alvaro L, Amélie HP. Research experimental platforms to study microgrids issues. Int J Interact Des Manuf (IJIDeM) (July) (2015) 1–13. [4] Tan D. Emerging system applications and technological trends in power electronics: power electronics is increasingly cutting across traditional boundaries. IEEE Power Electron Mag 2015;2(2):38–47. [5] Kyriakarakos G, Piromalis DD, Dounis AI, Arvanitis KG, Papadakis G. Intelligent demand side energy management system for autonomous polygeneration microgrids. Appl Energy 2013;103:39–51. [6] Ustun TS, Ozansoy C, Zayegh A. Recent developments in microgrids and example cases around the world—a review. Renew Sustain Energy Rev 2011;15(8):4030–41. [7] Agrawal M, Mittal A. Deployment of microgrids in the developed countries: an appraisal. Int J Electr Electron Eng Res (IJEEER) 2014;4(3):23–34. [8] Hossain E, Kabalci E, Bayindir R, Perez R. Microgrid testbeds around the world: state of art. Energy Convers Manage 2014;86:132–53. [9] Bracco S, Delfino F, Pampararo F, Robba M, Rossi M. The University of Genoa smart polygeneration microgrid test-bed facility: the overall system, the technologies and the research challenges. Renew Sustain Energy Rev 2013;18:442–59. [10] Cagnano A, De Tuglie E, Cicognani L. Prince—electrical energy systems lab: a pilot project for smart microgrids. Electr Power Syst Res July 2017;148:10–7. [11] Eto J, Lasseter R, Schenkman B, Stevens J, Klapp D, Volkommer H, et al. Overview of the CERTS microgrid laboratory test bed. In: Integration of Wide-Scale Renewable Resources Into the Power Delivery System, 2009 CIGRE/IEEE PES Joint Symposium. IEEE. July 2009. p. 1–7. [12] Lasseter RH, Eto JH, Schenkman B, Stevens J, Vollkommer H, Klapp D, et al. CERTS microgrid laboratory test bed. IEEE Trans Power Delivery 2011;26(1):325–32. [13] Ariyasinghe MNS, Hemapala KTMU. Microgrid test-beds and its control strategies. Smart Grid Renew Energy 2013;4(01):11–7. [14] Moneta D, Marciandi M, Tornelli C, Mora P. Voltage optimization of a LV microgrid with DERs: field test. In: 20th International Conference and Exhibition on Electricity Distribution – Part 1, 2009. CIRED 2009, pp. 1–4. IET. June 2009. [15] Barnes M, Dimeas A, Engler A, Fitzer C, Hatziargyriou N, Jones C et al. Microgrid laboratory facilities. In: 2005 International Conference on Future Power Systems, IEEE, November 2005. p. 1–6. [16] Loix T, Leuven KU. The first micro grid in the Netherlands: Bronsbergen. Available online on February 2009 < http://www.leonardo-energy.org/sites/leonardoenergy/files/root/pdf/2009/article2.pdf > . [17] Loix T, Leuven KU. The residential micro grid of Am Steinweg in Stutensee, Germany. Available online on February 2009 < http://www.leonardo-energy.org/ sites/leonardo-energy/files/root/pdf/2009/article3.pdf > . [18] Katiraei F, Abbey C, Tang S, Gauthier M. Planned islanding on rural feeders—utility perspective. In: power and energy society general meeting-conversion and delivery of electrical energy in the 21st century, 2008 IEEE, IEEE. July 2008. p. 1–6. [19] Patrao I, Figueres E, Garcerá G, González-Medina R. Microgrid architectures for low voltage distributed generation. Renew Sustain Energy Rev 2015;43:415–24. [20] Planas E, Andreu J, Gárate JI, de Alegría IM, Ibarra E. AC and DC technology in microgrids: a review. Renew Sustain Energy Rev 2015;43:726–49. [21] Memon AA, Kauhaniemi K. A critical review of AC microgrid protection issues and available solutions. Electr Power Syst Res 2015;129:23–31. [22] Hosseini SA, Abyaneh HA, Sadeghi SHH, Razavi F, Nasiri A. An overview of microgrid protection methods and the factors involved. Renew Sustain Energy Rev 2016;64:174–86. [23] Habib HF, Lashway CR, Mohammed OA. A review of communication failure impacts on adaptive microgrid protection schemes and the use of energy storage as a contingency. IEEE Trans Ind Appl 2018;54(2):1194–207. [24] Li HY, Yunus B. Assessment of switched communication network availability for state estimation of distribution networks with generation. IEEE Trans Power Delivery 2007;22(3):1424–32. [25] Gungor VC, Sahin D, Kocak T, Ergut S, Buccella C, Cecati C, et al. A survey on smart grid potential applications and communication requirements. IEEE Trans Ind Inf 2013;9(1):28–42. [26] Amin R, Martin J, Zhou X. Smart grid communication using next generation heterogeneous wireless networks. In: 2012 IEEE Third International Conference on Smart Grid Communications (SmartGridComm). IEEE; November 2012. p. 229–34. [27] Ahmed M, Kang Y, Kim YC. Communication network architectures for smart-house with renewable energy resources. Energies 2015;8(8):8716–35. [28] Usman A, Shami SH. Evolution of communication technologies for smart grid

15

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

[58] [59] [60]

[61] [62]

[63] [64] [65]

[66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88]

for the islanding transition of microgrids. In: 2016 IEEE 2nd International Forum on Research and Technologies for Society and Industry Leveraging a better tomorrow (RTSI); 2016. p. 1–6. CEI Comitato Elettrotecnico Italiano. Standard-CEI 0-21, reference technical rules for the connection of active and passive users to the LV electrical utilities, standards; 2011–2012. Cagnano A, Raza Sherazi HH, De Tuglie E. Communication system architecture of an industrial-scale microgrid: a case study. Internet Technology Letters. Cagnano A, De Tuglie E, Rasolomampionona DD, Klos M, Favuzza S, Massaro F et al. Transitions from grid-connected to island operation of Smart Microgrids. In: 2019 IEEE International Conference on Environment and Electrical Engineering and 2019 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe); 2019. p. 1–6. Cagnano A, Caldarulo Bugliari A, De Tuglie E. A cooperative control for the reserve management of isolated microgrids. Appl Energy 218; 2018:256–65. Cagnano A, De Tuglie E, Cervi A, Stecca R, Turri R, Vian A. Re-synchronization control strategy for master-slave controlled microgrids. In: 2019 1st International Conference on Energy Transition in the Mediterranean Area (SyNERGY MED). IEEE; 2019. p. 1–6. Cagnano A, De Tuglie E. On-line identification of simplified dynamic models: Simulations and experimental tests on the Capstone C30 microturbine. Electr Power Syst Res 2018;157:145–56. Cagnano A, De Tuglie E. Time domain identification of a simplified model of So–Nick BESS: a methodology validated with field experiments. Electr Power Syst Res 2018;165:229–37. Vian A, Cervi A, Stecca R, Cagnano A, De Tuglie E. Validation of the dynamic model of the PrInCE lab CHP through real-time measurements. In: 2019 1st International Conference on Energy Transition in the Mediterranean Area (SyNERGY MED). IEEE; 2019. p. 1–6. Manaz MM, Lu CN. Adaptive defense plan against anticipated islanding of microgrid. In: IEEE Transactions on Smart Grid; March 2018. Romankiewicz J, Marnay C, Zhou N, Qu M. Lessons from international experience for China's microgrid demonstration program. Energy Policy 2014;67:198–208. Qu M. Microgrid policy review of selected major countries, regions, and organizations; 2014. Palma-Behnke R, Ortiz D, Reyes L, Jimenez-Estevez G, Garrido N. A social SCADA approach for a renewable based microgrid—the Huatacondo project. In: 2011 IEEE Power and Energy Society General Meeting. IEEE; July 2011. pp. 1–7. Marnay C. International Microgrid Assessment: Governance, Incentives, and Experience (IMAGINE); 2014. Lasseter RH. Microgrids. In: power engineering society winter meeting. IEEE; 2002, Vol. 1. p. 305–8. Kroposki B, Lasseter R, Ise T, Morozumi S, Papathanassiou S, Hatziargyriou N. Making microgrids work. IEEE Power and Energy Magazine 2008;6(3):40–53. Mariam L, Basu M, Conlon FM. A review of existing microgrid architectures. J Eng 2013. Lidula NWA, Rajapakse AD. Microgrids research: a review of experimental microgrids and test systems. Renew Sustain Energy Rev 2011;15(1):186–202. Mariam L, Basu M, Conlon MF. Microgrid: architecture, policy and future trends. Renew Sustain Energy Rev 2016;64:477–89. Unamuno E, Barrena JA. Hybrid ac/dc microgrids—Part I: Review and classification of topologies. Renew Sustain Energy Rev 2015;52:1251–9. Haron AR, Mohamed A, Shareef H, Zayandehroodi H. Analysis and solutions of overcurrent protection issues in a microgrid. In: IEEE international conference on power and energy (PECon). February 2013. p. 644–49. Jin YT, Park SJ, Lee SJ, Choi MS. Intelligent agent based protection for smartdistribution systems. In: CIRED 21st International Conference on ElectricityDistribution, 6–9 June, 2011, Frankfurt; 2011. Al-Nasseri H, Redfern MA, Li F. A voltage based protection for micro-grids containing power electronic converters. IEEE Power Eng Soc General Meeting 2006:7. Al-Nasseri H, Redfern MA, O0Gorman R. Protecting micro-grid systems containing solid-state converter generation. In: Proceedings of the International Conference on Future Power Systems Amsterdam; 2005. p. 1–5. Sitharthan R, Geethanjali M, Pandy TKS. Adaptive protection scheme for smart microgrid with electronically coupled distributed generations. Alexandria Eng J 2016;55(3):2539–50. Brearley BJ, Prabu RR. A review on issues and approaches for microgrid protection. Renew Sustain Energy Rev 2017;67:988–97. Mirsaeidi S, Said DM, Mustafa M, Habibuddin M, Ghaffari K. Review and analysis of existing protection strategies for micro-grids. J Electr Syst 2014;10(1). Dewadasa M, Ghosh A, Ledwich G. Protection of microgrids using differential relays. In: AUPEC 2011. IEEE. September 2011. p. 1–6. Bui DM, Chen SL. Fault protection solutions appropriately proposed for ungrounded low-voltage AC microgrids: review and proposals. Renew Sustain Energy Rev 2017;75:1156–74. Dewadasa M, Ghosh A, Ledwich G. Line protection in inverter supplied net works. In: Australasian Universities Power Engineering Conference (AUPEC’08); 2008. p. 1–6. Lin YJ, Latchman HA, Lee M, Katar S. A power line communication network infrastructure for the smart home. IEEE Wirel Commun 2002;9(6):104–11. Elgargouri A, Virrankoski R, Elmusrati M. IEC 61850 based smart grid security. In: 2015 IEEE International Conference on Industrial Technology (ICIT). IEEE; March 2015. p. 2461–2465.

[89] Bracco S, Delfino F, Pampararo F, Robba M, Rossi M. A dynamic optimizationbased architecture for polygeneration microgrids with tri-generation, renewables, storage systems and electrical vehicles. Energy Convers Manage 2015;96:511–20. [90] Fu Y, Li Z, Wu L. Modeling and solution of the large-scale security-constrained unit commitment. IEEE Trans Power Syst 2013;28(4):3524–33. [91] Bonfiglio A, Bracco S, Brignone M, Delfino F, Pampararo F, Procopio R, et al. A receding-horizon approach for active and reactive power flows optimization in microgrids. In: 2014 IEEE Conference on Control Applications (CCA). IEEE. October 2014. p. 867–72. [92] Hawkes AD, Leach MA. Modelling high level system design and unit commitment for a microgrid. Appl Energy 2009;86(7):1253–65. [93] Wu X, Wang X, Bie Z. Optimal generation scheduling of a microgrid. In: 2012 3rd IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe). IEEE. October 2012. p. 1–7. [94] Carpinelli G, Mottola F, Proto D, Varilone P. Minimizing unbalances in low-voltage microgrids: optimal scheduling of distributed resources. Appl Energy 2017;191:170–82. [95] Yang Z, Wu R, Yang J, Long K, You P. Economical operation of microgrid with various devices via distributed optimization. IEEE Trans Smart Grid 2016;7(2):857–67. [96] Wang J, Wang J, Liu C, Ruiz JP. Stochastic unit commitment with sub-hourly dispatch constraints. Appl Energy 2013;105:418–22. [97] Luh PB, Yu Y, Zhang B, Litvinov E, Zheng T, Zhao F, et al. Grid integration of intermittent wind generation: a Markovian approach. IEEE Trans Smart Grid 2014;5(2):732–41. [98] Cui H, Li F, Hu Q, Bai L, Fang X. Day-ahead coordinated operation of utility-scale electricity and natural gas networks considering demand response based virtual power plants. Appl Energy 2016;176:183–95. [99] Analui B, Scaglione A. A dynamic multistage stochastic unit commitment formulation for intraday markets. In: IEEE Trans Power Syst 2017. [100] Zhang J, Wu Y, Guo Y, Wang B, Wang H, Liu H. A hybrid harmony search algorithm with differential evolution for day-ahead scheduling problem of a microgrid with consideration of power flow constraints. Appl Energy 2016;183:791–804. [101] Wang Y, Zhou Z, Zhao S, Xu Y, Chen R, Yin J. Chance-constrained goal-programming based day ahead scheduling in wind power integrated system. In: Power and Energy Society General Meeting (PESGM); 2016. p. 1–5. [102] Zachar M, Daoutidis P. Microgrid/macrogrid energy exchange: a novel market structure and stochastic scheduling. IEEE Trans Smart Grid 2017;8(1):178–89. [103] Zhao C, Guan Y. Unified stochastic and robust unit commitment. IEEE Trans Power Syst 2013;28(3):3353–61. [104] Blanco I, Morales JM. An efficient robust solution to the two-stage stochastic unit commitment problem. IEEE Trans Power Syst 2017;32(6):4477–88. [105] Zamani AG, Zakariazadeh A, Jadid S. Day-ahead resource scheduling of a renewable energy based virtual power plant. Appl Energy 2016;169:324–40. [106] Belgana A, Rimal BP, Maier M. Open energy market strategies in microgrids: a Stackelberg game approach based on a hybrid multiobjective evolutionary algorithm. IEEE Trans Smart Grid 2015;6(3):1243–52. [107] Bronzini M, Bruno S, La Scala M, Sbrizzai R. Coordination of active and reactive distributed resources in a smart grid. In: 2011 IEEE Trondheim PowerTech. IEEE. June 2011. p. 1–7. [108] Nwulu NI, Xia X. Optimal dispatch for a microgrid incorporating renewables and demand response. Renew Energy 2017;101:16–28. [109] Fu L, Zhang W, Dong Z, Meng K. A mixed logical dynamical model for optimal energy scheduling in microgrids. In: 2017 IEEE Conference on Energy Internet and Energy System Integration (EI2). November 2017. p. 1–6. [110] Ranaweera I, Midtgård OM. Optimization of operational cost for a grid-supporting PV system with battery storage. Renew Energy 2016;88:262–72. [111] Sobu A, Wu G. Dynamic optimal schedule management method for microgrid system considering forecast errors of renewable power generations. In: 2012 IEEE International Conference on Power System Technology (POWERCON); 2012. p. 1–6. IEEE. [112] Luna AC, Diaz NL, Andrade F, Graells M, Guerrero JM, Vasquez JC. Economic power dispatch of distributed generators in a grid-connected microgrid. In: 2015 9th International Conference on Power Electronics and ECCE Asia (ICPE-ECCE Asia). IEEE. June 2015. p. 1161–1168. [113] Luna AC, Aldana NLD, Graells M, Vasquez JC, Guerrero JM. Mixed-integer-linearprogramming based energy management system for hybrid PV-wind-battery microgrids: Modeling, design and experimental verification. IEEE Trans Power Electron PP(99);2016:1–1. [114] Bonfiglio A, Barillari L, Brignone M, Delfino F, et al. An optimization algorithm for the operation planning of the University of Genoa smart polygeneration microgrid. In: Bulk power system dynamics and control-IX optimization, security and control of the emerging power grid (IREP), 2013 IREP symposium. IEEE. 2013. p. 1–8. [115] Borghetti A, Bosetti M, Grillo S, Paolone M, Silvestro F. Short-term scheduling of active distribution systems. In: PowerTech, 2009 IEEE Bucharest. IEEE. June 2009. p. 1–7. [116] Bahramipanah M, Torregrossa D, Cherkaoui R, Paolone M. An adaptive modelbased real-time voltage control process for active distribution networks using battery energy storage systems. In: Power System Computation Conference (PSCC) (No. EPFL-CONF-215925); 2016. [117] Nayeripour M, Fallahzadeh-Abarghouei H, Waffenschmidt E, Hasanvand S. Coordinated online voltage management of distributed generation using network

16

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

for isolated microgrids. IEEE Trans Smart Grid 2014;5(4):1864–75. [147] Li Z, Xu Y. Optimal coordinated energy dispatch of a multi-energy microgrid in grid-connected and islanded modes. Appl Energy 2018;210:974–86. [148] Mazzola S, Astolfi M, Macchi E. A detailed model for the optimal management of a multigood microgrid. Appl Energy 2015;154:862–73. [149] Wang C, Liu Y, Li X, Guo L, Qiao L, Lu H. Energy management system for standalone diesel-wind-biomass microgrid with energy storage system. Energy 2016;97:90–104. [150] Wang K, Huang X, Fan B, Yang Q, Li G, Crow ML. Decentralized power sharing control for parallel-connected inverters in islanded single-phase micro-grids. In: IEEE Transactions on Smart Grid; 2017. [151] Gui Y, Wei B, Li M, Guerrero JM, Vasquez JC. Passivity-based coordinated control for islanded AC microgrid. Appl Energy 2018;229:551–61. [152] Nutkani IU, Loh PC, Wang P, Blaabjerg F. Linear decentralized power sharing schemes for economic operation of AC microgrids. IEEE Trans Ind Electron 2016;63(1):225–34. [153] Alharbi W, Raahemifar K. Probabilistic coordination of microgrid energy resources operation considering uncertainties. Electr Power Syst Res 2015;128:1–10. [154] Lee SY, Jin YG, Yoon YT. Determining the optimal reserve capacity in a microgrid with islanded operation. IEEE Trans Power Systems 31(2) 1369–76. [155] Solanki BV, Raghurajan A, Bhattacharya K, Canizares CA. Including smart loads for optimal demand response in integrated energy management systems for isolated microgrids. IEEE Trans Smart Grid 2017;8(4):1739–48. [156] El Moursi MS, Zeineldin HH, Kirtley JL, Alobeidli K. A dynamic master/slave reactive power-management scheme for smart grids with distributed generation. IEEE Trans Power Del 2014;29(3):1157–67. [157] Bose S, Liu Y, Bahei Eldin K, de Bedout J, Adamiak M. Tieline controls in microgrid applications. In: IREP Symposium on Bulk Power System Dynamics and Control -VII.Revitalizing Operational Reliability; August 2007. p. 1–9. [158] Khederzadeh M, Beiranvand A. Identification and prevention of cascading failures in autonomous microgrid. IEEE Syst J Issue 99; 2015:1–8. [159] Babalola A, Belkacemi R, Zarrabian S. Real-time cascading failures prevention for multiple contingencies in smart grids through a multi-agent system. IEEE Trans Smart Grid 2016. Issue 99. [160] Bai L, Li F, Cui H, Jiang T, Sun H, Zhu J. Interval optimization based operating strategy for gas-electricity integrated energy systems considering demand response and wind uncertainty. Appl Energy 2016;167:270–9. [161] Nottrott A, Kleissl J, Washom B. Energy dispatch schedule optimization and cost benefit analysis for grid-connected, photovoltaic-battery storage systems. Renew Energy 2013;55:230–40. [162] Hong YY, Hsiao MC, Chang YR, Lee YD, Huang HC. Multiscenario underfrequency load shedding in a microgrid consisting of intermittent renewables. IEEE Trans Power Delivery 2013;28(3):1610–7. [163] Balaguer IJ, Lei Q, Yang S, Supatti U, Peng FZ. Control for grid-connected and intentional islanding operations of distributed power generation. IEEE Trans Ind Electron 2011;58(1):147–57. [164] Liu G, Xiao B, Starke M, Ceylan O, Tomsovic K. A robust load shedding strategy for microgrid islanding transition. In: 2016 IEEE/PES Transmission and Distribution Conference and Exposition (T&D); 2016. p. 1–5. [165] Oliveira DQ, de Souza AZ, Almeida AB, Santos MV, Lopes BIL, Marujo D. Microgrid management in emergency scenarios for smart electrical energy usage. In: 2015 IEEE Eindhoven PowerTech; 2015. p. 1–6. [166] Nejabatkhah F, Li YW. Overview of power management strategies of hybrid AC/ DC microgrid. IEEE Trans Power Electron 2015;30(12):7072–89. [167] Lidula NWA, Rajapakse AD. Voltage balancing and synchronization of microgrids with highly unbalanced loads. Renew Sustain Energy Rev 2014;31:907–20. [168] Assis TML, Taranto GN. Automatic reconnection from intentional islanding based on remote sensing of voltage and frequency signals. IEEE Trans Smart Grid 2012;3(4):1877–84. [169] Hanif M, Khadkikar V, Kanjiya P. Control and SRF-q based re-synchronization of a master DG for microgrids. In: 6th IEEE Power India International Conference (PIICON). IEEE; 2014. p. 1–6. [170] Ashabani M, Gooi HB, Guerrero JM. Designing high-order power-source synchronous current converters for islanded and grid-connected microgrids. Appl Energy 2018;219:370–84. [171] Vandoorn TL, Meersman B, De Kooning JD, Vandevelde L. Transition from islanded to grid-connected mode of microgrids with voltage-based droop control. IEEE Trans Power Syst 28(3), 2545–53. [172] Lee CT, Jiang RP, Cheng PT. A grid synchronization method for droop-controlled distributed energy resource converters. IEEE Trans Industry Appl 49(2);2013:954–962. [173] Micallef A, Apap M, Spiteri-Staines C, Guerrero JM. Single-phase microgrid with seamless transition capabilities between modes of operation. IEEE Trans Smart Grid 2015;6(6):2736–45. [174] Wang J, Chang NCP, Feng X, Monti A. Design of a generalized control algorithm for parallel inverters for smooth microgrid transition operation. IEEE Trans Ind Electron 2015;62(8):4900–14. [175] Lopes JP, Moreira CL, Resende FO. Microgrids black start and islanded operation. In: 15th Power systems computation conference (PSCC), Liege. August 2015. [176] Lopes JP, Moreira CL, Resende FO. Control strategies for microgrids black start and islanded operation. Int J Distribut Energy Resour 2015;1(3):241–61.

partitioning. Electr Power Syst Res 2016;141:202–9. [118] Cagnano A, Torelli F, Alfonzetti F, De Tuglie E. Can PV plants provide a reactive power ancillary service? A treat offered by an on-line controller. Renew Energy 2011;36(3):1047–52. [119] Calderaro V, Galdi V, Lamberti F, Piccolo A, Graditi G. Voltage support control of unbalanced distribution systems by reactive power regulation. In: IEEE PES innovative smart grid technologies, Europe, IEEE; October 2014. p. 1–5. [120] Cagnano A, De Tuglie E. A decentralized voltage controller involving PV generators based on Lyapunov theory. Renew Energy 2016;86:664–74. [121] Fallahzadeh-Abarghouei H, Hasanvand S, Nikoobakht A, Doostizadeh M. Decentralized and hierarchical voltage management of renewable energy resources in distribution smart grid. Int J Electr Power Energy Syst 2018;100:117–28. [122] Fallahzadeh-Abarghouei H, Nayeripour M, Hasanvand S, Waffenschmidt E. Online hierarchical and distributed method for voltage control in distribution smart grids. IET Gener Transm Distrib 2016;11(5):1223–32. [123] Vandoorn TL, De Kooning JD, Meersman B, Guerrero JM, Vandevelde L. Voltagebased control of a smart transformer in a microgrid. IEEE Trans Ind Electron 2013;60(4):1291–305. [124] Savaghebi M, Jalilian A, Vasquez JC, Guerrero JM. Secondary control scheme for voltage unbalance compensation in an islanded droop-controlled microgrid. IEEE Trans Smart Grid 2015;3(2):797–807. [125] Hirase Y, Abe K, Sugimoto K, Sakimoto K, Bevrani H, Ise T. A novel control approach for virtual synchronous generators to suppress frequency and voltage fluctuations in microgrids. Appl Energy 2018;210:699–710. [126] Moradi MH, Eskandari M, Hosseinian SM. Cooperative control strategy of energy storage systems and micro sources for stabilizing microgrids in different operation modes. Int J Electr Power Energy Syst 2016;78:390–400. [127] Mahmood H, Michaelson D, Jiang J. Accurate reactive power sharing in an islanded microgrid using adaptive virtual impedances. IEEE Trans Power Electron 2015;30(3):1605–17. [128] Tang X, Hu X, Li N, Deng W, Zhang G. A novel frequency and voltage control method for islanded microgrid based on multienergy storages. IEEE Trans Smart Grid 2016;7(1):410–9. [129] Simpson-Porco JW, Shafiee Q, Dörfler F, Vasquez JC, Guerrero JM, Bullo F. Secondary frequency and voltage control of islanded microgrids via distributed averaging. IEEE Trans Ind Electron 2011;62(11):7025–38. [130] Nasirian V, Shafiee Q, Guerrero JM, Lewis FL, Davoudi A. Droop-free distributed control for AC microgrids. IEEE Trans Power Electron 2016;31(2):1600–17. [131] Mohamed YARI, Radwan AA. Hierarchical control system for robust microgrid operation and seamless mode transfer in active distribution systems. IEEE Trans Smart Grid 2011;2(2):352–62. [132] Cai N, Mitra J. A multi-level control architecture for master-slave organized microgrids with power electronic interfaces. Electr Power Syst Res 2014;109:8–19. [133] Li X, Zhang H, Shadmand MB, Balog RS. Model predictive control of a voltagesource inverter with seamless transition between islanded and gridconnected operations. IEEE Trans Ind Electron 2017;64(10):7906–18. [134] Shadmand MB, Li X, Balog RS, Rub HA. Constrained decoupled power predictive controller for a single-phase grid-tied inverter. IET Renew Power Gener 2017;11(5):659–68. [135] Jin N, Gan C, Guo L. Predictive control of bidirectional voltage source converter with reduced current harmonics and flexible power regulation under unbalanced grid. In: EEE Transactions on Energy Conversion; December 2017. [136] Sajadian S, Ahmadi R. Model predictive control of dual-mode operations Z-source inverter: islanded and grid-connected. IEEE Trans Power Electron 2018;33(5):4488–97. [137] Katiraei F, Iravani MR, Lehn PW. Micro-grid autonomous operation during and subsequent to islanding process. IEEE Trans Power Delivery 2005;20(1):248–57. [138] Caldognetto T, Tenti P. Microgrids operation based on master–slave cooperative control. IEEE J Emerg Select Top Power Electron 2014;2(4):1081–8. [139] Meegahapola LG, Robinson D, Agalgaonkar AP, Perera S, Ciufo P. Microgrids of commercial buildings: strategies to manage mode transfer from grid connected to islanded mode. IEEE Trans Sustain Energy 2014;5(4):1337–47. [140] Etemadi AH, Iravani R. Supplementary mechanisms for smooth transition between control modes in a microgrid. Electr Power Syst Res 2017;142:249–57. [141] Chee SJ, Lee Y, Son YK, Sul SK, Lim C, Huh S. Seamless transfer strategy considering power balance in parallel operation. In: 2016 IEEE Energy Conversion Congress and Exposition (ECCE) . IEEE; 2016. p. 1–7. [142] Talapur GG, Suryawanshi H, Xu L, Shitole A. A reliable micro-grid with seamless transition between grid connected and islanded mode for residential community with enhanced power quality. In: IEEE Transactions on Industry Applications; 2018. [143] Barklund E, Pogaku N, Prodanovic M, Hernandez-Aramburo C, Green TC. Energy management in autonomous microgrid using stability-constrained droop control of inverters. IEEE Trans Power Electron 2008;23(5):2346–52. [144] Chen F, Chen M, Li Q, Meng K, Zheng Y, Guerrero JM, et al. Cost-based droop schemes for economic dispatch in islanded microgrids. IEEE Trans Smart Grid 2017;8(1):63–74. [145] Ma Y, Yang P, Guo H, Wang Y. Dynamic economic dispatch and control of a standalone microgrid in dongao island. J Electr Eng Technol 10(4);1432–1440. [146] Olivares DE, Cañizares CA, Kazerani M. A centralized energy management system

17

Applied Energy 258 (2020) 114039

A. Cagnano, et al.

December 2003 he has been Associate Professor at the Polytechnic of Bari. His research interests include power system analysis and control and reregulated markets.

[177] Li J, Su J, Yang X, Zhao T. Study on microgrid operation control and black start. In: 2011 4th International Conference on Electric Utility Deregulation and Restructuring and Power Technologies (DRPT), IEEE; 2011. p. 1652–55. [178] Xu Z, Yang P, Zheng Q, Zeng Z. Study on black start strategy of microgrid with PV and multiple energy storage systems. In: 2015 18th International Conference on Electrical Machines and Systems (ICEMS), IEEE; 2015. p. 402–408. [179] Moreira CL, Resende FO, Lopes JP. Using low voltage microgrids for service restoration. IEEE Trans Power Syst 2007;22(1):395–403. [180] Ding T, Lin Y, Bie Z, Chen C. A resilient microgrid formation strategy for load restoration considering master-slave distributed generators and topology reconfiguration. Appl Energy 2017;199:205–16.

Pierluigi Mancarella is Chair Professor of Electrical Power Systems at The University of Melbourne (Australia) and part-time Professor of Smart Energy Systems at The University of Manchester (UK). He received his MSc and PhD in Power Systems from the Politecnico di Torino (Italy) and then worked as a post-doc at Imperial College London (UK). Pierluigi has also held visiting positions at the University of Cambridge (UK), NREL (US), Tsinghua University (China), Ecole Centrale de Lille (France), and Universidad de Chile. His research interests include techno-economic modelling of integrated energy systems, energy planning under uncertainty, and reliability and resilience of future networks. He has been involved in/led around 50 research projects worldwide and is author of several books and of over 200 research publications. He is an Editor of the IEEE Transactions on Power Systems, IEEE Transactions on Smart Grid, and IEEE Systems Journal, and is the convenor of the CIGRE Working Group C6/C2.34 on “Flexibility from Distributed Energy Resources”. Pierluigi is an IEEE PES Distinguished Lecturer and holds the 2017 veski Innovation Fellowship by the Victorian Government. He recently led the Melbourne Energy Institute’s power system security assessment studies in support of the “Finkel Review” of the Australian Energy Market and was awarded a 2018 international Newton Prize for his work on power systems resilience.

Alessia Cagnano received B.Sc. and M.Sc. degree in Electrical Engineering from the Polytechnic of Bari (Italy) in 2005 and 2007 respectively. She received the Ph.D. degree in Electrical Engineering at the Polytechnic of Bari in 2011. Her research interests are in power system and microgrid analysis and control. Enrico De Tuglie received the M.Sc. and Ph.D. degrees in electrical engineering from the Polytechnic of Bari. He has been visiting scientist at the Pacific Northwest National Laboratory (PNNL) operated by Battelle under the Department Of Energy (USA). Since

18