- Email: [email protected]
On Defining Information and Communication Technology Requirements and Associated Challenges for ‘Energy and Comfort Active’ Buildings
Available online at www.sciencedirect.com
ScienceDirect Procedia Computer Science 32 (2014) 979 – 984
The 4th International Symposium on Frontiers in Ambient and Mobile Systems (FAMS-2014)
On defining information and communication technology requirements and associated challenges for ‘energy and comfort active’ buildings Kennedy O. Aduda*, Wim Zeiler, Gert Boxem, Timilehil Labeodan † Department of Built Environment Eindhoven University of Technology Netherlands
Abstract The intention of this article is to highlight considerations and ensuing challenges encountered in attempts to define communication requirements for the proposed multi-agent based building energy management systems. This is within the framework of ‘TKI-Smart Grid BEMS’ project which aims at developing new generation intelligent Building Energy Management Systems having capacity to interact with the utility power systems distribution network. The article identifies the development of comfort and energy active buildings as key to deriving maximum benefits from electrical smart grids for the built environment. These buildings require well specified information and communication technology for operational success. The paper is based on critical literature review. This is followed by a discussion on the challenges associated with specifying ICT infrastructure for multi-agents systems-based energy and comfort active buildings. © 2014 by Elsevier B.V. is an B.V. open access article under the CC BY-NC-ND license © 2014 Published The Authors. Published byThis Elsevier (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and peer-review under responsibility of Elhadi M. Shakshuki. Selection and Peer-review under responsibility of the Program Chairs. Keywords: ICT, smart grids, buildings, multi-agents system
* Corresponding author. Tel.: +31 40 247 2039 E-mail address: [email protected]
1877-0509 © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and Peer-review under responsibility of the Program Chairs. doi:10.1016/j.procs.2014.05.521
980
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
1. Introduction The now common Net Zero Energy Buildings (NZEB) policy has consequently introduced the need to integrate building level energy generations to power grids [1]. This has in turn led to emergence of ‘smart electrical grid’ characterised by multi-directional flow of power at low voltage levels and requirement for an elaborate informational exchange between stakeholders and components. Smart electrical grids is defined as ‘electricity network that can cost-efficiently integrate the behavior and actions of all users connected to it in efforts towards economic efficiency, energy sustainability and better service delivery to end users’ [2]. Implications of smart grid to buildings are: (i) flexible operations for energy advantage, and (ii) dynamic exchange of information by component subsystems and actors. Informational exchange includes: energy generation and consumption profiles, occupants’ comfort profiles and preferences, building behavior, market behavior and activity flows for different environmental scenarios. This leads to a key term used in this paper: ‘energy and comfort active buildings’; this term is used to describe buildings that proactively use indoor comfort requirements to define limits of dynamic interactions with electrical grid. General guidelines on indoor comfort requirements for buildings are often described in terms of thermal comfort, indoor air quality and visual comfort [3-5]. Operations in electrical smart grids involve interactive coordination amongst multiple actors, processes and devices, this requires agile and robust controls. Some scholars propose multiagent systems (MAS) for interactions between buildings and electrical smart grids [6, 7]; this is also important in enabling robust user interactions and real time decision making in buildings [16]. However, these have implications in terms of communication requirement such as need for greater robustness, proactivity and real time dynamics [8]. Little attention has been given towards addressing these new developments. This paper outlines the requirements and challenges associated with defining ICT for multi-agent based energy and comfort active buildings. Discussions are further divided into the following subsections: ICT requirements; the challenges, illustrative cases and conclusion. 2. ICT requirements Multi-agent based energy and comfort active buildings are essentially knowledge based systems whose operations are characterised by informational flows illustrated in Figure 1. Subsequently their ICT requirements can be categorised as unique to the 4 stages of informational flows (see Figure 1). UTILITY GRID SIDE COMMUNICATION
ICT PARAMETERS Data rates<50 kbps Latency<1 s Range >10000m
IAN/NAN/FAN concentrator / level 1 t
WAN /Level 2 aggregator
`
Grid Control Centre
BUILDING SIDE COMMUNICATIONS
BUILDING domain illustrations are with black outline, those for UTILITY side domain are in red outlines.
Fig. 1. Figure 1: Informational flows for energy & comfort active buildings in a smart grid.
Further details on the ICT requirements in relation to Figure 1are as discussed below. Use level communication (stage 1) This is stage connects the real world with the Building Management System. Information communicated at this stage is modal, singular in objective and is delivered as a signal. There are two categories of communications at this
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
981
stage. The first denoted as ‘1a’ in the Figure 1 deals with communication between occupants, appliances or building equipment to the sensors or actuators. The second category (denoted as 1b in the Figure 1) transmits collected information from real word to the Building Management System and vice versa. The performance parameters desired at this level are: latency of less than a minute, data transfer rate of less than 100 Mbps and a coverage of 100 m. Technologies used could be wired or wireless ones; wireless technologies are preferred at this level due to ability for re-use, cost effectiveness, high flexibility and elimination of cable trunking which is at times a nuisance [9-11]. The content of information communicated at this stage include: environmental conditions at room and other zone levels, user requirements, user behaviour, user preference and aggregated energy requirements at all levels of operations. Building Management Level Communication (stage 2) The BMS exists to improve indoor comfort whilst also reducing costs through prudent energy management strategies during building operations [12, 13]. In order to achieve this a web based BMS undertakes load profiling, forecast energy demand, undertake preliminary energy consumption analysis, undertake automated control and benchmark with other systems.This requires a database system, intelligent system and connectivity infrastructure. Connectivity is enabled by: x application based communication protocols such as LonWorks and BACnet which poll sensor based data from the real world to the database and relays action oriented instructions to the real world. x standard internet based protocols. The BMS is equipped with a communication middleware to mediate in interactions between different protocols. The ICT performance parameters for this level of operation remain similar to those at stage 1 except for latency which should be to the level desired for onwards transmission to the smart meter. Also, information content communicated at this stage are similar to those at building level with the addition to grid power systems information and energy statuses and local energy sources. Agents/Agent Platform-Utility Grid Level Communication (stage 3) Services at this stage are data and intelligence based. Communication at this stage is based on standard internet protocol. Two agent communications protocols may be used for building operations: ‘Foundation of Intelligent Physical Agents-Agent Communication Language’(FIPA-ACL) and ‘Knowledge Query and Manipulation Language’ (KQML) [14]. FIPA-ACL is based on the FIPA protocol and defines the environment and protocol in which the agents acts and communicate [15]. In addition it specifies the agents management system with regards to creation, deletion of agents and, access rights and privileges. KQML is designed en-suite with protocol to support high level of communication amongst intelligent agents and having capacity to allow for interactions between multiple intelligent systems [14, 15]. KQML defines the format of messages in such a way that agents are recognized, data channels are created and informational exchange occurs; this is done in 3 communication layers ( that is communication, message and content layers). The ICT requirements for this stage are similar to those on stage 2. However, the multi-agents based Building Energy Management System (SG-BEMS) is equipped with a middleware to decipher communication from the Smart Meter; this is often in radio frequency or GSM based protocol. The content of information exchanged at this stage is similar to that at building management level of communication. Utility grid side communication, stage 4 These occur between the utility side and the smart meter. The smart meter interconnects the utility side (grid) to the Customer and market layers are interconnected by the smart meter. The smart meter transforms metering concept from a post consumption billing gadget to a comprehensive and dynamic information collection and processing infrastructure collectively known as ‘automated metering infrastructure’ (AMI). On request or predefined schedule, the AMI measures, saves and analyses energy consumption data received from an elaborate communication system and metering devices [16]. The content of information exchanged at this stage is similar to that at building management level of communication with addition of asset and utility cost management information. Smart meter communicates information in protocols that are either GSM or radio frequency based to the agent; this is then interpreted to standard internet based protocol. The information is then received and analysed by the agents for building control. 3. The challenges Indications from existing cases
982
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
Functionally, performance parameters such as latency, data range, data transfer rate, flexibility, bandwidth, availability and priority be of significance for ICT in buildings connected to electrical smart grids [9, 17], see previous section. Closely related to this is the choice for a protocol. Table 2, indicates that technologically Dash 7, ZigBee, and WLAN would be adequate for the ICT tasks demanded. Table 1. Wireless ICT technology evaluation for at building level operations Technology & Performance ZigBee r: 250 kbps R: 100 m L:
Benefits
Disadvantages
References
Appropriate for enabling wireless networking between devices connected to the grid; Robust nature make them ideal for hostile environments prone to node failures; Can be easily used with actuators and sensors & are hence ideal for load and conditions monitoring; Low powered & energy efficient.
Current buildings need retrofitting to accommodate usage thus leading to high initial cost; Limited by memory size and range for large buildings.
[18-20]
WLAN r: 2-600 kbps R: up to100 m L:
Ideal for shared spectrum and noisy; Supports all IP based protocols; Achieves required security & data authentication requirements during communication; Effectively communicate in HAN as it accommodates several devices to operate at the same time; Devices are cost effective plug & play type. 1. Simple yet reliable; Appropriate for control lights and appliances in a house setting
Operates in an unlicensed ISM band whose spectrum is crowded; Limited bandwidth availability; Industrial grade Wi-Fi largely unavailable; Highly susceptible to interference.
[18-21]
Data range limits it to residential houses or very small office buildings..
[22]
Appropriate for enabling wireless networking between devices connected to the grid; Robust nature make them ideal for hostile environments prone to node failures; Can be easily used with actuators and sensors & are hence ideal for load and conditions monitoring; Low powered & energy efficient.
Current buildings need retrofitting to accommodate usage thus leading to high initial cost; Limited by memory size and range for large buildings.
[18-21]
Z-Wave r: 40 kbps R: up to 30 m L:
L: latency; r:data transfer rate R: range or coverage
Dash 7 seems most preferable as it surpasses others in terms of the number of devices it can connect, data range and data rate [21]. It is however noted that Dash 7 is yet to penetrate the market and ZigBee remains popular for building level operations. Broadly speaking, the challenges ICT for comfort and energy active buildings emanate either as a result of relatively new age technological practice in the area or associated complex nature of operations; the former is evidenced by use of numerous standards and protocols which are continually shifting [17, 18]. The complex manifestations in the ICT system for comfort and energy active buildings occurs as a result of the use of numerous devices across equally numerous operational systems and protocols. These imply differential requirements and hence heterogeneity in technology, protocols and standards must be embraced [18-20]. Added to this is the fact that there has been increasing move to realise real time data processing and real time optimisation [8]. This further underlies additional difficulties in terms of requirement for rapid processing of enormous amount of data[18]. In our discussion in section 2, it is noted that the information flow between the stages rely heavily the use of middleware software; this may be a cause of concern taken due to associated proprietary nature and consequential high initial costs. However, standardisation of operations at different stages of information flow remains problematic: for example at the interface between the smart meter and the agent platform no guideline exists for interpretation/conversion of grid data to agent usable data and vice versa. Existing guidelines such as 1) IEC/TR 62051: Electricity metering - Data exchange for meter reading, tariff and load control and 2) IEC/TR 62059: Electricity metering equipment are non-committal on specifics and offer very little to a bold proposal. This is not to mention that as at the end of 2013, no comprehensive standards for smart metering for buildings existed [23]. Also requiring attention are issues of security and reliability requirements for building operations which may become an issue with increased traffic and increased cyber security scares. This may render present technologies very costly. 4. Comments from illustrative cases
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
It is important to use some illustrative cases to highlight further challenges for energy and comfort active buildings. We discuss two illustrative cases that highlight some practical challenges associated with defining multi-agents based comfort and energy active buildings. Case 1: ‘HOMEBOTS’ system field test [24] HOMEBOTS is an agent-based energy management services for buildings using power line communication (PLC). In relation to this a series of tests was were performed in energy distribution area in the South-East of Sweden to evaluate real-time requirements for agent communication. The system connected various components a total of 28 electrical radiators able to communicate with each other over power line communication using LonTalk. The purpose of communication was to ensure market based direct load management. Communication test results indicates the following: (1) average latency of 1.29 seconds, (2) message lengths ranged from 10-160 bits, (3) response times for messages were from 0.540.99 seconds and (4) 13201 messages were successfully exchanged. PLC proved adequate for the task. Case 2: ABB-Zurich [25] The ABB experiment was meant to offer a proof of concept for design of a user centric Building Automation System based on multi-agents systems. The idea was to improve interactions with users and actualise dynamic control reconfigurations for building systems. The system used utilised KNX operation protocol for the building management system and ZigBee for the sensor and actuator network. Interprocess communication was XML based. Key results indicated that using a common communication bus by multiple devices led to congestion of messages and hence message delay. Also, multi-agents system need to operate on servers external to the BMS infrastructure to ensure continued basic functionality during server downtimes. Two main lessons from above cases are: (1) simple architectural frameworks with target specific devices connection such as the case with HOMEBOTS [24] has higher chances of success than generalised systems that are associated with technological capabilities such as in ABB-Zurich [25], (2) use of multiple communication buses and servers could decrease communication latency and by default increase response times. Few practical instances are reported for multiagent system in comfort and energy management in office buildings [26, 27, 28]; few practical learning opportunities therefore exist. However studies from other fields suggest that agent to agent communication in comfort and energy active buildings may be complicated by the following [14, 29]: x existence multiple intelligent systems, for example a grid based control system, comfort system and building energy management systems. This requires communication specification in a manner that allows interoperability amongst agents, x elaborate protocols for dialogue, negotiation and competition amongst agents to increase reliability, x administration of communication and intelligent processes such as modalities of handling inconsistencies and mismatches ensuing from different world views and ontological framework is necessary to guarantee timely and universal interpretation of the exchanged information. 5. Conclusions We have defined the requirements for ICT in comfort and energy active buildings within the context of smart grid operations and use of multi-agent systems. Little practical evidence exist for similar scenarios thus emphasising the need for practical studies to evaluate the performance of the ICT systems in energy and comfort active buildings. We have also highlighted some key challenges that are encountered in ICT systems for comfort and energy active systems. Central to these challenge is the heterogeneity in devices, technologies and systems for comfort and energy active buildings. It is acknowledged that heterogeneity in protocols, standards and technologies is essential for realisation of desired performances [18, 20]. Consequently open platforms, standards and protocols remain a key concern. This need to be considerate of both the present and future needs of the building and electrical grids. References 1. Li DH, Yang WL, Lam JC. Zero energy buildings and sustainable development implications-A review, Energy 2013; 54:1-10.
983
984
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
2. European Commission. Smart Grids: from Innovation to deployment, Communication from the Commission to the European Parliament. The Council, the European Economic and Social Committee and the Committee of the Regions, COM(2011)202 final, 2011. 3. ANSI/ASHRAE Standard 55, Thermal Environmental Conditions for Human Occupancy, ASHRAE Inc., Atlanta, 2004. 4. ANSI/ASHRAE Standard 62-2001-Ventilation for Acceptable Indoor Air Quality, American Society of Heating, Refrigerating and AirConditioning Engineers, 2001. 5. Mills E, Borg N. Trends in recommended illuminance levels: An international comparison. Journal of the Illuminating Engineering Society 1999; 8: 155-163 6. Pipattanasomporn, M., H. Feroze, and S. Rahman. Multi-agent systems in a distributed smart grid: Design and implementation. In Power Systems Conference and Exposition, 2009(PSCE'09), IEEE, 2009. p. 1-8. 7. Dounis AI, Christos C. Advanced control systems engineering for energy and comfort management in a building environment-A review. Renewable and Sustainable Energy Reviews 2009; 13: 1246-1261. 8. Klein L, Kwak JY, Kavulya G, Jazizadeh F, Becerik-Gerber B, Varakantham P, Tambe M. Coordinating occupant behavior for building energy and comfort management using multi-agent systems, Automation in Construction 2012; 22: 525-536. 9. Doh YM, Kim SJ, Hoe TW. A Trend Analysis of Smart Grid Technology: The Convergence of Electric Power Network and IT Technologies. ETTRENDS 2009;24:74-86. 10. Kuzlu M, Manisa P. Assessment of communication technologies and network requirements for different smart grid applications. In 2013 IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013. p. 1-6. 11. Al-Omar B, Al-Ali AR, Ahmed R, Landolsi T. Role of information and communication technologies in the smart grid, Journal of Emerging Trends in Computing and Information Sciences 2012; 3: 707-716. 12. Figueiredo J, and José S. A SCADA system for energy management in intelligent buildings. Energy and Buildings 2011; 49: 85-98. 13. Han J, Jeong YK, Lee I. Efficient Building Energy Management System Based on Ontology, Inference Rules, and Simulation, International Conference on Intelligent Building and Management, Proceedings of CSIST, IACSIT Press, Singapore, 2011; 5: 295-300 14. Berna-Koes M, Illah N, Katia S. Communication efficiency in multi-agent systems. Proceedings of ICRA'04. 2004 IEEE International Conference on Robotics and Automation 2004;3:2129-2134. 15. Fang L, Panos JA. On communication requirements for multi-agent consensus seeking. Networked Embedded Sensing and Control. Springer Berlin Heidelberg; 2006. p.53-67. 16. Siano, P. Demand response and smart grids-A survey. Renewable and Sustainable Energy Reviews 2014; 30: 461-478. 17. Jürgen Heiles, D1.3.1 Smart Grid Standardization Analysis Version 2.0 (February 2012), Nokia Siemens Networks. Helsinki, 2012. 18. Singh, A, Jyotsna B, and Debabrata D. Two tier communication architecture for smart meter. (COMSNETS), 2013 Fifth International Conference on Communication Systems and Networks. IEEE, 2013. 19. W. Kastner, G. Neugschwandtner, S. Soucek, H. M. Newmann. Communication systems for building automation and control, Proceedings of the IEEE 93 (2005): 1178-1203. 20. Khamphanchai W., M. Kuzlu, M. Pipattanasomporn, M. A smart distribution transformer management with multi-agent technologies, 2013 IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013. 21. Ma JR, Chen HH, Huang YR, MengW. Smart grid communication: Its challenges and opportunities. IEEE Transactions on Smart Grids 2013; 1:1-11. 22. Haase J. Wireless Network Standards for Building Automation, Embedded Systems for Smart Appliances and Energy Management, Springer New York; 2013. p. 53-65. 23. The Open Meter Consortium. Description of Current State of the Art Technologies and Protocols -General Overview of State of the Art Technological Alternatives Task 2.1.0, 2009. 24. Ygge FJ, Akkermans M, Andersson A, Krejic M, Boertjes E. The HOMEBOTS system and field test: A multi-commodity market for predictive power load management, Proceedings 4th Int. Conf. on the Practical Application of Intelligent Agents and Multi-Agent Technology PAAM-99, 1999. p. 363-382. 25. Yu D, Ettore F, Hadeli H. An intelligent building that listens to your needs. In Proceedings of the 28th Annual ACM Symposium on Applied Computing, ACM; 2013. p. 58-63 26. Kwak J, Pradeep V, Rajiv M, Yu-Han C, Milind T, Burcin B, and Wendy W. TESLA: An energy-saving agent that leverages schedule flexibility. In Proceedings of the 2013 international conference on Autonomous agents and multi-agent systems. International Foundation for Autonomous Agents and Multiagent Systems; 2013. p. 965-972 27. Clearwater SH, Bernardo A. H. Thermal markets for controlling building environments. Energy Engineering 1994; 91: 26-56. 28. Kok K, Warmer C, Kamphuis R. PowerMatcher: multiagent control in the electricity infrastructure. Proceedings AAMAS ’05, Int. conf. on Autonomous Agents and Multiagent Systems, volume industry track, New York; 2005:75–82. 29. Finin, T, Don M, Rich F. An overview of KQML: A knowledge query and manipulation language. Technical report, Department of Computer Science, University of Maryland Baltimore County, 1992.
ScienceDirect Procedia Computer Science 32 (2014) 979 – 984
The 4th International Symposium on Frontiers in Ambient and Mobile Systems (FAMS-2014)
On defining information and communication technology requirements and associated challenges for ‘energy and comfort active’ buildings Kennedy O. Aduda*, Wim Zeiler, Gert Boxem, Timilehil Labeodan † Department of Built Environment Eindhoven University of Technology Netherlands
Abstract The intention of this article is to highlight considerations and ensuing challenges encountered in attempts to define communication requirements for the proposed multi-agent based building energy management systems. This is within the framework of ‘TKI-Smart Grid BEMS’ project which aims at developing new generation intelligent Building Energy Management Systems having capacity to interact with the utility power systems distribution network. The article identifies the development of comfort and energy active buildings as key to deriving maximum benefits from electrical smart grids for the built environment. These buildings require well specified information and communication technology for operational success. The paper is based on critical literature review. This is followed by a discussion on the challenges associated with specifying ICT infrastructure for multi-agents systems-based energy and comfort active buildings. © 2014 by Elsevier B.V. is an B.V. open access article under the CC BY-NC-ND license © 2014 Published The Authors. Published byThis Elsevier (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and peer-review under responsibility of Elhadi M. Shakshuki. Selection and Peer-review under responsibility of the Program Chairs. Keywords: ICT, smart grids, buildings, multi-agents system
* Corresponding author. Tel.: +31 40 247 2039 E-mail address: [email protected]
1877-0509 © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and Peer-review under responsibility of the Program Chairs. doi:10.1016/j.procs.2014.05.521
980
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
1. Introduction The now common Net Zero Energy Buildings (NZEB) policy has consequently introduced the need to integrate building level energy generations to power grids [1]. This has in turn led to emergence of ‘smart electrical grid’ characterised by multi-directional flow of power at low voltage levels and requirement for an elaborate informational exchange between stakeholders and components. Smart electrical grids is defined as ‘electricity network that can cost-efficiently integrate the behavior and actions of all users connected to it in efforts towards economic efficiency, energy sustainability and better service delivery to end users’ [2]. Implications of smart grid to buildings are: (i) flexible operations for energy advantage, and (ii) dynamic exchange of information by component subsystems and actors. Informational exchange includes: energy generation and consumption profiles, occupants’ comfort profiles and preferences, building behavior, market behavior and activity flows for different environmental scenarios. This leads to a key term used in this paper: ‘energy and comfort active buildings’; this term is used to describe buildings that proactively use indoor comfort requirements to define limits of dynamic interactions with electrical grid. General guidelines on indoor comfort requirements for buildings are often described in terms of thermal comfort, indoor air quality and visual comfort [3-5]. Operations in electrical smart grids involve interactive coordination amongst multiple actors, processes and devices, this requires agile and robust controls. Some scholars propose multiagent systems (MAS) for interactions between buildings and electrical smart grids [6, 7]; this is also important in enabling robust user interactions and real time decision making in buildings [16]. However, these have implications in terms of communication requirement such as need for greater robustness, proactivity and real time dynamics [8]. Little attention has been given towards addressing these new developments. This paper outlines the requirements and challenges associated with defining ICT for multi-agent based energy and comfort active buildings. Discussions are further divided into the following subsections: ICT requirements; the challenges, illustrative cases and conclusion. 2. ICT requirements Multi-agent based energy and comfort active buildings are essentially knowledge based systems whose operations are characterised by informational flows illustrated in Figure 1. Subsequently their ICT requirements can be categorised as unique to the 4 stages of informational flows (see Figure 1). UTILITY GRID SIDE COMMUNICATION
ICT PARAMETERS Data rates<50 kbps Latency<1 s Range >10000m
IAN/NAN/FAN concentrator / level 1 t
WAN /Level 2 aggregator
`
Grid Control Centre
BUILDING SIDE COMMUNICATIONS
BUILDING domain illustrations are with black outline, those for UTILITY side domain are in red outlines.
Fig. 1. Figure 1: Informational flows for energy & comfort active buildings in a smart grid.
Further details on the ICT requirements in relation to Figure 1are as discussed below. Use level communication (stage 1) This is stage connects the real world with the Building Management System. Information communicated at this stage is modal, singular in objective and is delivered as a signal. There are two categories of communications at this
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
981
stage. The first denoted as ‘1a’ in the Figure 1 deals with communication between occupants, appliances or building equipment to the sensors or actuators. The second category (denoted as 1b in the Figure 1) transmits collected information from real word to the Building Management System and vice versa. The performance parameters desired at this level are: latency of less than a minute, data transfer rate of less than 100 Mbps and a coverage of 100 m. Technologies used could be wired or wireless ones; wireless technologies are preferred at this level due to ability for re-use, cost effectiveness, high flexibility and elimination of cable trunking which is at times a nuisance [9-11]. The content of information communicated at this stage include: environmental conditions at room and other zone levels, user requirements, user behaviour, user preference and aggregated energy requirements at all levels of operations. Building Management Level Communication (stage 2) The BMS exists to improve indoor comfort whilst also reducing costs through prudent energy management strategies during building operations [12, 13]. In order to achieve this a web based BMS undertakes load profiling, forecast energy demand, undertake preliminary energy consumption analysis, undertake automated control and benchmark with other systems.This requires a database system, intelligent system and connectivity infrastructure. Connectivity is enabled by: x application based communication protocols such as LonWorks and BACnet which poll sensor based data from the real world to the database and relays action oriented instructions to the real world. x standard internet based protocols. The BMS is equipped with a communication middleware to mediate in interactions between different protocols. The ICT performance parameters for this level of operation remain similar to those at stage 1 except for latency which should be to the level desired for onwards transmission to the smart meter. Also, information content communicated at this stage are similar to those at building level with the addition to grid power systems information and energy statuses and local energy sources. Agents/Agent Platform-Utility Grid Level Communication (stage 3) Services at this stage are data and intelligence based. Communication at this stage is based on standard internet protocol. Two agent communications protocols may be used for building operations: ‘Foundation of Intelligent Physical Agents-Agent Communication Language’(FIPA-ACL) and ‘Knowledge Query and Manipulation Language’ (KQML) [14]. FIPA-ACL is based on the FIPA protocol and defines the environment and protocol in which the agents acts and communicate [15]. In addition it specifies the agents management system with regards to creation, deletion of agents and, access rights and privileges. KQML is designed en-suite with protocol to support high level of communication amongst intelligent agents and having capacity to allow for interactions between multiple intelligent systems [14, 15]. KQML defines the format of messages in such a way that agents are recognized, data channels are created and informational exchange occurs; this is done in 3 communication layers ( that is communication, message and content layers). The ICT requirements for this stage are similar to those on stage 2. However, the multi-agents based Building Energy Management System (SG-BEMS) is equipped with a middleware to decipher communication from the Smart Meter; this is often in radio frequency or GSM based protocol. The content of information exchanged at this stage is similar to that at building management level of communication. Utility grid side communication, stage 4 These occur between the utility side and the smart meter. The smart meter interconnects the utility side (grid) to the Customer and market layers are interconnected by the smart meter. The smart meter transforms metering concept from a post consumption billing gadget to a comprehensive and dynamic information collection and processing infrastructure collectively known as ‘automated metering infrastructure’ (AMI). On request or predefined schedule, the AMI measures, saves and analyses energy consumption data received from an elaborate communication system and metering devices [16]. The content of information exchanged at this stage is similar to that at building management level of communication with addition of asset and utility cost management information. Smart meter communicates information in protocols that are either GSM or radio frequency based to the agent; this is then interpreted to standard internet based protocol. The information is then received and analysed by the agents for building control. 3. The challenges Indications from existing cases
982
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
Functionally, performance parameters such as latency, data range, data transfer rate, flexibility, bandwidth, availability and priority be of significance for ICT in buildings connected to electrical smart grids [9, 17], see previous section. Closely related to this is the choice for a protocol. Table 2, indicates that technologically Dash 7, ZigBee, and WLAN would be adequate for the ICT tasks demanded. Table 1. Wireless ICT technology evaluation for at building level operations Technology & Performance ZigBee r: 250 kbps R: 100 m L:
Benefits
Disadvantages
References
Appropriate for enabling wireless networking between devices connected to the grid; Robust nature make them ideal for hostile environments prone to node failures; Can be easily used with actuators and sensors & are hence ideal for load and conditions monitoring; Low powered & energy efficient.
Current buildings need retrofitting to accommodate usage thus leading to high initial cost; Limited by memory size and range for large buildings.
[18-20]
WLAN r: 2-600 kbps R: up to100 m L:
Ideal for shared spectrum and noisy; Supports all IP based protocols; Achieves required security & data authentication requirements during communication; Effectively communicate in HAN as it accommodates several devices to operate at the same time; Devices are cost effective plug & play type. 1. Simple yet reliable; Appropriate for control lights and appliances in a house setting
Operates in an unlicensed ISM band whose spectrum is crowded; Limited bandwidth availability; Industrial grade Wi-Fi largely unavailable; Highly susceptible to interference.
[18-21]
Data range limits it to residential houses or very small office buildings..
[22]
Appropriate for enabling wireless networking between devices connected to the grid; Robust nature make them ideal for hostile environments prone to node failures; Can be easily used with actuators and sensors & are hence ideal for load and conditions monitoring; Low powered & energy efficient.
Current buildings need retrofitting to accommodate usage thus leading to high initial cost; Limited by memory size and range for large buildings.
[18-21]
Z-Wave r: 40 kbps R: up to 30 m L:
L: latency; r:data transfer rate R: range or coverage
Dash 7 seems most preferable as it surpasses others in terms of the number of devices it can connect, data range and data rate [21]. It is however noted that Dash 7 is yet to penetrate the market and ZigBee remains popular for building level operations. Broadly speaking, the challenges ICT for comfort and energy active buildings emanate either as a result of relatively new age technological practice in the area or associated complex nature of operations; the former is evidenced by use of numerous standards and protocols which are continually shifting [17, 18]. The complex manifestations in the ICT system for comfort and energy active buildings occurs as a result of the use of numerous devices across equally numerous operational systems and protocols. These imply differential requirements and hence heterogeneity in technology, protocols and standards must be embraced [18-20]. Added to this is the fact that there has been increasing move to realise real time data processing and real time optimisation [8]. This further underlies additional difficulties in terms of requirement for rapid processing of enormous amount of data[18]. In our discussion in section 2, it is noted that the information flow between the stages rely heavily the use of middleware software; this may be a cause of concern taken due to associated proprietary nature and consequential high initial costs. However, standardisation of operations at different stages of information flow remains problematic: for example at the interface between the smart meter and the agent platform no guideline exists for interpretation/conversion of grid data to agent usable data and vice versa. Existing guidelines such as 1) IEC/TR 62051: Electricity metering - Data exchange for meter reading, tariff and load control and 2) IEC/TR 62059: Electricity metering equipment are non-committal on specifics and offer very little to a bold proposal. This is not to mention that as at the end of 2013, no comprehensive standards for smart metering for buildings existed [23]. Also requiring attention are issues of security and reliability requirements for building operations which may become an issue with increased traffic and increased cyber security scares. This may render present technologies very costly. 4. Comments from illustrative cases
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
It is important to use some illustrative cases to highlight further challenges for energy and comfort active buildings. We discuss two illustrative cases that highlight some practical challenges associated with defining multi-agents based comfort and energy active buildings. Case 1: ‘HOMEBOTS’ system field test [24] HOMEBOTS is an agent-based energy management services for buildings using power line communication (PLC). In relation to this a series of tests was were performed in energy distribution area in the South-East of Sweden to evaluate real-time requirements for agent communication. The system connected various components a total of 28 electrical radiators able to communicate with each other over power line communication using LonTalk. The purpose of communication was to ensure market based direct load management. Communication test results indicates the following: (1) average latency of 1.29 seconds, (2) message lengths ranged from 10-160 bits, (3) response times for messages were from 0.540.99 seconds and (4) 13201 messages were successfully exchanged. PLC proved adequate for the task. Case 2: ABB-Zurich [25] The ABB experiment was meant to offer a proof of concept for design of a user centric Building Automation System based on multi-agents systems. The idea was to improve interactions with users and actualise dynamic control reconfigurations for building systems. The system used utilised KNX operation protocol for the building management system and ZigBee for the sensor and actuator network. Interprocess communication was XML based. Key results indicated that using a common communication bus by multiple devices led to congestion of messages and hence message delay. Also, multi-agents system need to operate on servers external to the BMS infrastructure to ensure continued basic functionality during server downtimes. Two main lessons from above cases are: (1) simple architectural frameworks with target specific devices connection such as the case with HOMEBOTS [24] has higher chances of success than generalised systems that are associated with technological capabilities such as in ABB-Zurich [25], (2) use of multiple communication buses and servers could decrease communication latency and by default increase response times. Few practical instances are reported for multiagent system in comfort and energy management in office buildings [26, 27, 28]; few practical learning opportunities therefore exist. However studies from other fields suggest that agent to agent communication in comfort and energy active buildings may be complicated by the following [14, 29]: x existence multiple intelligent systems, for example a grid based control system, comfort system and building energy management systems. This requires communication specification in a manner that allows interoperability amongst agents, x elaborate protocols for dialogue, negotiation and competition amongst agents to increase reliability, x administration of communication and intelligent processes such as modalities of handling inconsistencies and mismatches ensuing from different world views and ontological framework is necessary to guarantee timely and universal interpretation of the exchanged information. 5. Conclusions We have defined the requirements for ICT in comfort and energy active buildings within the context of smart grid operations and use of multi-agent systems. Little practical evidence exist for similar scenarios thus emphasising the need for practical studies to evaluate the performance of the ICT systems in energy and comfort active buildings. We have also highlighted some key challenges that are encountered in ICT systems for comfort and energy active systems. Central to these challenge is the heterogeneity in devices, technologies and systems for comfort and energy active buildings. It is acknowledged that heterogeneity in protocols, standards and technologies is essential for realisation of desired performances [18, 20]. Consequently open platforms, standards and protocols remain a key concern. This need to be considerate of both the present and future needs of the building and electrical grids. References 1. Li DH, Yang WL, Lam JC. Zero energy buildings and sustainable development implications-A review, Energy 2013; 54:1-10.
983
984
Kennedy O. Aduda et al. / Procedia Computer Science 32 (2014) 979 – 984
2. European Commission. Smart Grids: from Innovation to deployment, Communication from the Commission to the European Parliament. The Council, the European Economic and Social Committee and the Committee of the Regions, COM(2011)202 final, 2011. 3. ANSI/ASHRAE Standard 55, Thermal Environmental Conditions for Human Occupancy, ASHRAE Inc., Atlanta, 2004. 4. ANSI/ASHRAE Standard 62-2001-Ventilation for Acceptable Indoor Air Quality, American Society of Heating, Refrigerating and AirConditioning Engineers, 2001. 5. Mills E, Borg N. Trends in recommended illuminance levels: An international comparison. Journal of the Illuminating Engineering Society 1999; 8: 155-163 6. Pipattanasomporn, M., H. Feroze, and S. Rahman. Multi-agent systems in a distributed smart grid: Design and implementation. In Power Systems Conference and Exposition, 2009(PSCE'09), IEEE, 2009. p. 1-8. 7. Dounis AI, Christos C. Advanced control systems engineering for energy and comfort management in a building environment-A review. Renewable and Sustainable Energy Reviews 2009; 13: 1246-1261. 8. Klein L, Kwak JY, Kavulya G, Jazizadeh F, Becerik-Gerber B, Varakantham P, Tambe M. Coordinating occupant behavior for building energy and comfort management using multi-agent systems, Automation in Construction 2012; 22: 525-536. 9. Doh YM, Kim SJ, Hoe TW. A Trend Analysis of Smart Grid Technology: The Convergence of Electric Power Network and IT Technologies. ETTRENDS 2009;24:74-86. 10. Kuzlu M, Manisa P. Assessment of communication technologies and network requirements for different smart grid applications. In 2013 IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013. p. 1-6. 11. Al-Omar B, Al-Ali AR, Ahmed R, Landolsi T. Role of information and communication technologies in the smart grid, Journal of Emerging Trends in Computing and Information Sciences 2012; 3: 707-716. 12. Figueiredo J, and José S. A SCADA system for energy management in intelligent buildings. Energy and Buildings 2011; 49: 85-98. 13. Han J, Jeong YK, Lee I. Efficient Building Energy Management System Based on Ontology, Inference Rules, and Simulation, International Conference on Intelligent Building and Management, Proceedings of CSIST, IACSIT Press, Singapore, 2011; 5: 295-300 14. Berna-Koes M, Illah N, Katia S. Communication efficiency in multi-agent systems. Proceedings of ICRA'04. 2004 IEEE International Conference on Robotics and Automation 2004;3:2129-2134. 15. Fang L, Panos JA. On communication requirements for multi-agent consensus seeking. Networked Embedded Sensing and Control. Springer Berlin Heidelberg; 2006. p.53-67. 16. Siano, P. Demand response and smart grids-A survey. Renewable and Sustainable Energy Reviews 2014; 30: 461-478. 17. Jürgen Heiles, D1.3.1 Smart Grid Standardization Analysis Version 2.0 (February 2012), Nokia Siemens Networks. Helsinki, 2012. 18. Singh, A, Jyotsna B, and Debabrata D. Two tier communication architecture for smart meter. (COMSNETS), 2013 Fifth International Conference on Communication Systems and Networks. IEEE, 2013. 19. W. Kastner, G. Neugschwandtner, S. Soucek, H. M. Newmann. Communication systems for building automation and control, Proceedings of the IEEE 93 (2005): 1178-1203. 20. Khamphanchai W., M. Kuzlu, M. Pipattanasomporn, M. A smart distribution transformer management with multi-agent technologies, 2013 IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013. 21. Ma JR, Chen HH, Huang YR, MengW. Smart grid communication: Its challenges and opportunities. IEEE Transactions on Smart Grids 2013; 1:1-11. 22. Haase J. Wireless Network Standards for Building Automation, Embedded Systems for Smart Appliances and Energy Management, Springer New York; 2013. p. 53-65. 23. The Open Meter Consortium. Description of Current State of the Art Technologies and Protocols -General Overview of State of the Art Technological Alternatives Task 2.1.0, 2009. 24. Ygge FJ, Akkermans M, Andersson A, Krejic M, Boertjes E. The HOMEBOTS system and field test: A multi-commodity market for predictive power load management, Proceedings 4th Int. Conf. on the Practical Application of Intelligent Agents and Multi-Agent Technology PAAM-99, 1999. p. 363-382. 25. Yu D, Ettore F, Hadeli H. An intelligent building that listens to your needs. In Proceedings of the 28th Annual ACM Symposium on Applied Computing, ACM; 2013. p. 58-63 26. Kwak J, Pradeep V, Rajiv M, Yu-Han C, Milind T, Burcin B, and Wendy W. TESLA: An energy-saving agent that leverages schedule flexibility. In Proceedings of the 2013 international conference on Autonomous agents and multi-agent systems. International Foundation for Autonomous Agents and Multiagent Systems; 2013. p. 965-972 27. Clearwater SH, Bernardo A. H. Thermal markets for controlling building environments. Energy Engineering 1994; 91: 26-56. 28. Kok K, Warmer C, Kamphuis R. PowerMatcher: multiagent control in the electricity infrastructure. Proceedings AAMAS ’05, Int. conf. on Autonomous Agents and Multiagent Systems, volume industry track, New York; 2005:75–82. 29. Finin, T, Don M, Rich F. An overview of KQML: A knowledge query and manipulation language. Technical report, Department of Computer Science, University of Maryland Baltimore County, 1992.