- Email: [email protected]
Options for an autarkic operation of a communal power grid using a battery and renewable energies
Available online at www.sciencedirect.com
ScienceDirect ScienceDirect Energy Procedia 00 (2018) 000–000 Energy Procedia 00 (2018) 000–000
Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2018) 000–000 329–335 EnergyProcedia Procedia155 00 (2017) www.elsevier.com/locate/procedia
12th International Renewable Energy Storage Conference, IRES 2018 12th International Renewable Energy Storage Conference, IRES 2018
Options for an autarkic operation of a communal power grid using a Options for an autarkic operation of a communal power grid using a battery and renewable energies The 15th International Symposium on District Heating and Cooling battery and renewable energies a Silvan Rummenya,of Eberhard Waffenschmidt * Assessing the feasibility using the heat demand-outdoor a Silvan Rummeny , Eberhard Waffenschmidta* Cologne Institute for Renewable Energy (CIRE), Cologne University of Applied Sciences, Betzdorfer Str. 2, 50679 Cologne, Germany temperature function for a long-term district heat demand forecast Cologne Institute for Renewable Energy (CIRE), Cologne University of Applied Sciences, Betzdorfer Str. 2, 50679 Cologne, Germany a a
Abstract I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc Abstract WithaIN+ a growing of renewable energy, thePolicy structure of the electrical powerTécnico, grid structure should their distributed nature. Center share for Innovation, Technology and Research - Instituto Superior Av. Rovisco Paisreflect 1, 1049-001 Lisbon, Portugal b Veolia Recherche Innovation, Avenue Dreyfous Daniel, 78520 Limay, France A cellular grid structure meetenergy, this demand. In caseofof291 global blackout each cellstructure would even be able to their operate autarkic.nature. Then, With a growing share of would renewable the&structure the electrical power grid should reflect distributed c Systèmes Énergétiques et Environnement - IMT Atlantique, ruethe Alfred Kastler, 44300 Nantes, still a minimum level ofwould supply would be possible. An of innovative community in north of be Germany aims France toautarkic. pursue such A cellular gridDépartement structure meet this demand. In case global blackout each4cell would even able to operate Then,a concept. The local energy provider, Versorgungsbetriebe Bordesholm, is planning install large battery aims (12 MWh / +/ 8such MW)a still a minimum level of supply would be possible. An innovative community in tothe northa of Germany to pursue to provideThe control dailyVersorgungsbetriebe operation. In case ofBordesholm, a global blackout it should supplya the community in MWh combination with concept. local power energyduring provider, is planning to install large battery (12 / +/ 8 MW) available However, supply consisting two biogas generators 800the kWcommunity each and several photovoltaic to provideenergy controlsources. power during dailythe operation. In case of of a global blackout it shouldwith supply in combination with Abstract systems are not sufficient for an infinite autarkicconsisting supply. Therefore, this publication investigates possible operation time for available energy sources. However, the supply of two biogas generators with 800 kWthe each and several photovoltaic such an autarkic operation.for Typical self-supply times for this community from 3investigates to 20 hours.the possible operation time for systems are not sufficient an infinite autarkic supply. Therefore, thisrange publication District heatingoperation. networks Typical are commonly addressed in this the community literature asrange one from of the3 to most effective solutions for decreasing the such an autarkic self-supply times for 20 hours. greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat © 2018 The Authors. Published by Elsevier Ltd. Due to access the changed climate and building renovation policies, heat demand in the future could decrease, ©sales. 2018 The Authors. Published by Elsevier Ltd. This is an open article under theconditions CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) prolonging the investment return period. Selection peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage This is an and open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific of the –12th International Renewable The main scope of this paper is to assess the feasibility of usingcommittee the heat demand outdoor temperature functionEnergy for heatStorage demand Conference. Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 Conference. buildingscommunity that vary power in both construction and typology. Three weather scenarios medium, Keywords: grid; isolated grid period operation; renewable energy; cellular power grid; grid (low, forming battery; high) and three district renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were Keywords: community power grid; isolated grid operation; renewable energy; cellular power grid; grid forming battery; compared with results from a dynamic heat demand model, previously developed and validated by the authors. results showed that when only weather change is considered, the margin of error could be acceptable for some applications 1.The Introduction error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation 1.(the Introduction scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). With a growing share of renewable energy, the structure of the electrical power grid structure should reflect their The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the With a growing shareapproach of renewable energy,grid the divides structurethe ofgrid the electrical powercan gridoperate structure should[1]. reflect their distributed nature. The of a cellular in cells, which autarkic The basic decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and distributed nature. The approach of a cellular grid divides the grid in cells, which can operate autarkic [1]. The basic principle is to balance supply and demand deviations on the lowest grid level as possible. In case of global blackout renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the principle is towould balance supply and demand deviations on lowest grid levelthe as transmission possible. of global blackout certain be to operate autarkic. Byused synchronizing these cells grid structure can be and coupledcells scenarios). Theable values suggested could be to the modify the function parameters for In thecase scenarios considered, certain cells would be able to operate autarkic. By synchronizing these cells the transmission grid structure can be improve the accuracy of heat demand estimations. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of8275 the Scientific Committee of The 15th International Symposium on District Heating and * Corresponding author. Tel.: +49 221 2020 (office). Cooling. E-mail address: [email protected] * Corresponding author. Tel.: +49 221 8275 2020 (office).
E-mail address: [email protected] Keywords:©Heat Forecast; Climatebychange 1876-6102 2018demand; The Authors. Published Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Selection under responsibility of the scientific of the 12th International Renewable Energy Storage Conference. This is an and openpeer-review access article under the CC BY-NC-ND licensecommittee (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference. 10.1016/j.egypro.2018.11.045
330 2
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
recovered quickly. Commonly, distributed generators are shut down to prevent uncoordinated islanding grids. However, distributed system operators (DSO) are able to coordinate isolated micro grids. There are already various initiatives to apply and analyse the capability and applicability of isolated emergency operations of grid-tied micro grids. The Bavarian communities Wilpoldried and Niederschönenfeld want to apply this system. Wilpoldsried has a supply surplus by up to 500% of its load. The power is supplied by 6 MWp photovoltaic, 1 MW biogas, and 12 MW wind power. Within the project IREN2, the communal micro grid wants to apply switching to an isolated and grid tied operation without interruption [2]. The power supply in Niederschönenfeld includes a big hydro power plant. Primarily, in project LINDA an isolated emergency operation will installed including blackstart with this leading power plant. Within the communal micro grid supply is supported by biogas and photovoltaic power plants. By means of field tests with a load bank or consumers the impact and control of load steps is analyzed [3]. The initiative Micro grid Brooklyn investigates the emergency operation of Brooklyn, New York. Primarily, critical infrastructure will be supplied, followed by the supply of residential buildings. An isolated operation is provided by photovoltaic systems and batteries and load management. Combined heat and power plants can blackstart the micro grid. In daily grid tied operation power within the grid is traded by block-chain method [4]. As one of various local pioneers within the integrated European network, the communal power grid operator Versorgungsbetriebe Bordesholm wants to enable an isolated emergency operation. A big battery storage will supply the town without interruption as supervisory control [5]. First switching and isolated operation tests will be done in summer of 2019. However, Bordesholm is not in excess equipped with decentralized generators. It turned out that it cannot supply itself for unlimited time without restrictions. In this paper it is analysed, how long a self-sufficient islanding operation would be possible. For this purpose, feed-in profiles of the available generators and load profiles of the existing loads are compared taking the battery storage into account. With these data it is calculated for each hour of the year, how long an autarkic operation would be possible, if an islanding event would occur at that point of time. Details about the used algorithms can be found in [6]. Finally, it is analysed, how an existing biogas plant would have to be extended to allow an unlimited autarkic operation. 2. Communal power grid The communal power grid consist of three medium voltage branches, which can be looped together (see Figure 1). Each square represents a low voltage transformer station. The current demand and supply structure is shown in Figure 2. Two biogas generators with 800 kW each (at Positions 1.1 and 1.2) provide the main power generation. Additionally, several distributed photovoltaic (PV) systems support generation within the power grid. Their installed power can be derived from Figure 2. The loads mainly consist of households with some distributed business consumer. The load is rather equally distributed and there are no single power consumers dominating the demand. Even the street lighting has no significant contribution, because it was converted to LED recently. Figure 2 shows the max. current demand of consumers and the installed power of generators in every branch and substation. The demand data is obtained from drag pointer measurements in the low voltage transformer stations and therefore indicates the worst load case. The total demand of the communal grid is about 10 kA respectively 7 MW, the total supply is about 3.5 MW. The following time-dependent simulations are based on measured profiles in quarter hour increments for the year 2015. The biogas power plants operate nearly constantly with smaller changes of the power level for longer periods. Profiles of nine of the larger PV systems are measured. The profiles of the further smaller PV systems are derived from one of these profiles by scaling it to their installed power. Furthermore, load profiles of 47 large consumers are measured. These are besides others: street lighting, water supply, waste water treatment station, town hall, offices, hotels, schools. It turned out that the peaks of the load profiles relate reasonably to the peak power values shown in Figure 2. In addition, the total residual power flow at the transmission grid connection point is measured (see Figure 3). In most of the times the residual load is positive, which means that the consumption exceeds the available generation with renewable energy sources. Therefore, the autarkic operation time can be assumed as limited for the whole period.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
331 3
By subtracting the feed-in profiles and the known load profiles from the residual load the resulting power profile can be attributed to the households in the community.
Branch 1 Branch 2
Position 1.5
Branch 3 Position 3.1 Fig. 1: Topology of the medium voltage part of the investigated power grid.
Fig. 2: Maximum power feed in and demand at the nodes in the three branches of the medium voltage part of the investigated power grid.
Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
4 332
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335
Fig. 3: Residual load measured at the point of common coupling.
3. Self-sufficiency time The self-sufficiency time is discussed with three storage scenarios: 1. No battery: self-sufficient operation by means of negative residual loads (supply surpluses) 2. 5 MW/ 5 MWh: initial plan of the communal grid operator 3. 8 MW/ 12 MWh: current intended construction plan. For scenario 1 only times with negative residual load (feed-in excess) are considered. If the residual load becomes positive the self sufficient time is reached. For scenarios 2 and 3 the battery is taken into account. Figure 4 shows the method to calculate the period of selfsufficiency (ta) for any possible start time (t) of isolated operation is shown in Figure 4. During normal grid-tied operation, the battery is intended to provide primary reserve power. Therefore, an average state of charge (SoC) of 50% is assumed as the initial SOC (SOC(0)) for the calculation of the self sufficient time. During islanding operation, the battery takes over any residual load or excess power. For charging and discharging the battery, it is assumed that the SOC is linear to the energy of residual load (ERL). If a maximum SoC of 90% is reached, it is assumed that feedin sources are cut, but operation continues. If the SoC is below 10% (SOCmin) the system is assumed to shut down and the self sufficient time is reached. In this publication no controllable loads and no energy losses are considered. This way, for every point of time it was calculated, how long the community could supply itself. The results are shown in Figure 5 as time dependence and in Figure 6 as a sorted distribution.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
333 5
As the time dependence of the self-sufficient time shows (Figure 5) it is difficult to predict, when a long self-sufficient time is possible. In general, the probability for long times increases in summer due to the larger contribution of PV systems. A more general statement can be made from the sorted distribution in Figure 6: Without a battery, only during a small time of maximum 20 days per year, an autarkic operation is possible anyhow (see grey part in Figure 6). Therefore, for this community a battery is a precondition for any autarkic operation throughout the year. Typical self-supply times range from 3 h to 20 h for the larger 12 MWh battery. On a few times in the year (in total less than 15 days) the selfsufficient time is longer than 20 h. These are e.g. sunny days with low load, as exemplarily shown in in the lower part of Figure 5. As could be assumed a larger battery results in longer self-sufficient times. The battery capacity (Cbat) is relevant for this result, because for both batteries the maximum battery power was not needed at any time. The selfsufficient time seems more or less to scale linearly with the battery capacity. However, in detail it does not simply relate to a larger energy provision. In some cases, the larger energy provision can prevent falling into blackout by bridging times to a state, where again generation is available. Such an event happens early on Sat in Figure 5. At that time, the smaller battery (red) is not able to provide enough energy, while the larger battery (blue) can bride the supply until further energy sources are available.
Fig. 4: Calculation of the period of self-sufficiency.
Fig. 5: Calculated period of self-sufficiency for every possible point of time in a year (2015).
6334
Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000 Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335
Figure 6: Annual sorted period of self-sufficiency which can be attained (2015).
To provide Bordesholm completely autarkic, the missing residual load needs to be covered by decentralized generation. This adds up to an energy of 21 GWh. To achieve this amount of energy, it would be an option to increase the biogas generator to a power of 2.4 MW including the related increased generation of biogas. If it operates constantly throughout the year, it could cover the missing energy. Simulations show that then the remaining residual load can be covered by the battery. Concluding a full autarkic operation can be achieved with an upgrade of the biogas generator to 2.4 MW including related biogas energy provision. 4. Conclusion In most of the times the consumption exceeds the available generation with renewable energy sources, such that the autarkic operation time is limited in the investigated community Bordesholm. Without any battery, only on less than 20 days per year autarkic operation is possible for a short time. Typical self-supply times range from 3 h to 20 h for the planned 12 MWh battery. On a few times in the year (in total less than 15 days) the self-sufficient time is longer than 20 h. To provide Bordesholm completely autarkic, it would be an option to increase the biogas generator to a power of 2.4 MW including the related increased generation of biogas. The presented calculation model is well suited for a quick estimation of the potential period of self-sufficiency. Nevertheless, for an isolated operation grid operators have to assure voltage and power control and protection of the system. Additionally, critical processes like dynamic switching operations in and out of the isolated operation have to be managed successfully. Only then such a community grid can self-sufficiently be supplied in case of an emergency. Furthermore, the simulation can be improved by replacing the simplified battery model with e.g. the shepherd model [7]. This model can calculate the SOC behavior more realistically depending on the terminal voltage and experimentally measured discharge curve of a particular battery technology. 5. Acknowledgements The authors like to thank Mr. Frank Günther, Versorgungsbetriebe Bordesholm, and his co-workers for the provision of data about the grid and measurements of the profiles and for the opportunity to investigate a case of a real grid.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
335 7
References [1] VDE Verband der Elektrotechnik Elektronik Informationstechnik e.V., „Der zellulare Ansatz - Grundlage einer erfolgreichen, regionenübergreifenden Energiewende,“ Frankfurt am Main, 2015 [2] R. Körberle, K. Mayr, B. Rindt, T. Sowa, D. Buchstaller, A. Armstorfer und H. Biechl, „IREN2: Zukunftsfähige Netze zur Integration Regenerativer Energiesysteme,“ Von Smart Grids zu Smart Markets 2015 - Beiträge der ETG-Fachtagung, Kassel, 2015. [3] G. Kerber, M. Finkel, K. Schaarschmidt, C. J. Steinhart, M. Gratza und R. Witzmann, „Konzept für eine lokale Inselnetzversorgung mit dezentralen Erzeugungsanlagen bei großflächigen Stromausfällen,“ 14. Symposium Energieinnovation, Graz, Österreich, 2016. [4] Brooklyn Microgrid, „Brooklyn Microgrid Team's Response to NY Prize Community,“ New York, 2015. [5] Versorgungsbetriebe Bordesholm, „EU-Ausschreibung der Versorgungsbetriebe Bordesholm GmbH: "Stromversorgungsnetz-Inselbetrieb in Bordesholm aus Erneuerbarer Energie mit Batteriespeicher",“ 2017. [6] Silvan Faßbender, "Autarker Notbetrieb des Stromnetzes einer Gemeinde mit Erneuerbaren Energien", Master Thesis, TH-Köln, 24.Nov.2016. [7] C. M. Shepherd, Design of Primary and Secondary Cells – Part 2. An Equation Describing Battery Discharge. Journal of Electrochemical Sciety, vol. 112, pp. 657-664, January 1965
ScienceDirect ScienceDirect Energy Procedia 00 (2018) 000–000 Energy Procedia 00 (2018) 000–000
Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy (2018) 000–000 329–335 EnergyProcedia Procedia155 00 (2017) www.elsevier.com/locate/procedia
12th International Renewable Energy Storage Conference, IRES 2018 12th International Renewable Energy Storage Conference, IRES 2018
Options for an autarkic operation of a communal power grid using a Options for an autarkic operation of a communal power grid using a battery and renewable energies The 15th International Symposium on District Heating and Cooling battery and renewable energies a Silvan Rummenya,of Eberhard Waffenschmidt * Assessing the feasibility using the heat demand-outdoor a Silvan Rummeny , Eberhard Waffenschmidta* Cologne Institute for Renewable Energy (CIRE), Cologne University of Applied Sciences, Betzdorfer Str. 2, 50679 Cologne, Germany temperature function for a long-term district heat demand forecast Cologne Institute for Renewable Energy (CIRE), Cologne University of Applied Sciences, Betzdorfer Str. 2, 50679 Cologne, Germany a a
Abstract I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc Abstract WithaIN+ a growing of renewable energy, thePolicy structure of the electrical powerTécnico, grid structure should their distributed nature. Center share for Innovation, Technology and Research - Instituto Superior Av. Rovisco Paisreflect 1, 1049-001 Lisbon, Portugal b Veolia Recherche Innovation, Avenue Dreyfous Daniel, 78520 Limay, France A cellular grid structure meetenergy, this demand. In caseofof291 global blackout each cellstructure would even be able to their operate autarkic.nature. Then, With a growing share of would renewable the&structure the electrical power grid should reflect distributed c Systèmes Énergétiques et Environnement - IMT Atlantique, ruethe Alfred Kastler, 44300 Nantes, still a minimum level ofwould supply would be possible. An of innovative community in north of be Germany aims France toautarkic. pursue such A cellular gridDépartement structure meet this demand. In case global blackout each4cell would even able to operate Then,a concept. The local energy provider, Versorgungsbetriebe Bordesholm, is planning install large battery aims (12 MWh / +/ 8such MW)a still a minimum level of supply would be possible. An innovative community in tothe northa of Germany to pursue to provideThe control dailyVersorgungsbetriebe operation. In case ofBordesholm, a global blackout it should supplya the community in MWh combination with concept. local power energyduring provider, is planning to install large battery (12 / +/ 8 MW) available However, supply consisting two biogas generators 800the kWcommunity each and several photovoltaic to provideenergy controlsources. power during dailythe operation. In case of of a global blackout it shouldwith supply in combination with Abstract systems are not sufficient for an infinite autarkicconsisting supply. Therefore, this publication investigates possible operation time for available energy sources. However, the supply of two biogas generators with 800 kWthe each and several photovoltaic such an autarkic operation.for Typical self-supply times for this community from 3investigates to 20 hours.the possible operation time for systems are not sufficient an infinite autarkic supply. Therefore, thisrange publication District heatingoperation. networks Typical are commonly addressed in this the community literature asrange one from of the3 to most effective solutions for decreasing the such an autarkic self-supply times for 20 hours. greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat © 2018 The Authors. Published by Elsevier Ltd. Due to access the changed climate and building renovation policies, heat demand in the future could decrease, ©sales. 2018 The Authors. Published by Elsevier Ltd. This is an open article under theconditions CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) prolonging the investment return period. Selection peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage This is an and open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific of the –12th International Renewable The main scope of this paper is to assess the feasibility of usingcommittee the heat demand outdoor temperature functionEnergy for heatStorage demand Conference. Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 Conference. buildingscommunity that vary power in both construction and typology. Three weather scenarios medium, Keywords: grid; isolated grid period operation; renewable energy; cellular power grid; grid (low, forming battery; high) and three district renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were Keywords: community power grid; isolated grid operation; renewable energy; cellular power grid; grid forming battery; compared with results from a dynamic heat demand model, previously developed and validated by the authors. results showed that when only weather change is considered, the margin of error could be acceptable for some applications 1.The Introduction error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation 1.(the Introduction scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). With a growing share of renewable energy, the structure of the electrical power grid structure should reflect their The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the With a growing shareapproach of renewable energy,grid the divides structurethe ofgrid the electrical powercan gridoperate structure should[1]. reflect their distributed nature. The of a cellular in cells, which autarkic The basic decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and distributed nature. The approach of a cellular grid divides the grid in cells, which can operate autarkic [1]. The basic principle is to balance supply and demand deviations on the lowest grid level as possible. In case of global blackout renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the principle is towould balance supply and demand deviations on lowest grid levelthe as transmission possible. of global blackout certain be to operate autarkic. Byused synchronizing these cells grid structure can be and coupledcells scenarios). Theable values suggested could be to the modify the function parameters for In thecase scenarios considered, certain cells would be able to operate autarkic. By synchronizing these cells the transmission grid structure can be improve the accuracy of heat demand estimations. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of8275 the Scientific Committee of The 15th International Symposium on District Heating and * Corresponding author. Tel.: +49 221 2020 (office). Cooling. E-mail address: [email protected] * Corresponding author. Tel.: +49 221 8275 2020 (office).
E-mail address: [email protected] Keywords:©Heat Forecast; Climatebychange 1876-6102 2018demand; The Authors. Published Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Selection under responsibility of the scientific of the 12th International Renewable Energy Storage Conference. This is an and openpeer-review access article under the CC BY-NC-ND licensecommittee (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of the 12th International Renewable Energy Storage Conference. 10.1016/j.egypro.2018.11.045
330 2
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
recovered quickly. Commonly, distributed generators are shut down to prevent uncoordinated islanding grids. However, distributed system operators (DSO) are able to coordinate isolated micro grids. There are already various initiatives to apply and analyse the capability and applicability of isolated emergency operations of grid-tied micro grids. The Bavarian communities Wilpoldried and Niederschönenfeld want to apply this system. Wilpoldsried has a supply surplus by up to 500% of its load. The power is supplied by 6 MWp photovoltaic, 1 MW biogas, and 12 MW wind power. Within the project IREN2, the communal micro grid wants to apply switching to an isolated and grid tied operation without interruption [2]. The power supply in Niederschönenfeld includes a big hydro power plant. Primarily, in project LINDA an isolated emergency operation will installed including blackstart with this leading power plant. Within the communal micro grid supply is supported by biogas and photovoltaic power plants. By means of field tests with a load bank or consumers the impact and control of load steps is analyzed [3]. The initiative Micro grid Brooklyn investigates the emergency operation of Brooklyn, New York. Primarily, critical infrastructure will be supplied, followed by the supply of residential buildings. An isolated operation is provided by photovoltaic systems and batteries and load management. Combined heat and power plants can blackstart the micro grid. In daily grid tied operation power within the grid is traded by block-chain method [4]. As one of various local pioneers within the integrated European network, the communal power grid operator Versorgungsbetriebe Bordesholm wants to enable an isolated emergency operation. A big battery storage will supply the town without interruption as supervisory control [5]. First switching and isolated operation tests will be done in summer of 2019. However, Bordesholm is not in excess equipped with decentralized generators. It turned out that it cannot supply itself for unlimited time without restrictions. In this paper it is analysed, how long a self-sufficient islanding operation would be possible. For this purpose, feed-in profiles of the available generators and load profiles of the existing loads are compared taking the battery storage into account. With these data it is calculated for each hour of the year, how long an autarkic operation would be possible, if an islanding event would occur at that point of time. Details about the used algorithms can be found in [6]. Finally, it is analysed, how an existing biogas plant would have to be extended to allow an unlimited autarkic operation. 2. Communal power grid The communal power grid consist of three medium voltage branches, which can be looped together (see Figure 1). Each square represents a low voltage transformer station. The current demand and supply structure is shown in Figure 2. Two biogas generators with 800 kW each (at Positions 1.1 and 1.2) provide the main power generation. Additionally, several distributed photovoltaic (PV) systems support generation within the power grid. Their installed power can be derived from Figure 2. The loads mainly consist of households with some distributed business consumer. The load is rather equally distributed and there are no single power consumers dominating the demand. Even the street lighting has no significant contribution, because it was converted to LED recently. Figure 2 shows the max. current demand of consumers and the installed power of generators in every branch and substation. The demand data is obtained from drag pointer measurements in the low voltage transformer stations and therefore indicates the worst load case. The total demand of the communal grid is about 10 kA respectively 7 MW, the total supply is about 3.5 MW. The following time-dependent simulations are based on measured profiles in quarter hour increments for the year 2015. The biogas power plants operate nearly constantly with smaller changes of the power level for longer periods. Profiles of nine of the larger PV systems are measured. The profiles of the further smaller PV systems are derived from one of these profiles by scaling it to their installed power. Furthermore, load profiles of 47 large consumers are measured. These are besides others: street lighting, water supply, waste water treatment station, town hall, offices, hotels, schools. It turned out that the peaks of the load profiles relate reasonably to the peak power values shown in Figure 2. In addition, the total residual power flow at the transmission grid connection point is measured (see Figure 3). In most of the times the residual load is positive, which means that the consumption exceeds the available generation with renewable energy sources. Therefore, the autarkic operation time can be assumed as limited for the whole period.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
331 3
By subtracting the feed-in profiles and the known load profiles from the residual load the resulting power profile can be attributed to the households in the community.
Branch 1 Branch 2
Position 1.5
Branch 3 Position 3.1 Fig. 1: Topology of the medium voltage part of the investigated power grid.
Fig. 2: Maximum power feed in and demand at the nodes in the three branches of the medium voltage part of the investigated power grid.
Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
4 332
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335
Fig. 3: Residual load measured at the point of common coupling.
3. Self-sufficiency time The self-sufficiency time is discussed with three storage scenarios: 1. No battery: self-sufficient operation by means of negative residual loads (supply surpluses) 2. 5 MW/ 5 MWh: initial plan of the communal grid operator 3. 8 MW/ 12 MWh: current intended construction plan. For scenario 1 only times with negative residual load (feed-in excess) are considered. If the residual load becomes positive the self sufficient time is reached. For scenarios 2 and 3 the battery is taken into account. Figure 4 shows the method to calculate the period of selfsufficiency (ta) for any possible start time (t) of isolated operation is shown in Figure 4. During normal grid-tied operation, the battery is intended to provide primary reserve power. Therefore, an average state of charge (SoC) of 50% is assumed as the initial SOC (SOC(0)) for the calculation of the self sufficient time. During islanding operation, the battery takes over any residual load or excess power. For charging and discharging the battery, it is assumed that the SOC is linear to the energy of residual load (ERL). If a maximum SoC of 90% is reached, it is assumed that feedin sources are cut, but operation continues. If the SoC is below 10% (SOCmin) the system is assumed to shut down and the self sufficient time is reached. In this publication no controllable loads and no energy losses are considered. This way, for every point of time it was calculated, how long the community could supply itself. The results are shown in Figure 5 as time dependence and in Figure 6 as a sorted distribution.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
333 5
As the time dependence of the self-sufficient time shows (Figure 5) it is difficult to predict, when a long self-sufficient time is possible. In general, the probability for long times increases in summer due to the larger contribution of PV systems. A more general statement can be made from the sorted distribution in Figure 6: Without a battery, only during a small time of maximum 20 days per year, an autarkic operation is possible anyhow (see grey part in Figure 6). Therefore, for this community a battery is a precondition for any autarkic operation throughout the year. Typical self-supply times range from 3 h to 20 h for the larger 12 MWh battery. On a few times in the year (in total less than 15 days) the selfsufficient time is longer than 20 h. These are e.g. sunny days with low load, as exemplarily shown in in the lower part of Figure 5. As could be assumed a larger battery results in longer self-sufficient times. The battery capacity (Cbat) is relevant for this result, because for both batteries the maximum battery power was not needed at any time. The selfsufficient time seems more or less to scale linearly with the battery capacity. However, in detail it does not simply relate to a larger energy provision. In some cases, the larger energy provision can prevent falling into blackout by bridging times to a state, where again generation is available. Such an event happens early on Sat in Figure 5. At that time, the smaller battery (red) is not able to provide enough energy, while the larger battery (blue) can bride the supply until further energy sources are available.
Fig. 4: Calculation of the period of self-sufficiency.
Fig. 5: Calculated period of self-sufficiency for every possible point of time in a year (2015).
6334
Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000 Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335
Figure 6: Annual sorted period of self-sufficiency which can be attained (2015).
To provide Bordesholm completely autarkic, the missing residual load needs to be covered by decentralized generation. This adds up to an energy of 21 GWh. To achieve this amount of energy, it would be an option to increase the biogas generator to a power of 2.4 MW including the related increased generation of biogas. If it operates constantly throughout the year, it could cover the missing energy. Simulations show that then the remaining residual load can be covered by the battery. Concluding a full autarkic operation can be achieved with an upgrade of the biogas generator to 2.4 MW including related biogas energy provision. 4. Conclusion In most of the times the consumption exceeds the available generation with renewable energy sources, such that the autarkic operation time is limited in the investigated community Bordesholm. Without any battery, only on less than 20 days per year autarkic operation is possible for a short time. Typical self-supply times range from 3 h to 20 h for the planned 12 MWh battery. On a few times in the year (in total less than 15 days) the self-sufficient time is longer than 20 h. To provide Bordesholm completely autarkic, it would be an option to increase the biogas generator to a power of 2.4 MW including the related increased generation of biogas. The presented calculation model is well suited for a quick estimation of the potential period of self-sufficiency. Nevertheless, for an isolated operation grid operators have to assure voltage and power control and protection of the system. Additionally, critical processes like dynamic switching operations in and out of the isolated operation have to be managed successfully. Only then such a community grid can self-sufficiently be supplied in case of an emergency. Furthermore, the simulation can be improved by replacing the simplified battery model with e.g. the shepherd model [7]. This model can calculate the SOC behavior more realistically depending on the terminal voltage and experimentally measured discharge curve of a particular battery technology. 5. Acknowledgements The authors like to thank Mr. Frank Günther, Versorgungsbetriebe Bordesholm, and his co-workers for the provision of data about the grid and measurements of the profiles and for the opportunity to investigate a case of a real grid.
Silvan Rummeny et al. / Energy Procedia 155 (2018) 329–335 Rummeny, Waffenschmidt / Energy Procedia 00 (2018) 000–000
335 7
References [1] VDE Verband der Elektrotechnik Elektronik Informationstechnik e.V., „Der zellulare Ansatz - Grundlage einer erfolgreichen, regionenübergreifenden Energiewende,“ Frankfurt am Main, 2015 [2] R. Körberle, K. Mayr, B. Rindt, T. Sowa, D. Buchstaller, A. Armstorfer und H. Biechl, „IREN2: Zukunftsfähige Netze zur Integration Regenerativer Energiesysteme,“ Von Smart Grids zu Smart Markets 2015 - Beiträge der ETG-Fachtagung, Kassel, 2015. [3] G. Kerber, M. Finkel, K. Schaarschmidt, C. J. Steinhart, M. Gratza und R. Witzmann, „Konzept für eine lokale Inselnetzversorgung mit dezentralen Erzeugungsanlagen bei großflächigen Stromausfällen,“ 14. Symposium Energieinnovation, Graz, Österreich, 2016. [4] Brooklyn Microgrid, „Brooklyn Microgrid Team's Response to NY Prize Community,“ New York, 2015. [5] Versorgungsbetriebe Bordesholm, „EU-Ausschreibung der Versorgungsbetriebe Bordesholm GmbH: "Stromversorgungsnetz-Inselbetrieb in Bordesholm aus Erneuerbarer Energie mit Batteriespeicher",“ 2017. [6] Silvan Faßbender, "Autarker Notbetrieb des Stromnetzes einer Gemeinde mit Erneuerbaren Energien", Master Thesis, TH-Köln, 24.Nov.2016. [7] C. M. Shepherd, Design of Primary and Secondary Cells – Part 2. An Equation Describing Battery Discharge. Journal of Electrochemical Sciety, vol. 112, pp. 657-664, January 1965