MicroGrid Laboratory Facilities

Transcription

1 1 MicroGrid aboratory Facilities M. Barnes, A. Dimeas, A. Engler, C. Fitzer,. Hatziargyriou, C. Jones, S. Papathanassiou, M. Vandenbergh Abstract Within the frame of the European project Microgrids several test set-ups have been installed or enlarged at different laboratories. Three of them are presented in this contribution: a specially designed single phase system of the TUA with agent control software, the DeMoTec at ISET, which is a general test site for DER and the flywheel test rig, designed by the University of Manchester. These three sites allow the tests of different components, control strategies and different storage technologies. Though this contribution is aimed at the presentation of the facilities themselves, some test results are included. turbine are connected to the AC grid via fast-acting DC/AC power converters. The battery converter in particular is suitably controlled to permit the operation of the system either interconnected to the V network (grid-tied), or in stand-alone (island) mode, with a seamless transfer from one mode to the other. A. System description The configuration of the microgrid system is shown in Fig. 1, along with a picture of the actual installation. Index Terms laboratories, DER, battery inverter, flywheel, wind turbine W I. ITRODUCTIO ithin the frame of the European project Microgrids several test facilities have been installed or enlarged at different laboratories. Three of them are presented in this contribution: a specially designed single phase system of the TUA with agent control software, the DeMoTec at ISET, which is a general test site for DER and a flywheel test rig, at the University of Manchester. These three sites allow the tests of different components, control strategies and different storage technologies. This paper presents the facilities and illustrates their behaviour with some test results. II. TEST FACIITY AT TUA The laboratory-scale microgrid system, installed at the ational Technical University of Athens, comprises two PV generators, one wind turbine, a battery energy storage, controllable loads and a controlled interconnection to the local V grid. The battery unit, the PV generators and the wind The authors wish to express their gratitude to the European Commission for the financial support provided to this work within the MicroGrids project (EK5-CT ). M. Barnes, C. Fitzer and C. Jones are with the School of Electrical and Electronic Engineering, University of Manchester, M60 1QD, Manchester, England ( m.barnes@manchester.ac.uk, chris.fitzer@manchester.ac.uk and c.jones@manchester.ac.uk ) A. Engler and M. Vandenbergh are with the Department of Engineering and Power Electronics of the Institut für Solare Energieversorgungstechnik (ISET) e. V., Königstor 59, D Kassel Germany ( s: aengler@iset.uni-kassel.de and mvandenbergh@iset.uni-kassel.de). Hatziargyriou, S. Papathanassiou and A. Dimeas are with the Department of Electrical and Computer Engineering, ational Technical University of Athens (TUA), 9 Iroon Polytechniou str., Athens, Greece ( s: nh@power.ece.ntua.gr, st@power.ece.ntua.gr and adimeas@power.ece.ntua.gr) Fig. 1: Main switchboard of the TUA system. It is a modular system, comprising a wind turbine and a PV generator as the primary source of power, along with a second smaller PV module. All microsources are interfaced to the single-phase AC bus via DC/AC inverters. A battery bank is

2 2 also included, interfaced to the AC system via a bi-directional PWM voltage source converter. The microgrid is connected to the local V grid, as shown in Fig. 1. The central component of the microgrid system is the battery inverter, which regulates the voltage and frequency when the system operates in island mode, taking over the control of active and reactive power. The battery inverter operates in voltage control mode (regulating the magnitude and phase/frequency of its output voltage), acting as a gridforming unit, when the microgrid operates in island mode, i.e. setting the voltage and frequency of the system. When the microgrid operates in parallel to the grid, in which case the latter defines the operating frequency and voltage, the inverter operates as a grid-following unit. The grid-connection scheme is shown in Fig. 2. In this configuration, the transfer from island to grid-tied mode and vice versa is controlled by the relays K3 and K6 of the battery inverter, the former of which controls the operation of the interconnection switch S. Sensing of grid voltage presence is made via the Diesel synch. input of the inverter, which is wired in series to relay K6. With this scheme, the transfer from one mode of operation to the other is now uninterruptible under normal operating conditions (no grid failure). However, a function is still needed to be implemented for the fast detection of abnormal grid conditions and immediate isolation of the microgrid. ACoutlet K3 F2 K6 Diesel synh. input Sunny Island AC Bus 2A Utility grid Grid Fig. 2: Grid connection scheme of the TUA system B. MultiAgent System for MicroGrid Operation In this section, the capabilities offered by the MultiAgent System (MAS) technology in the operation of the MicroGrid are presented [9]. The use of MAS technology can solve a number of specific operational problems: Small DG (Distributed Generation) units have different owners, so centralized control is difficult. Several decisions should be taken locally. ack of dedicated communication facilities. Microgrids will operate in a liberalized market, so market decisions of the controller of each unit should have a certain degree of «intelligence». The local DG units have also other tasks besides selling power to the network: producing heat for local installations, keeping the voltage locally at a certain level or providing a backup system for critical local loads in case of a failure of the main system. These tasks demonstrate the importance of the distributed control and autonomous operation. In the application developed for the Microgrids MAS there are 4 kinds of agents: Production Unit: This agent controls the Battery Inverter of the Microgrid. The main tasks of this agent are to control the overall status of the batteries and to adjust the power flow depending on the Market Condition (prices). Consumption Unit: This agent represents the controllable loads in the system. It knows the current demand and makes estimations of the energy demand for the next 15 minutes. Every 15 minutes it makes bids to the available Production Units in order to cover the estimated needs. Power System: This agent represents the Main Grid to which the Microgrid is connected. According to the Market Model adopted, the Power System Agent announces to all participants the Selling and Buying prices. It does not participate in the market operation since it is obliged to buy or sell any amount of energy asked for (as long as there are no security issues for the network). MGCC: This agent has only coordinating tasks. It announces the beginning and the end of a negotiation for a specific period and records final power exchanges between the agents in every period. The development of the application is in conformity with the standards proposed by the international Foundation of intelligent Physical Agents (FIPA) [4]. This organization aims to standardize the development of such systems, especially in the area of communication between the agents and the organization of the MAS. For our implementation, the Java Agent Development Framework (Jade) 3.0 platform is selected [5]. Jade is a Java-based tool for developing MAS systems. C. Controllable load For the effective implementation of the MAS technology in the microgrid, a controllable load is required, which will be linked to the Consumption Unit agent. The Consumption Unit agent will use measurements in order to estimate the consumption and make more realistic bids. Furthermore, this agent will have the ability to control the load, applying limitations depending on the market status or the MicroGrid security considerations. So a controllable load scheme has been implemented, using various load types and a relay panel, controlled either manually or remotely, by a PC (Programmable ogic Controller) or a suitable PC-Card. The relays are interfaced to the PC or to the PC Card via a 25-pin parallel port, to facilitate the individual control of each relay and its associated load. The panel is sufficient for controlling up to eight different loads of rated current up to 16A. D. Monitoring and control system

3 3 Τwo monitoring and control systems were implemented, developed in WinCC and abview environments. These systems provide, in the first place, measurements from the battery, PV and wind turbine inverters (voltage, current and frequency, state of the batteries etc.). In addition, they can alter the settings of the battery inverter droop controller, in order to regulate its active and reactive power output, and can also control the synchronization of the system to the main grid. Fig. 3: WinCC control panel III. DEMOTEC The DeMoTec promotes design, development and presentation of systems for the utilization of renewable energies and the rational use of energy. ISET and the University of Kassel s Institute for Electrical Energy Technology (IEE) have joined forces with companies and other research institutions to perform experiments and demonstrate the state of the art. The latest R&D findings and products, which are mainly related to the generation or consumption of electrical energy, are demonstrated. A supply technology, which allows series-produced modules from different manufacturers to be assembled together, is presented. By working closely together with companies, ISET develops the basic systems functions required and contributes to the standardization of interfaces. In order to guarantee the suitability of the components, whole supply systems are set up in DeMoTec and compatibility tests are performed. ISET also checks and optimizes power supply systems with dynamic load profiles and long-term tests, in order to improve reliability. DeMoTec mainly focuses on electrification with renewable energies using modularly expandable and grid-compatible hybrid power supply systems. A step-by-step expansion of such power supply systems is demonstrated for applications in developing countries, starting with small isolated systems and progressing to grid connected power supply units. The supply units may be connected via a flexible crossbar distributor. Remote monitoring and control systems are developed and presented in DeMoTec, which allow the economic operation of such modularly structured supply systems e.g. by independent power producers (IPP). To investigate basic issues of distributed generation in interconnected grids, currently additional low-voltage distribution grids, grid simulators and corresponding decentralized generators are being set up. A DeMoTec master display is being used to monitor the operations of a widely dispersed wind power plant system, which comprises about 80 representatively, selected systems throughout Germany. In this master display, moreover, the remote monitoring of remote isolated systems in Greece and Spain as well as the control of active low-voltage grids can Fig. 4: ayout of DeMoTec facility at ISET

4 4 also be demonstrated. For investigations in the test field a portable energy container can be used. With its PV-diesel hybrid system the operation of small wind mills at an island grid and the evaluation of the monitored data are possible. The total available generation capacity is approximately 200 kw. All generators and loads can be connected via a central crossbar switch cabinet to a local grid. Up to three independent grids can be realized simultaneously. These grids may be coupled via a medium voltage network simulator to study the effect of interconnected micro-grids. Control of the connection of the components to the grid, data acquisition, and visualization is managed by a software for visualization and industrial process control. The communication needed for this purpose is done via an Interbus-S control line. Extra space is available to integrate custom devices on request. In order to enable a common control of the generators and to enable a monitoring of the operating states of the system a Supervisory Control and Data Acquisition System (SCADA) for the laboratory network was developed. XM-RPC was selected as communication protocol between the generators, the communication is done via a separate Ethernet communication line. island operation and the transition from island to interconnected operation: Test 1: high impedance fault on main grid Test 2: microgrid overloaded by main grid Test 3: low impedance fault on main grid Test 4: re-connection to mains after fault Test 5: microgrid black start Microgrid Switch UPS Switch UPS Switch OAD 1 DIESE BATTERY 16 kw OAD 2 BATTERY 10 kw 10 kw PV WEC 15 kw OAD 3 2 kw OAD 10 OAD 4 OAD 9 OAD 7 OAD 5 diesel Switch DIESE 10 kw Protected user: health centre OAD 8 Community generators and deferrable loads Fig. 6: Microgrid test configuration OAD 6 Standard user: households Protected user: Small business IV. FYWHEE ISTAATIO AT THE UIVERSITY OF MACHESTER Fig. 5: View of DeMoTec at ISET In order to validate the different functions of a microgrid, a specific test configuration has been set up in the DeMoTec, which includes the following components: 4 grid forming units (2 battery units and 2 diesel gensets) 2 renewable energy generators: PV and Wind several loads with different priority levels several automatic switches for sectionalizing the microgrid into up to 3 low voltage island grids supervisory control for a fully automatic operation of the microgrid ( disconnection, re-connection, black-start, optimal dispatch) Using the above microgrid configuration, five tests have been realized for studying the transition from interconnected to Fig. 7 illustrates the hardware topology used in the University of Manchester Microgrid/ energy storage laboratory prototype. The overall system is nominally rated at a 20kVA, although the flywheel and power electronics are rated much higher (100kW). The immediate work for 2005 has involved the development of control systems for real time control of the Microgrid hardware, using the Simulink/dSPACE control environment. The test-rig has been designed to allow the investigation of power-electronic interfaces for generation, loads or energy storage. The AC/DC inverter labeled flywheel inverter in figure 7 can be configured in software to allow the interfacing of the flywheel storage system to the remaining microgrid unit. Alternatively the interface may be fed from a shunt connection from the power supply (emulating the DC link of a source of generation) or it may feed a DC load bank (emulating a load). The Microgrid may be operated in islanded operation (breakers 1 and 2 open in fig 7). Breaker 1 is opened to emulate a loss of mains condition. Breaker 2 is under the control of the Microgrid controller system. Thus the onset of islanding, and resynchronization to mains, as well as mains connected operation may be tested. The design, build and commissioning of the laboratory Microgrid hardware/flywheel energy storage system is now complete.

5 5 Main Grid Microgrid Coupling reactors and transformers Inverter M-G set G M 12kW load oad(s) Unit 7 8 DC-link 9 Variac and rectifier Synchronous GenSet Dump resistors Breakers aboratory Supply Fig. 9: inverter P and Q during microgrid grid reconnection to mains (P top trace) Figure 9 shows real and reactive power flows from the flywheel energy storage unit before and after reconnection to the mains unit. Despite some noise on the trace, it can be seen that about 400W of real power and 2kVAr of reactive power are needed before mains connection to balance the mismatch between generation and load in the islanded microgrid and to move the voltage and frequency of the microgrid to match mains. After reconnection at about 6.5s, the load mismatch is supplied from mains and the flywheel net power injected falls to zero. MICROGRID V. REFERECES Measured Power Microprocessor Signal Electronics & ow Based Control Information evel Protection Centre Rectifier for Highway Hardware Energy Storage Fig. 7: MicroGrid ayout Schematic at Manchester Websites: [1] a database of relevant DG testing facilities [2] Virtual Visit in DeMoTec (Microgrid aboratory facility of ISET) [3] Microgrids project homepage [4] Foundation of Intelligent Physical Agents (FIPA) [Online]. Available: [5] Java Agent DEvelopment Framework [Online]. Available: Technical Reports: [6] O. Osika, T. Degner, C. Hardt, H. ange, M. Vandenbergh, A. Dimeas, D. Georgakis, A. Kamarinopoulos, S. Papathanassiou, A. J. Goodwin, K. Elmasides, G. Iliadis, M. Barnes, C. Fitzer. G. Kariniotakis, Description of the laboratory micro-grids, Microgrids project report DH1 Fig. 8: View of flywheel installation Papers from Conference Proceedings (Published): [7] P. Strauß and T. Degner European aboratories for Distributed Energy Resources - Technology and International Co-operation, in Proc. World Renewable Energy Congress VIII, 29. Aug Sept. 2004, Denver, Colorado, USA [8] A. Engler, D. Georgakis. Hatziargyriou C. Hardt S. Papathanassiou Operation of a prototype Microgrid system based on micro-sources equipped with fast-acting power electronics interfaces, in Proc. 35th Annual IEEE Power Electronics Specialists Conference, , Aachen [9] Dimeas A.,Hatziargyriou.: A MultiAgent System for MicroGrid Operation, IEEE PES General Meeting 2004 Denver Colorado

6 6 VI. BIOGRAPHIES Dr. Mike Barnes, received his BEng and PhD degrees from the University of Warwick in 1993 and 1998 respectively. From 1997 he has worked as a lecturer, and subsequently senior lecturer in power electronics at the Univeristy of Manchester/UMIST. He has more than a decade of experience in the field of power electronic interfaces for machines, renewable generation and power quality improvement. Aris. Dimeas: was born in Athens, Greece in He received the diploma in Electrical and Computer Engineering from TUA. He is currently a Ph.D. student at Electrical and Computers Engineering Department of TUA. His research interests include dispersed generation, artificial intelligence techniques in power systems and computer applications in liberalized energy markets. He is member of the Technical Chamber of Greece and student member of IEEE. Dr.-Ing. Alfred Engler is head of the group ``Electricity Grids" of ISET's division ``Engineering and Power Electronics". He has been with ISET e.v., Kassel, Germany, which he joined in He received his Dipl.-Ing. (Master's) in 1995 from the Technical University of Braunschweig with a thesis on control of induction machines. In 2001 Dr. Engler received the degree Dr.-Ing. (Ph.D.) for the development of control algorithms for inverters in modular and expandable island systems. He is mainly involved with inverter control, island grids, micro grids, power quality and grid integration of wind power. He has presented the results of his work in about 40 publications and patents. He regularly lectures in distributed generation and control of power electronics. Stavros Papathanassiou received the Diploma in Electrical Engineering from the ational Technical University of Athens (TUA), Greece, in 1991 and the Ph.D. degree in 1997 from the same University. He worked for the Distribution Division of the Public Power Corporation (PPC) of Greece, where he was engaged in power quality and distributed generation studies, being responsible for the elaboration of DG interconnection guidelines. In 2002, he joined the Electric Power Division of TUA as a lecturer. His research mainly deals with distributed generation technologies and the integration of DG in distribution networks. He is a member of the IEEE, CIGRE and a registered professional engineer and member of the Technical Chamber of Greece. Michel Vandenbergh was born in Brussels, Belgium. He received his degree of electromechanical engineering from the Université ibre de Bruxelles in 1987, and his PhD in energy engineering in 1997 from the Ecole des Mines de Paris, France. He has more than 15 years experience in using renewable energy technologies for the production of electricity in rural areas. In 2000, he joined the Engineering and Power Electronics Division of ISET, where he is responsible of rural electrification activities within ISET s group on hybrid power systems. Chris Fitzer received a BSc in electronics from the University of Central ancashire, UK in 1999 and a PhD from the University of Manchester Institute of Science and Technology (UMIST), UK in Since 2003 he has been a research associate at the University of Manchester/UMIST. atterly this his become part-time as he has started his own engineering design company. His research interests include dynamic voltage restorers, energy storage systems, microgrids, switched mode power supplies and digital control / digital signal processing. ikos D. Hatziargyriou was born in Athens, Greece. He received the Diploma in Electrical and Mechanical Engineering from TUA and MSc and PhD degrees from UMIST, Manchester, UK. He is professor at the Power Division of the Electrical and Computer Engineering Department of TUA. His research interests include dispersed generation, artificial intelligence techniques in power systems, modeling and digital techniques for power system analysis and control. He is a senior IEEE member, member of CIGRE SCC6 and the Technical Chamber of Greece. Catherine Jones received the MEng degree in Electronics and Electrical Engineering from the University of Glasgow, Glasgow, UK in Since 2003 she has been studying for her PhD at the University of Manchester/UMIST. Areas of interest include Microgrids, in particular the local control of microgrids.