A STUDY ON ENERGY MANAGEMENT SYSTEM FOR STABLE OPERATION OF ISOLATED MICROGRID Kwang Woo JOUNG Hee-Jin LEE Seung-Mook BAEK Dongmin KIM KIT South Korea Kongju National University - South Korea DongHee CHOI jinlee@kumoh.ac.kr smbaek@kongju.ac.kr Jung-Wook PARK Hak Ju LEE Yonsei University South Korea Soo Hyoung LEE Jun Bo SIM {kwjoung93, ys0641056, igo87, KERI South Korea KEPCO South Korea jungpark}@yonsei.ac.kr slee82@keri.re.kr {jureeya, jbsim}@kepco.co.kr ABSTRACT This paper proposes an energy management system (EMS) for stable operation of isolated microgrid which is composed of diesel generators, wind turbines, photovoltaic generation systems and battery energy storage systems (BESS). In the EMS, a power dispatch control is proposed in order to minimize diesel operation. There is also a state-of-charge (SOC) control to allow continuous use of the BESS. By dispatching the scheduled power of the diesel generators, the SOC can be regulated within its limits. A simulation model for isolated microgrid was developed using DIgSILENT Power Factory software. To verify the performance of the proposed EMS, case studies are performed using practical data of Dukjuk island in Korea. INTRODUCTION Most island power systems in South Korea largely depend on diesel generators because of their small sizes and low installation costs. However, their operation and maintenance cost are excessively high compared to those of the main grid. As the result, the installation of distributed generators (DGs) based on renewable energy sources (RES) such as wind turbine generators (WTGs), photovoltaic generators (PVs), and battery energy storage systems (BESSs) are increasing. In the grid-connected microgrid, the main grid gives a strong support to the microgrid for stable frequency and voltage behaviors. However, the isolated microgrid does not have external supports for its reliable operation. Therefore, the energy management system (EMS) is used to control active power of each generator to stabilize the microgrid. The EMS maintains the power balance between the generation and the load demand in terms of frequency and voltage stability by considering the unpredictable output of renewable DGs [1]. The BESS plays an important role in the EMS control algorithm by virtue of its charge/discharge characteristics. The integration of BESSs improves system reliability and reduces peak demand, giving the benefits to DG owners and customers. In addition, the BESS can provide enhanced frequency/voltage support, improved power quality and faster transient response compared to an island microgrid without BESS [2]. Accordingly, the dispatchable control of BESS can help reduce the use of diesel generator as well as let the microgrid operate with the higher efficiency and lower operational cost [3, 4]. This paper proposes the energy management strategy suited for isolated microgrid. In the case study, three DGs that consist of WTGs, PVs, BESS, and diesel generators are considered in the EMS. The EMS will give a proper signal to each generator by considering not only the power difference between generation and demand but also the state-of-charge (SOC) of the battery. The BESS operates as a slack bus and forms the system frequency while the diesel generator is used to charges or discharges the BESS enabling the BESS to overcome its capacity limitation. To validate the proposed EMS algorithm, the Dukjuk Island power system in South Korea is modeled. Its practical parameters were obtained from Korea Electric Power Corporation (KEPCO). The proposed EMS algorithm is applied to Dukjuk Island power system by considering environmental factors such as wind speed and irradiance. The autonomous operation of isolated microgrid will be verified with simulation tests by using the DIgSILENT Power Factory software. SYSTEM CONFIGURATION Dukjuk Island is an island with the area of 21 km 2, situated 50 km from the west coast of South Korea. The population is 1,754. Figure 1 shows the load profile in 2015, with a peak load of 1,875 kw. Dukjuk Island was made the target island for a governmental project the green island project, which aims to install eco-friendly, stand-alone power systems in remote islands. Hence, 500 kw photovoltaic generators, 1500 kw wind turbines, and
2500 kw/ 9000 kwh BESS are to be built in Dukjuk Island in order to minimize the operation of diesel generators. It is expected to reduce the fuel consumption and the carbon footprint of Dukjuk Island. ENERGY MANAGEMENT SYSTEM FOR ISOLATED MICROGRID An EMS algorithm suited to an isolated microgrid is needed for stable operation of the system. In the proposed EMS, diesel generators operate with constant power output, while the output of PVs and WTGs supports the diesel generator. The remaining load demand is handled by the BESS. More specifically, the BESS generates the nominal system frequency and acts as a slack bus. An additional algorithm for controlling SOC is needed to keep the SOC within normal operating range. Fig.1. Load profile of Dukjuk Island in 2015. Figure 2 shows the power system of Dukjuk Island. Currently, there are three 300 kw, four 500 kw diesel generators and three main transformers. There also are 170 kw PVs, and 60 kw WTGs in total. PVs smaller than 10 kw are not considered in this model. The system is modelled with the energy sources mentioned above, and added the BESS that are planned to be built. Distribution lines can roughly be divided into two: the north and the west. Also, there are two tie lines which connect the two regions. Location of the loads and lengths of the lines were obtained from KEPCO. The line impedance is considered as a type of AL-ACSR 160-mm 2, 6.9-kV cable. The nominal system voltage and frequency is 6.9 kv and 60 Hz, respectively. Power dispatch control The sum of the PV, WTG, and diesel output primarily handles the load demand. The surplus energy is used to charge the BESS. However, if the power output of DGs (except for the BESS) is less than the load demand, the BESS releases the saved energy to meet power balance. If we neglect the transmission loss of the system, the power balance equation can be expressed as P Diesel + P RES + P BESS = P L (1) where P L is load demand, P RES is the sum of the output power of PV and WTG, and P Diesel is the power output of the diesel generator. P BESS is the power from the BESS. The BESS discharges when P BESS is positive and charges when P BESS is negative. PVs, WTGs and BESS must be used efficiently in order to reduce the operation of diesel generators. Maximum power point tracking (MPPT) algorithm is used for that goal. Each type of the generators has constraints of minimum and maximum generation limits which can be expressed as follows G1 G2 G3 DFIG g g G4 G5 G6 G7 Fig.2. Power system of Dukjuk Island.
max 0 P Diesel P Diesel (2) MPPT 0 P RES P RES (3) P charge,max discharge,max BESS P BESS P BESS (4) max where P Diesel is a maximum power output of diesel generator, P MPPT RES is an output calculated by MPPT, and P charge,max BESS and P discharge,max BESS are a maximum charging and discharging power of the BESS. As a result, the imal dispatched power of the diesel generator, P Diesel will be within the limits such that P Diesel P Diesel P L P MPPT discharge,max RES P BESS P L P MPPT charge,max RES + P BESS (5) (6) SOC control SOC must be limited within normal operating range which is one of the most crucial constraint in BESS. It can be expressed as follows SOC L SOC(t) SOC U (7) where SOC U and SOC L is upper limit and lower limit of SOC, respectively. If we can forecast with confidence the load demand and generation of RES for a given time interval T, the power from the BESS can be calculated by (1) [5, 6]. Furthermore, SOC for the next dispatch can be calculated as follows SOC(t i + T) = SOC(t i ) + 1 t i +T [P E BESS (τ)] dτ BESS t i (8) where t i is a start time of i-th dispatch control, and E BESS is the capacity of the BESS. If the SOC after T goes beyond the limits, power flow of the BESS can be regulated by power dispatch control of the diesel generators. When predicted SOC is lower than SOC L, P Diesel will increase so as to charge the BESS for time interval T. In addition, when predicted SOC is higher than SOC U, the BESS is controlled to discharge for time interval T and P Diesel (t) will be decreased. Figure 3 shows the proposed EMS algorithm. First, the algorithm checks whether MMPT control is being used or not. Then, the EMS compares SOC value with its upper and lower limits for the next dispatching interval. When the SOC is within limits, the BESS operates in order to meet the power balance. When the SOC is beyond limits, the BESS can be controlled, and the lack of power can be compensated by changing P Diesel. Fig. 3. The proposed energy management system algorithm for stable operation of isolated microgrid. CASE STUDY Case study description To validate the proposed EMS algorithm, the proposed algorithm is compared with one that only considers present value of the SOC. The time interval is set to be 15 minutes (i.e., T = 15 min) and the lower and upper limits of SOC are 0.1 and 0.9, respectively (i.e., SOC L = 0.1, SOC L = 0.9). The data of wind speed ranges from 9.5 m/s to 11 m/s, and solar irradiance ranges from 800 W/m 2 to 1000 W/m 2. Simulation Results Without SOC control algorithm, the BESS is controlled to balance peak load calculating only present value of the SOC. Figure 4(a) shows the load demand and power output from diesel generator and RES without the SOC control. Initially, the load demand is bigger than power from diesel and RES. Therefore, the BESS discharges, and the SOC falls from its initial value, as shown in Figure 4(b). Then, SOC reaches 0.1, lower limit of the BESS at t = 33.9 min. As the result, the BESS cannot produce power nor forms the nominal system frequency until the sum of RES and diesel goes higher than load at t = 41 min. Instead, diesel generators support the system frequency which drops below 59.9 Hz as shown in Figure 4(c). However, the proposed EMS can predict that the SOC will drop below its lower limit at t = 33.9 min. Thus, the battery will be charged by diesel generator between t = 30 min and t = 45 min.
range. Furthermore, Figure 5(c) shows that the system frequency with SOC control algorithm has a better response than one without SOC control in Figure 4(c), since the response of a BESS is faster than a diesel generator. As the result, the isolated microgrid can be operated with greater stability with the proposed EMS algorithm. Fig. 4. Simulation results without SOC control (a) Load demand and power of diesel generator and RES. (b) Active power of BESS. (c) Frequency. CONCLUSION This paper concludes that the proposed EMS algorithm can control each DGs with power dispatch control for stable operation of the isolated microgrid. Various constraints of DGs are considered including power balance and SOC limits of the BESS. With the proposed algorithm, the SOC can always be kept within its limits. Also, the system frequency can be made more stable since the response of the BESS is much faster than that of the diesel generator. The proposed EMS was applied to Dukjuk island power system and simulated by DIgSILENT Power Factory. Acknowledgments This work was supported in part by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2016R1E1A1A0292-0095) and in part by the Power Generation & Electricity Delivery Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20141020-402340). Also, it was supported by Korea Electric Power Company (KEPCO) Research Institute (KEPRI). Fig. 5. Simulation results with SOC control (a) Load demand and power of diesel generator and RES. (b) Active power of BESS. (c) Frequency. Figure 5(a) shows that the power from the diesel generators increases up to 1400 kw at t = 30 min and maintains for T =15 min. Then, the BESS can charge the energy with surplus power as shown in Figure 5(b). Thus, the SOC can always stay within the normal operational REFERENCES [1] C. Wang and M. H. Nehrir, 2008, "Power management of a stand-alone wind/photovoltaic/fuel cell energy system," IEEE transactions on energy conversion, vol. 23, pp. 957-967. [2] W.-H. Hwang, S.-K. Kim, J.-H. Lee, W.-K. Chae, J.- H. Lee, H.-J. Lee, et al., 2014, " Autonomous Microgrid Design for Supplying Electricity in Carbon-Free Island ", Journal of Electrical Engineering & Technology. vol. 9, 1112-1118. [3] Y. S. Kim, E. S. Kim, and S. I. Moon, 2016, "Frequency and Voltage Control Strategy of Standalone Microgrids With High Penetration of Intermittent Renewable Generation Systems," IEEE Transactions on Power Systems, vol. 31, pp. 718-728. [4] K. Kusakana, 2015, "Optimal scheduled power flow for distributed photovoltaic/wind/diesel generators with battery storage system," IET Renewable Power Generation, vol. 9, pp. 916-924. [5] N. Cong-Long, L. Hong-Hee, and C. Tae-Won, 2015,
"Cost-Optimized Battery Capacity and Short-Term Power Dispatch Control for Wind Farm", IEEE Transactions on Industry Applications. vol. 51, 595-606. [6] M. A. Abdullah, K. M. Muttaqi, D. Sutanto, and A. P. Agalgaonkar, 2015, "An effective power dispatch control strategy to improve generation schedulability and supply reliability of a wind farm using a battery energy storage system," IEEE Transactions on Sustainable Energy, vol. 6, pp. 1093-1102.