Peak and valley regulation of distribution network with electric vehicles
2018; Institution of Engineering and Technology; Volume: 2019; Issue: 16 Linguagem: Inglês
10.1049/joe.2018.8540
ISSN2051-3305
AutoresFucun Li, Hongxia Guo, Zhen Jing, Zhaojun Wang, Xinku Wang,
Tópico(s)Smart Grid Energy Management
ResumoThe Journal of EngineeringVolume 2019, Issue 16 p. 2488-2492 Session – Poster EFOpen Access Peak and valley regulation of distribution network with electric vehicles Fucun Li, Corresponding Author Fucun Li fengxiao0516@yeah.net State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorHongxia Guo, Hongxia Guo State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorZhen Jing, Zhen Jing State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorZhaojun Wang, Zhaojun Wang State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorXinku Wang, Xinku Wang State Grid Dezhou Power Supply Company, Dezhou, Shandong, People's Republic of ChinaSearch for more papers by this author Fucun Li, Corresponding Author Fucun Li fengxiao0516@yeah.net State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorHongxia Guo, Hongxia Guo State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorZhen Jing, Zhen Jing State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorZhaojun Wang, Zhaojun Wang State Grid Shandong Electric Power Research Institute, Jinan, People's Republic of ChinaSearch for more papers by this authorXinku Wang, Xinku Wang State Grid Dezhou Power Supply Company, Dezhou, Shandong, People's Republic of ChinaSearch for more papers by this author First published: 07 December 2018 https://doi.org/10.1049/joe.2018.8540Citations: 10AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract With the increasing number of electric vehicles (EVs), how to make full use of EVs to a peak shaving and valley filling effect on the electrical load, realise the effective interaction between EVs and power grid has become a hot spot of research. However, EV charging is random and uncertain on the characteristics of time and space, in this study, the authors consider the interaction between the EV and the power grid as the foraging behaviour of birds combined with the construction of EV charging station, and build a charge–discharge model based on a particle swarm optimisation algorithm to achieve optimal results through group cooperation. Finally, the feasibility of the scheme is verified by a numerical example. 1 Introduction The large-scale application of electric vehicles (EVs) is an effective way to deal with the global energy shortage and environmental pollution [1]; however, EV access to the power grid is random and uncertain on the characteristics of time and space, a large number of EVs will affect the safe and stable operation of the power system inevitably. At the same time, the current daily load peak-valley difference is still large [2]; Fig. 1 is a typical daily load diagram. Fig. 1Open in figure viewerPowerPoint Typical daily load diagram It can be seen that the load trough time is between 0:30 and 7:00, peak load occurs at 10:00–22:00 and 20:00–21:30, the daily load rate is only 87%, and these data shows the generator utilisation is low; however, we must take the peak capacity of the power consumption as the standard in the construction of power transmission and distribution. On the other hand, it will cause the peak overlapping peak if we access the EV for charging at the peak of electricity consumption. In order to reduce the peak and valley difference, give full play to the role of EV batteries, vehicle-to-grid (V2G) technology came into being. 2 Peak regulation characteristics of charging station 2.1 Analysis of vehicle travel characteristics As a traffic tool, trip departure time and the end time of the vehicle are in accordance with certain statistical rules. Fig. 2 shows the statistics of departure time and end time of private cars in a certain area. Fig. 2Open in figure viewerPowerPoint Statistics of vehicle travel time As can be seen from Fig. 2, the peak vehicle travel time is 6:00–8:00 in the morning, which has no effect on daily load; the end of the time gathered at 17:00–19:00 in the evening, most users will charge the EV into the grid when they return home. From Fig. 1 we can see that the 20:00–21:30 period is the peak of the day, if the EV access to the grid for charge, which will overlap with the traditional peak period, if the charge quantity is expanded to a certain scale, it will bring great impact to the power transmission equipment, from this aspect, the realisation of V2G interactive participation in the hands of users with great risk, if the charging and discharging plan unified centralised control, and then unified centralised charging in the electricity trough period, in this way we can avoid these risks [3-5]. 2.2 Comparison between V2G and traditional peak shaving In the development of EVs, the supply of electricity has been the bottleneck of further expansion of EVs, even now there has been a dispute over the charging mode and the battery replacement. In the current context of science and technology, the use of charging mode or battery replacement is still not conclusive. EV charging station is a power station which can realise the electric energy supplement of EVs through two ways: vehicle charging and battery replacement. Which way to adopt, user can choose according to their own circumstances. At the same time, the operation enterprise can rent the battery to the user, so that users can reduce the cost of buying a car, and the battery can be concentrated in the charging station by specialised personnel to maintain, which can extend battery life. Realisation of V2G based on charging station does not use EVs as a unit [6-8], but with its battery and charger unit. This realisation of V2G has characteristics of large scale and convenient centralised control, avoiding the mobility of the EV and the uncertainty of access to the power grid. The charging station location is generally determined in large substation around (such as 35 kV substation), this is to avoid the expensive cost of land in the city center, but also can expand the capacity, If we want to cooperate with renewable energy, charging station can be built around the wind farm, which can improve the quality of grid connected wind power by energy storage. One of the main reasons for the research of V2G is to reduce the peak and valley difference of daily load, the commonly used method of peak shaving and valley filling is to build a special pumped storage power station, which is the earliest method to deal with the peak and valley difference of power load, its working principle is: in the electricity trough, we use the extra power to raise the water level; in the peak period, the water is released from the high water level, and the electrical energy generated is used to supplement the energy required during peak power consumption. In addition, the pumped storage power station can also play the functions of spinning reserve, frequency regulation, phase modulation, black start and so on. However, the pumped storage power station is usually located near the large hydropower station close to the water source, far away from the load centre and the dynamic process of starting and closing is slower. The biggest advantage of charging station is fast response compared to pumped storage power station; this feature will play a more effective role in the peak load regulation of the power grid. Whether it is from full load to no-load or from no-load to full, it can be quickly realised through charging station; this feature will play an important role in the peak load regulation of power grid [9], which is very important for peak and valley regulation of distribution network. 3 Peak regulation based on particle swarm optimisation Through the above analysis, the realisation form of V2G based on charging station, avoiding the shortcomings of EVs mobility, random access to power grid, achieved the centralised management of the battery and realised the function of V2G technology. The technology based on the charging station has a great advantage over traditional peak shaving station, but specific to each charge and discharge plan, how to participate in peak load regulation through EVs, there are many ways [10, 11]. In this paper, the particle swarm optimisation algorithm is used to achieve the optimal combination of charging units in charging station. 3.1 Sequential peak shaving method Particle swarm optimisation (PSO) algorithm was first developed by simulating the foraging behaviour of birds, and they achieved the optimal combination through group cooperation. Each particle in the particle swarm has no weight and volume, only contains two parameters: velocity and position. By constantly updating their position and speed, the particle can seek the optimal position, which is the optimal solution of the system. First, PSO initialises a group of random particles [12, 13], which is represented by a D dimension, its location information is: ; velocity: . The particle updates its velocity and position by the formula (1): (1) here: is the velocity and position of the D dimension of the particle in the K iteration; is the inertia weight, which is the parameter of inertia in the particle motion; is the individual optimal position of particle I; is the global optimal position of the whole particle swarm; is a learning factor, usually taking 2; is a random number between 0 and 1. Usually in order to avoid the particle out of the optimal solution, we set the upper velocity limit , at the same time, in order to avoid the lack of search in feasible space, we set the minimum velocity limit . According to the formula (1), the next iteration speed has three parts: the first part is a reflection of the current velocity, the second part reflects the particle's individual learning ability, the third part reflects the particle's social learning ability. The latter two parts reflect the particle update. If the weight of the first part has been a constant (close to 1), it will affect the local search ability; if its weight is small at the beginning stage, is not conducive to jump out of the local minimum point, so we need to add a weight loss function (2) where , respectively, are the maximum and minimum inertia weight; are the number of iterations and the maximum number of iterations, respectively. The following is the objective function: (3) where F is the function of power grid load fluctuation suppression, the objective function for V2G; is the current area load that does not contain battery load, that is, the load before adjustment, and can be predicted by the load curve; The load power P of the predicted period j is equal to the average value of the load power in the first 7 days, is the average daily load, where the average value of 24 h: (4) where is the charge or discharge loads of group i in period j, and n is the number of charging units connected to the local charging station. It is necessary to indicate that it is not accurate to predict the load power P of the period j with the average value of the load power in the first 7 days (In fact, any treatment cannot be accurate, as long as the prediction, there must be an error), but it can meet the requirements of practical application, after all, electricity load does not have much change in a short term. The main constraints of charging stations include the maximum charging and discharging power of the battery, the remaining line transmission and distribution capacity, as well as all the number of batteries whose state-of-charge (SOC) value is greater than the critical value of and the number whose SOC is less than the critical value . , is the battery boundary value as energy, energy/load and load of the three functions and can be set by different circumstances, as shown in Fig. 3. Fig. 3Open in figure viewerPowerPoint SOC determines the function of the battery The constraint conditions of the peak regulation power of the charging station are obtained: (5) here: is the maximum charge power for the charging device of group i in the period j; is the maximum discharge power for the charging device of group i in the period j; is the maximum available transmission capacity of transmission lines in period j; is the hysteresis width for the objective function F. is mainly affected by the size of charge and discharge current, a small current can prolong the service life of the battery, and a large current can speed up the adjustment process, it needs to take a compromise. In order to maintain the battery life, the battery cannot be excessive discharge and cannot be over charged in the peak regulation, the energy involved in the battery pack needs to meet certain limits, in discharge state SOC needs to be greater than the lower limit set , in charge state SOC needs to be less than the upper limit set . It can be concluded that the energy involved in peak regulation: (6) here: : charge and discharge energy; : battery remaining power; In addition, the number of standby batteries (battery SOC between ) in charging station also need to be determined in accordance with local EV operating conditions and load conditions. Fig. 4 shows the flow chart of particle swarm optimisation algorithm: Fig. 4Open in figure viewerPowerPoint Flow chart of particle swarm optimisation algorithm 4 Peak load calculation example Taking a forecast load curve of a region in October 2020 as an example, curve data can be predicted by the past load data, the curves are shown in Table 1. Table 1. Forecast daily load curve data Time/h Load power, MW Time, h Load power, MW Time, h Load power, MW 1 59.1 9 186.4 17 190.9 2 45.6 10 207.9 18 220.9 3 47.6 11 204.7 19 240.3 4 37.3 12 187.9 20 264.6 5 36.1 13 163.3 21 270.5 6 46.5 14 167.3 22 216.1 7 82.3 15 163.5 23 154.4 8 118.1 16 173.3 24 97.5 From the table we can calculate: (7) Now we set up a total of 20 charging units in the charging stations and each unit contains 50 chargers, the maximum charging power of each charger is 15 kW, the maximum discharge power is 10 kW. Hysteresis loop of objective function is set to and the transmission capacity of the transmission line is large enough. The specific parameters of power battery are shown in Table 2. Table 2. Battery parameters for peak regulation Parameter name Parameter values number of batteries in charging station 200 maximum discharge power of each charger 10 kW maximum charge power of each charger 15 kW SOC minimum 0.5 SOC maximum 0.9 capacity of a single battery 5 kW h number of batteries with initial SOC = 0.5 100 number of batteries with initial SOC = 0.9 100 Parameters of particle swarm optimisation algorithm are shown in Table 3 : Table 3. Parameters in PSO algorithm Parameter name Parameter values maximum update speed 1 minimum update speed −1 inertia weight 1 learning factor 2 particle swarm size 20 maximum number of iterations 200 The charging and discharging power scheme of charging station can be obtained by particle swarm optimisation algorithm. The corresponding histogram is shown in Fig. 5 : Fig. 5Open in figure viewerPowerPoint Charging and discharging power histogram We add the load data in Table 2 and the daily charge discharge data in Table 4 to the order load data. The simulation results can be obtained from the original load data, the ordered charging load data and the random charging load data as shown in Fig. 6. Table 4. Charging and discharging power scheme Time/h Load power, MW Time, h Load power, MW Time, h Load power, MW 1 14.2 9 1.2 17 0 2 14.8 10 −2.3 18 −5.8 3 14.9 11 0 19 −5.7 4 14.5 12 0 20 −9.5 5 14.8 13 0 21 −9.7 6 14.5 14 0 22 −9.8 7 13.7 15 0 23 0 8 1.5 16 0 24 10.7 Fig. 6Open in figure viewerPowerPoint Simulation results of peak regulation As can be seen in Fig. 6, the orderly charging and discharging based on particle swarm optimisation can play the role of peak shaving and valley filling for power grid, the random charging will bring the power grid a peak plus peak. Due to the peak valley difference of the original load curve too large and the energy of the charging station is limited, the power load curve cannot be fully leveled; however, the daily load rate has been raised. With the development of EVs and charging stations, the orderly charging and discharging will be able to play a very good peak load regulation in the future. 4 Conclusions In order to give full play to the role of EVs in the peak shaving and valley filling for power grid, in this paper, we build a power grid peak load control model based on particle swarm optimisation algorithm combined with EV charging characteristics, and a simulation analysis is carried out with a specific example. According to the simulation results, it can be seen that the optimal results can be obtained quickly according to the objective function and the particle velocity and learning characteristics. The charging and discharging results are also suitable for the load characteristics. 5 References 1Yang B., Li J., Han Q. et al.: 'Distributed control for charging multiple electric vehicles with overload limitation', IEEE Trans. Parallel Distrib. Syst., 2016, 27, (12), pp. 3441 – 3454 2Abousleiman R., Scholer R.: 'Smart charging: system design and implementation for interaction between plug-in electric vehicles and the power grid', IEEE Trans. Transp. Electrification, 2015, 1, (1), pp. 18 – 25 3Zhang X., Wang G.: 'Optimal dispatch of electric vehicle batteries between battery swapping stations and charging stations'. 2016 IEEE Power and Energy Society General Meeting (PESGM), Boston, USA, 2016, pp. 1 – 5 4Paudel A., Supingklad W., Ongsakul W.: 'Optimal scheduling of active distribution network with electric vehicles'. 2016 Int. Conf. on Cogeneration, Small Power Plants and District Energy (ICUE), Bangkok, Thailand, 2016, pp. 1 – 6 5Panwar L.K., Srikanth Reddy K., Verma A. et al.: 'Optimal schedule of plug in electric vehicles in smart grid with constrained parking lots'. 2016 IEEE 6th Int. Conf. Power Systems (ICPS), New Delhi, India, 2016, pp. 1 – 6 6Shengfang A., Xiangning L., Yunfei W. et al.: 'Study of the active distribution network planning considering multiple electric vehicle Ch Arging stations participating in V2G applications', Proc. CSEE, 2013, 33, (34), pp. 122 – 129 7Ding X., Yonghua S., Zechun H. et al.: 'Research on optimal time of use price for electric vehicle participating V2G', Proc. CSEE, 2013, 33, (31), pp. 15 – 25 8Zhiwei X., Zechun H., Yonghua S. et al.: 'Coordinated charging of plug-in electric vehicles in charging stations', Autom. Electr. Power Syst., 2012, 36, (11), pp. 38 – 43 9Jian W., Kuihua W., Zhizhen L. et al.: 'Impact of electric vehicle charging on distribution network load and coordinated control', Electr. Power Autom. Equip., 2013, 33, (8), pp. 47 – 52 10Yujiao L., Chuanwen J., Xu W. et al.: 'Energy management with stochastic constraint and multi-objective optimization algorithm for electric vehicle battery swap station', Electr. Power Autom. Equip., 2013, 33, (8), pp. 59 – 63 11Shaolun X., Zheng Y., Donghan F. et al.: 'Cooperative charging control strategy of electric vehicles based on multi-agent', Electr. Power Autom. Equip., 2014, 31, (11), pp. 7 – 13 12Tan J., Wang L.: 'Integration of plug-in hybrid electric vehicles into residential distribution grid based on two-layer intelligent optimization', IEEE Trans. Smart Grid, 2014, 5, (4), pp. 1774 – 1784 13Abousleiman R., Rawashdeh O.: 'Electric vehicle modelling and energy-efficient routing using particle swarm optimisation', IET Intell. Transp. Syst., 2016, 10, (2), pp. 65 – 72 Citing Literature Volume2019, Issue16March 2019Pages 2488-2492 FiguresReferencesRelatedInformation
Referência(s)