A Game Theoretical Approach for Effective Demand Side Management

Abstract

Due to many advances in rechargeable batteries and environment awareness, electric vehicles (EVs) are making their way in the daily lives. Faster arrival of EVs has brought many challenges in all areas of technology including demand side management (DSM). To meet the peak electricity demand, extra generating units are switched on, which re duce the sustainability of the system and increase the cost of electricity. Therefore, effective DSM techniques with the renewable energy sources (RESs) are employed to efficiently utilize the existing generating capacity. The primary goal of this thesis is to propose cooperative game theory concept Shapley value for the fair distribution of the available resources/payoffs among the participants (Utility, Smart homes (SHs), CES, EVs) of the game. Multiple scenarios are proposed here, community energy storage (CES) is considered in scenario I, to reduce PAR, cost of electricity and to maximize the revenue of CES. Next, in scenario II, common EV parking is considered, which is connected to all SHs. Excess energy is shared with the Parking, then based on stay time, the available charge is distributed among EVs. Next, in Scenario III, coalition of smart homes, having their local energy generation is considered. Due to dynamic behavior of the consumers, the real load is different from the predict load. To tackle these interrup tions in real time, game theory concept Shapley value is employed to minimize the role of MG and to fairly distribute the energy among the smart homes in real time. last, in scenario IV, we have considered group of EVs. The objective is to minimize the impact of charging load at peak hours and to minimize the charging cost of EVs. To attain these objectives EVs are optimally charged and discharged multiple times. Shapley value is used to distribute the total charging cost among all EVs. Simulation results of scenario I, demonstrate the effectiveness of CES, Both the cost of SHs and PAR of the grid are reduced. In Scenario II, the available energy is fairly distributed among the EVs with the minimum wastage of energy. Results, in scenario III, shows the robustness to the dynamic changes in the power consumption. The involvement of the MG is reduced in coordination case. Results of scenario IV shows that the charging cost of all EVs are reduced. The storage capacities of the EVs are effectively used to decrease the burden on the grid at peak hours.