Game Theory Based Energy Sharing Strategy Of Prosumer Microgrids

Energy sharing of prosumer microgrids has been an important topic in the field of smart distribution system and demand-side management in smart grids,and has the characteristics of diversified and complex decision-making bodies and structures.The typical decisionmaking optimization approaches in the traditional power system are difficult and unsuitable to be directly utilized to solve the energy sharing problems of prosumer microgrids.Thus the game theory,focusing on the complex multi-agent and multi-objective optimization decision-making,is an effective theory and methodology to solve such issues in smart grids.Based on the game theory,this dissertation studies the game theory based energy sharing strategy of prosumer microgrids.The main contexts are as follows:For the leader-follower energy sharing problem considering prosumers with flexible loads,a Stackelberg game based energy sharing strategy is designed.According to the energy sharing system characteristics,the day-ahead and real-time Stackelberg games are modeled.The existence and uniqueness of the Stackelberg equilibriums are strictly proved,and a distributed probing method with the golden section search is proposed to seek the equilibrium.The simulation studies show that energy sharing can promote the local renewable energy consumption and the utilities.Note that the proposed algorithm can achieve the Stackelberg equilibrium of a certain class of Stackelberg games with limited iterations without exposure of privacy of the participants.For the leader-follower energy sharing problem considering the energy storage system and multiple uncertainties,a robust bi-level optimization model is established.The dayahead energy sharing optimization problem for the prosumers and microgrid operator is modeled as a bi-level problem,which is transformed into a mixed integer programming problem by the single-level reduction and related linearization techniques.An online energy optimization model is designed for each prosumer to update the robust energy schedules for the remaining period according to the real-time state and prediction information,in which the operator's penalty mechanism for prosumers' adjusting their day-ahead energy sharing profiles is embedded.Numerical simulations based on real scenarios shows that the proposed strategy can significantly reduce the energy cost of prosumers.For the aggregation energy sharing problem considering the sharing of uncertain community photovoltaic resources,a stochastic game based energy sharing strategy is modeled.A community aggregated linear dynamic price is designed to drive the energy sharing of prosumers.A risk-averse common photovoltaic sharing method,i.e.,the conditional value at risk,is proposed.A stochastic energy optimization problem for typical residential prosumers is formulated.Accordingly,a stochastic energy sharing game among the community prosumers is established.A sample weighted average approximation method is proposed to estimate the expected cost of prosumers,and then a sample weighted average approximation equilibrium with its existence is analyzed.In addition,the engineering application based on the blockchain technology for community energy sharing is discussed.Note that the proposed sample weighted average approximation method achieves a more effective utilization of community photovoltaic resources.For the peer-to-peer(P2P)energy sharing problem of an energy building prosumer microgrid,a two-stage game based energy sharing strategy is designed.According to the main controllable loads,three typical types of building energy systems are modeled including office buildings,industrial buildings,and commercial buildings.In the first stage,the energy buildings collaborate in energy management to achieve global optimal energy costs with the optimal energy sharing profiles;in the second stage,the buildings carry out the non-cooperative game and obtain the equilibrium based end-to-end energy sharing prices.Distributed optimization algorithms are presented to solve the related optimization and equilibrium-seeking problems,which can reduce the computation and communication burden and can achieve privacy protection.Note that the proposed energy sharing strategy not only ensures the social energy benefits but also realizes the equilibrium-based energy sharing clearing among the participants.Numerical simulations show that the energy sharing reduces the net peak valley demand by 60%.For the P2 P energy sharing of multiple community prosumers,a two-stage game based hierarchical energy sharing strategy is proposed.A day-ahead inter-community energy sharing model is designed allowing all prosumers sharing energy with any prosumer communities.The optimal energy profiles of prosumers for optimal global energy costs are derived,and the equilibrium based energy sharing prices between all pairs of communities are achieved.A real-time inner-community energy sharing model is designed allowing prosumers within the same community to cooperate to deal with the real-time imbalance due to the real-time generation deviations.A distributed algorithm,named as the privacy guaranteed Alternating Direction Method of Multipliers with inner loop,is developed to solve the global optimization problem.A simulation case of a typical community distribution network with three consumer communities is studied.Note that the research mainly realizes an efficient P2 P energy sharing among multiple community prosumer,and address the challenges of decision in the complex P2 P structures.For the energy sharing problem of an energy building prosumer microgrids in P2 P mode,an energy sharing strategy based on the non-cooperative game and benefits distribution mechanism is proposed.A new P2 P energy sharing framework is investigated,and a non-cooperative energy sharing game is discussed.It is found that the general Nash equilibrium is not influenced by the energy sharing payment of the prosumers.The energy sharing profiles are derived by seeking the game equilibrium.The energy sharing payments are determined by the designed cost reduction ratio allocation method considering fairness criteria.The simulation results show that the P2 P energy sharing realizes the full utilization of local renewable energy equipment,and the designed benefits distribution model brings a relatively fairer benefits sharing scheme than the unified electricity price mechanism.For the storage sharing problem for multiple prosumers,an economic energy sharing model based on the asymmetric Nash bargaining model is proposed.A storage/energy sharing system modeling in P2 P mode is established.The optimal storage sharing scheme of the energy storage providers and energy sharing profiles of prosumers are obtained by solving the social energy costs optimization problem.The ways to quantify the contributions of energy storage providers and prosumers in optimizing the social energy costs are designed,and then the respective contributions are taken as the bargaining power of the participants to share the energy cooperation benefits.The simulation studies show that the storage sharing and energy sharing play important roles in improving the energy efficiency and economy of prosumer microgrids.Compared with the traditional Nash bargaining model or the coalition formation game,the proposed model can both ensure the social benefits and the fairness among the participants.