Research On Energy Management Optimization Of Micro Energy Grid Based On Game Theory

Micro energy grid(MEG)is a novel energy system that supports the access of a large number of distributed energies,the wide participation of users,and the deep integration of information technology.It can realize multi-energy complementary and multi-network coupling,improving the flexibility and economics of the energy supply system.Bidirectional interaction with the power supply company(PSC)and users is a key characteristic of the MEG.Achieving win-win results for the PSC,MEGs and users with such interactions remains a challenge.In this dissertation,a multi-stakeholder energy trading system with the MEG as the core is constructed as non-cooperative,hierarchical Stackelberg,and coalition cooperative game trading models.Theoretical analysis and numerical verification are performed to indicate the advantages and disadvantages of different game models.Finally,a theoretical system for energy management optimization of MEG based on game theory is established.Main contents of the dissertation are shown below.Firstly,for solving the problem of the MEG modeling and the user modeling in day-ahead energy market,the MEG system model is constructed based on the energy bus structure,and the devices in the MEG are modeled and analyzed for MEG operation constraints.Then,the deviation between the actual energy consumption loads and the energy demands of the user is used to quantify the user's dissatisfaction with the actual energy consumption adjustment.The user's objective function is built by considering both the energy-purchasing expenditure and the dissatisfaction through the fuzzy algorithm.The modelings of the MEG and the user lay the foundation for the subsequent energy management optimization of MEG in different game modes.Secondly,for solving the problem of non-cooperative energy trading between the MEG and the user,the model for the energy management optimization of single MEG under non-cooperative game framework is constructed.The equilibrium of the game is proved to exist,and the game solving method of single MEG and the user is proposed.In the proposed game mode,the MEG publishes energy sales prices,and the user adjusts the energy loads according to the published prices.The simulation results reveal that,in the proposed non-cooperative game mode,the MEG can realize the energy optimization and obtain better profits,and the user can better balance the energy-purchasing expenditure and the experience of energy usage.Thirdly,for solving the problem of non-cooperative energy trading between multiple MEGs and multi-type players,the model for MEGs' hierarchical energy trading under Stackelberg game framework is constructed.The Stackelberg equilibrium of the game is proved to exist,and the distributed method for solving the hierarchical game equilibrium without revealing objective function information is proposed.Considering the diversity of MEG cluster operation,three quintessential trading modes of MEGs,i.e.,independent,connected and centralized management,are investigated to analyze the impact on game results.The simulation results reveal that,the proposed hierarchical Stackelberg game modeling and solving method have good applicability to the energy management optimization of multiple MEGs under different trading modes.Fourthly,for solving the problem of cooperative energy trading between MEGs and different players,the model for multi-player energy trading with the MEG as the core under coalition cooperative game framework is constructed.The superadditivity of different coalitions' objective function is proved,and the methods for solving game equilibrium in different coalition cooperative modes are proposed.The concept of Core is introduced to analyze the stability of coalitions,and the profits of coalitions are reasonably distributed through the Shapley value.Numerical experiments verify the feasibility and effectiveness of the proposed model and method.