Multi-Agent Integrated Energy System Optimization Method Based On Game Theory

Energy is an important material basis for the survival and development of mankind,and it has a bearing on the national economy and security.The current energy structure is unreasonable.Energy supply systems such as electricity,heat,and gas are individually designed,planned,and operated without coordination,which is neither scientific nor economical.The traditional energy system is in urgent need of transformation.The integrated energy system(IES)can use advanced information and control technology to coordinate and optimize the generation,transmission,conversion,storage,and transaction of multiple energy sources.IES can realize efficient comprehensive utilization and diversified supply of energy,as well as improve energy efficiency,which has become an inevitable choice for the energy revolution.However,the IES is essentially a kind of complex heterogeneous system with strong coupling,nonlinearity,and multiple dimensions.Many characteristics such as flexible working conditions,cyber-physical integration,heterogeneous multi-energy flow coupling,make the optimization problems such as capacity configuration,operation strategy,and market transactions extremely difficult.Especially with the reform of the energy market and the expansion of the scale of the IES,the multi-agent characteristics are particularly obvious.Solving optimization not only pays attention to overall optimality,but also takes into account individual fairness,which undoubtedly aggravates the complexity of the problem.Therefore,this thesis focuses on the multi-agent characteristics of the IES,using game theory methods to conduct research on the capacity configuration,operation strategy,and transaction mechanism.The main research contents and innovations of this thesis includes:(1)Considering the difference in energy forms and the coupling of energy supply links in the IES,from the part to the whole,the multi-agent relationship of the IES is studied from three aspects:multi-energy complementation inside energy hub(EH),coordinated operation among EHs,and multi-level transactions between EH and users.The interest relationship and the decision-making order are analyzed,and different game models are established to describe optimization problems at various levels.Furthermore,the unit model of the equipment inside EH,the segment model of the EH,and the hierarchical model of the integrated energy market is established,which lays the foundation for the subsequent research on system capacity configuration,operation strategy and market transactions.(2)There are various energy supply modes and diverse interest relationships in EH.A game theory-based multi-agent capacity optimization method for an EH is proposed.The EH is composed of four subsystems:solar photovoltaic,wind turbine,combined heating and power system,and compressed air energy storage,wherein their operators act as players,and the net present value is selected as the utility function.The Nash equilibrium is proven to exist and is solved by the best response algorithm for analyzing self-interested optimization.The Shapley value method is adopted to deal with benefit allocation in cooperative coalition,considering both stability and fairness.Several case studies are conducted to analyze all fifteen possible game models among four players.The results show that the completely cooperative game model yields better economic performance for the whole and for individuals;its total NPV is 20.15%higher than when individuals act alone.(3)Considering the coupling between capacity configuration and operational strategy,a capacity and operation at one-stage joint optimization method is proposed.Compared with the traditional two-stage optimization,this method has higher calculation speed and accuracy.Furthermore,considering the differences in natural resources and load profiles in different geographical locations,this method is extended to IES with interconnection of EHs.This section proposed a capacity and operation joint optimization method for multiple interconnected EHs based on the Nash bargaining game.Neighboring EHs reach cooperation through bargaining over the transaction energy,and realize the transfer of energy in time and space.Finally,this game model is solved by the alternating direction method of multipliers algorithm.Case studies show that the bargaining solution can not only effectively reduce the cost of the IES,but also help achieve a fair benefit balance among EHs,to realize joint planning and coordinated operation of multiple EHs.(4)Aiming at the hierarchical interaction structure between EHs and users,this section proposed a bi-level trading model of IES based on the Stackelberg game with one-leader multi-follower.Taking the integrated energy retailer as the leader,the EH operator and the user as the followers respectively.The Stackelberg equilibrium of the proposed model was proven to be unique,and solved by a distributed solution based on genetic algorithm combined with quadratic programming.The pricing strategy of energy retailer,the operation strategy of EH and the demand response of user can be optimized at the same time.Finally,practical examples show that the proposed method is helpful to solve the balanced strategy of all parties in pursuit of the optimal objective.The price is used to guide the output scheme and adjust the energy consumption plan,the benefit of EH operator and consumer surplus of user are improved simultaneously.(5)Based on the bi-level trading model,this section proposed a distributed coordinative transaction mechanism of an IES utilizing a tri-level game model.The transaction of integrated energy market is divided into horizontal and vertical dimensions.The vertical trading aspect among EHs,load aggregators,and resident users is formulated as a dual-Stackelberg game.Horizontal bidding among the energy hubs is modeled as a non-cooperative game,wherein each stakeholder pursues the optimal benefit.The equilibrium of this tri-level game is proven to exist and is solved by means of a distributed algorithm.The tri-level model is initially transformed into a bi-level problem through the Karush-Kuhn-Tucker condition and convexification,and is then solved by using the distributed iterative algorithm.Finally,the effectiveness of the proposed tri-level game model and distributed solution algorithm was verified by case study,which confirmed that all stakeholders could benefit from the proposed transaction mechanism.In summary,this article has carried out research on the multi-agent capacity configuration,operation strategy and market transactions of the IES.A game framework of multi-energy complementation inside EH,coordinated operation among EHs,and multi-level transactions between EH and users is formed.These researches improve the overall economy of the system while ensuring the fairness of individuals,and provide theoretical and technical support for the development and construction of the IES.