Multi-agent Interest Balance Optimization Scheduling Of Inte-grated Energy System Considering Flexible Load

Through interconnecting production,transmission,storage,use and distribution of energy,the Integrated Energy System(IES)is the most effective way to resolve the energy crisis and environmental issue in the world.With the orderly advancement of energy market-oriented transactions,the role of the load in the system has become increasing important,while the characteristics of multi-energy coupling and multi-agent games presented by the IES demonstration project have brought a new challenge to coordinate optimization of the integrated energy system.Based on this,in order to solve the challenges caused by load uncertainty,multi-energy coupling and multi-agent game,the major research content is as follows:Based on the decisive role of the flexible load in Integrated Demand Response(IDR)technology currently,an optimal scheduling model for building-type integrated energy systems designed for users’ behavior is proposed.Firstly,three characteristic models of different flexible loads are introduced,and the elastic coefficient matrix method is introduced to analyze the benefits of flexible loads participating in IDR;Secondly,in order to analyze the impact of human comfort on energy dispatch,the Predicted Mean Vote(PMV)was used to temperature the human comfort,and the equivalent thermal resistance model of the building was established,which expresses the heat exchange process of the building by the form of a circuit to demonstrate the relationship between cooling/heating power and temperature in the building;Then based on the operating and economic characteristics of energy supply equipment and energy storage equipment and combined with users’ behavior,an optimal scheduling model was established,and MATLAB’s Yalmip/Cplex toolbox was used to solve the problem to obtain the optimal scheduling solution.Finally,through the comparison of five scenarios,the role of IDR in improving system economics and energy consumption is analyzed.By comparing whether the comfort in dex is considered,the validity of the proposed PMV and equivalent building thermal resistance model is verified.Aiming at the characteristics of multi-objective optimization and multi-agent games of the integrated energy system,this thesis proposes a multi-agent benefit equilibrium optimal scheduling model considering for the integrated energy system based on the characteristics of users’ behavior.Firstly,based on the equipment attributes and resource endowments of the integrated energy system,the integrated energy system is divided into three main agents: IESP,REO and users,and established their objective functions respectively;Secondly,due to the hyperplane projection(Hyperplane Projection,HP)can balance the convergence and distribution cha racteristics of the solution set,an NSGA-Ⅱ algorithm based on HP is proposed to solve the multi-objective problem;Then,a " One-Leader Two-Follower " games model is constructed to represent the multi-agent interest equilibrium,and the existence and uni queness of the model’s equilibrium solution are analyzed based on the Stackelberg game model.Finally,an example analysis demonstrates the effectiven ess of the proposed improved NSGA-Ⅱ algorithm,analyzes the results of the collaborative optimizati on scheduling of the proposed multi-agent game model,and compares it with the global unified scheduling scheme,which verifies the practicability of proposed multi-agent balanced scheduling.