Research On Distributed Multi-objective Energy Flow Optimization Method Under The Background Of Integrated Energy System

With the transition of China's power industry to the third-generation power grid and the gradual establishment of modern integrated energy system,the link between the interconnected regional power grids and the link between the different energy networks are becoming increasingly close,and the ever-expanding scale of the interconnected power grid has placed increasing demands on the power grid optimization calculation,data storage,and information transmission bandwidth.In addition,under the situation of the open electricity market,the distributed method is used to solve the energy flow optimization problem,which is easier to meet the information security requirements of decision-making bodies of the integrated electric and gas system and the muti-area interconnected power system.Moreover,in engineering practice,it is often desirable to coordinate the contradictions between multiple objectives so that several indicators are simultaneously optimal,thus energy flow optimization under multiple objectives needs to be solved.In order to solve the above problems,this paper is devoted to the research of distributed multi-objective energy flow optimization algorithm under the background of integrated energy,and the algorithm is deeply discussed and studied from the following three aspects: how to design the distributed multi-objective optimization algorithm,how to apply the algorithm to solve the distributed multi-objective optimal power flow of interconnected power grid,how to apply the algorithm to solve the distributed multi-objective optimal energy flow of integrated electric and gas system.The specific research contents and results are as follows:(1)The principle of distributed computing method and multi-objective optimization method is analyzed.According to the normal boundary inter-section and the auxiliary problem principle method,a distributed Pareto optimization algorithm based on target value exchange is designed and applied to solve the multi-objective power flow optimization problem of multiarea interconnected power system and obtain the Pareto frontier.Unlike traditional centralized optimization methods,the algorithm do not need to rely on the upper optimization center.In the optimization process,only a small number of boundary variable and target variable information need to be exchanged between the interconnected regions,thus realizing the distribution autonomy of each region and the distributed calculation mode guarantees the privacy of the system parameters in each region.(2)In order to obtain the optimal compromise solution of multi-objective optimization problem directly,three distributed multi-objective fuzzy optimization algorithms based on target value,membership degree and modified variable exchange are designed and applied to solve multi-objective power flow optimization problem of the multi-area interconnected power system.Then,the differences between the three algorithms are compared from the perspective of optimization principle and information security effect.Finally,simulations carried out on model of the IEEE 118 node and IEEE 300 node system verified the feasibility and effectiveness of the three algorithms.(3)A multi-objective optimization model of integrated electric and gas system is established,where three objectives,e.g.,the cost of energy supply,carbon emission and the smoothness of load curve,are taken into account.Then,the distributed multi-objective fuzzy optimization algorithm based on target value exchange is applied to solve the multi-objective energy flow optimization problem of the integrated electric and gas system,in order to meet the needs of the decentralized autonomy of the energy network and realizing distributed scheduling of power network and natural gas network.The work of this paper shows that several distributed multi-objective optimization algorithms can effectively realize the distributed solution of multi-objective optimal energy flow(optimal power flow)and realize decentered calculation of multi-objective optimization.