Operation Optimization And Comprehensive Evaluation Algorithm For Active Distribution Network

With the rapid development of the social economy and the increasing demand for electricity,traditional centralized large-scale power generation can no longer meet the requirements of clean energy production and efficient energy utilization.The distributed generations have been rapidly developed due to its environmental friendly and flexible operation mode.Most of the traditional distribution network optimization strategies control the output power of distributed generations in a multi-time framework.The active and reactive power coordination optimization of various flexible resources such as energy storage system and reactive power compensation devices is not considered,and the evaluation of the operational status of the optimized distribution network is not studied.To make up for the deficiency of existing research,based on the multi-source optimization and comprehensive evaluation of distribution network,the two-stage optimization mechanism of'feeder-distribution network'is established from the spatial scale in this dissertation,and combining with the characteristics of distribution power system,the comprehensive evaluation system is constructed,and the index weight assignment method is designed for quantitative evaluation of distribution network operation status.The main work includes:Firstly,the random probability characteristics and time series characteristics of wind energy and photovoltaic power generation are analyzed.The content includes that the Weibull distribution model is used to describe the probability distribution characteristics of wind energy resources;the beta distribution model is used to describe the probability distribution characteristics of illumination resources;and the time series analysis algorithm is used to describe the timing characteristics of renewable energy generation.At the same time,combining with scene analysis method,K-Medoids clustering algorithm is used to implement multi-scenario clustering and reduction.Secondly,the dynamic coordination optimization model of active and reactive power for active distribution network is established for optimal scheduling in the feeder.The content includes considering the timing characteristics of the distribution network,the mathematical model of the energy storage device,distributed generation,capacitor bank switching,SVC reactive power compensation device are established and the relaxation technique is used to deal with the relaxation of the tidal equation and solve the original model with an algorithm package such as MOSEK.Then,distribution network reconfiguration optimization model with distributed generation is established for network reconfiguration between feeders.The content includes the operating characteristics of distributed genrations are analyzed,four types power flow models of distributed generations including PQ,PI,PV and PQ(V)are established,and combining the swarm intelligence of the binary particle swarm optimization algorithm with the probabilistic kicking advantage of the simulated annealing algorithm,the improved hybrid particle swarm optimization algorithm is used to solve the original model.Combining with the IEEE-33 node system,the effects of distributed generation access capacity,location and power factor on the distribution network are studied,and various grid-connected scenarios are reconstructed and analyzed.Finally,the index weight assignment and evaluation model is established for the evaluation of the operation status of the distribution network and combining with the GUI tool of MATLAB,the program interface is created.Based on the operation characteristics of distribution power system,the comprehensive evaluation system is established.Considering the subjectivity and objectivity of index weight,network analysis method and entropy weight method are employed to assign the weights of indicators and evaluate the operational status.