Multi-objective Dynamic Reactive Power Optimization Of Distribution Network With DG And EVCS

Driven by the development of The Times and various national policies,The permeability of Distributed Generation(DG)and Electric Vehicle Charging Station(EVCS)in the distribution network is increasing year by year,which is conducive to the realization of the national dual-carbon target.Is the practice of green,health,environmental protection and other concepts.However,both DG output and EVCS load demand have randomness and volatility to a certain extent.Their access not only changes the original power flow distribution of the network,but also brings many adverse effects to the system under certain conditions,such as the increase of system active power loss,node voltage fluctuation,and the overload of EVCS A-nodes.Faced with the complex and changeable operating status of the distribution network,the traditional static reactive power optimization method can not meet the requirements of the system for operating stability and reliability.In view of this,in order to maintain the stability and reliability of distribution network operation as the principle,the multi-objective dynamic reactive power optimization research of distribution network including DG and EVCS is carried out.The details are as follows:Firstly,the classification and operation principle of distributed power supply are introduced,and the characteristics of EVCS users and power consumption behavior are analyzed.With the influence of their access on the active power network loss and node voltage of the system as the focus,the different access situations of DG and EVCS are analyzed in detail,and the conclusion is drawn.Secondly,in order to minimize the adverse effects on the distribution network system caused by the randomness and volatility of DG output and EVCS load demand,as well as the unreasonable configuration planning of both,and reduce the adjustment pressure of reactive power regulation equipment,the optimal configuration of DG and EVCS in the distribution network is studied.The process is as follows: 1)Based on scene analysis method,a typical scene is generated taking into account the correlation and uncertainty of wind-light-conventional load;2)Generating EVCS charging load demand scenarios based on Monte Carlo method;3)Establish a two-stage joint optimal configuration model of DG and EVCS,which aims to minimize the cost of equipment investment,operation and maintenance,the cost of the user’s journey to the charging station,and the amount of wind and light abandoned,and solve it to obtain the optimal access position and access capacity of DG and EVCS in the distribution network.In order to better solve such problems as multi-objective dynamic reactive power optimization of distribution network including DG and EVCS with many solving variables and large variable dimension.In this thesis,a Kernel-based hybrid multi-objective optimization algorithm(KHMO)is adopted.The innovation of this algorithm lies in: using the updating rule based on the concept of reproduction kernel,the numerical gradient is approximated better.A new method based on normal vector computation to determine the direction of the search target is used to guide the non-dominated solution to a more favorable region.Finally,two sets of test functions are used to analyze the performance of KHMO algorithm.By matching Multi-objective bacterial foraging optimization(MOBFO),Multi-objective particle swarm optimization(MOPSO),Multi-objective genetic algorithm(NSGA-II),The feasibility and effectiveness of KHMO algorithm are verified.Finally,for distribution networks containing DG and EVCS,a multi-objective dynamic reactive power optimization strategy based on maximum reactive power compensation fluctuation capacity identification is proposed.The definition and identification method of maximum reactive power compensation fluctuation capacity are given.The solution process of strategy is discussed.A multi-objective dynamic reactive power optimization model with minimum network loss and node voltage deviation is established.Finally,KHMO algorithm was used to solve the model,and the advantages and disadvantages of the optimization strategy used in this thesis and that only using reactive compensation capacitors as reactive compensation devices were compared and analyzed,and the feasibility and effectiveness of the multi-objective dynamic reactive power optimization strategy proposed in this thesis based on the maximum reactive compensation fluctuation capacity identification was verified.