Multi-objective Optimal Planning Of Soft Open Point And Shunt Capacitors In Distribution Network With Wind Power

Wind energy is a widely distributed and non-emission renewable source.The integration of wind power into the distribution network is of great significance for accelerating energy transformation and implementing sustainable development strategies.However,the uncertainty of wind power output also brings great challenges to the stable operation and economic dispatch of the distribution network.As a commonly used reactive power compensation device,shunt capacitors have unique advantages in dealing with the adverse impacts of wind power integration.But it is difficult to meet the voltage regulation requirements when wind power and load fluctuate frequently by simply switching the shunt capacitors.As a new type of flexible distribution equipment,soft open point(SOP)has certain advantages in the continuous regulation of power flow and promotion of wind power consumption.Therefore,on the basis of considering the uncertainty of wind power output,this thesis studies the coordinated planning of SOP and shunt capacitors.The specific work is as follows:(1)To accurately describe the uncertainty of wind power output,a method for generating wind power output scenarios based on bidirectional generative adversarial network(BIGAN)is proposed.Aiming at the problems of traditional generative adversarial network(GAN)training process instability and low quality of generated data,BIGAN model is constructed.The discriminator,generator and encoder are alternately trained using the multi-layer perceptron,and their parameters are updated using back propagation algorithm.After approximating Nash balance,new wind power output scenarios are generated by the trained generator.Using the wind power output data from the American Renewable Energy Laboratory for simulation,the effectiveness of the proposed method is verified from the four aspects of time correlation,probability distribution characteristics,spatial correlation and probability power flow calculation accuracy.(2)To solve the subjectivity of setting the weight coefficients and the uneven distribution of the Pareto optimal solution set when solving multi-objective optimization problems,a multi-objective optimization method based on the improved elitist non-dominated sorting genetic algorithm(NSGA-Ⅱ)is proposed.In view of the shortcomings of the traditional NSGA-Ⅱ algorithm in terms of convergence and uniformity of solution set,four improvement measures are proposed.Through the simulation on the test function,combined with the optimization algorithm evaluation index,the effectiveness of the improvement measures is verified.And the improved NSGA-Ⅱ algorithm is used as the effective solution algorithm for the coordinated planning of SOP and shunt capacitors.(3)For the coordinated planning of SOP and shunt capacitors,fully considering the uncertainty of wind power output,the typical scenario of wind power output is constructed using the BIGAN scenario generation method,and a multi-objective optimal planning model is established with the minimum system annual comprehensive cost and minimum voltage offset as the optimization objectives,and the improved NSGA-Ⅱ algorithm is used to solve the model.The method is tested on improved IEEE33-bus distribution network to determine the location and capacity of SOP and shunt capacitors,which improves the voltage level of the distribution network and the economic benefits of the system.The improved NSGA-Ⅱ algorithm proposed in this thesis shows good ability to find optimization.