Research On Reactive Power Optimization Of Power System Based On Improved Particle Swarm Optimization

Reactive power optimization in power systems is very complex.It is a nonlinear programming problem with multiple variables and multiple constraints.Given the coexistence of variables,we need to deal with continuous and discrete variables appropriately.The power flow equation is a non-convex high-order equation group equation.Its constraint condition contains multiple variables and constraints.The ordinary algorithm is not easy to find the solution of the power flow equation.Only selecting the appropriate algorithm can solve the equation better..In the process of reactive power optimization,an optimization algorithm that meets the requirements is also selected.Currently,the most widely used optimization algorithms are traditional algorithms and artificial intelligence algorithms.This paper compares the advantages and disadvantages of the two and selects an algorithm that is suitable for reactive power optimization.Particle Swarm Optimization(PSO)is the method used in this paper.Its characteristics are in good agreement with the characteristics of reactive power optimization.The model with the minimum active power loss is a combination of objective function and penalty function,and the power flow is calculated by Newton-Raphson method.The basic principle and optimization process of particle swarm optimization algorithm are introduced in detail.The improvement schemes are presented for the problems of premature and easy to form local optima.The simple particle method is used to constitute the initial particle group,and the quality of the initial particle is improved.The inertia weight of the grouping variation is selected.Based on this feature,the population can be divided into two groups: the large group strategy is a typical linear decrement,and the group strategy is based on any The nonlinear degression of the tangent function;the use of a linear strategy for the acceleration factor and a combination of the selection operations in the genetic algorithm to improve the algorithm,improve the performance of the algorithm,the improved algorithm can quickly jump out of the local optimal solution and can quickly converge to The global optimal value,at the same time adding the disturbance factor,solves the problem that the selection operation will damage the diversity of the population.The problem of discrete variables is the problem that the improved particle swarm algorithm must deal with when it is used in reactive power optimization.In this paper,the problem of discrete variables is solved by an integer-valued real-time hybrid coding scheme with no error and high precision.In this paper,the calculation process and flow chart of the improved particle swarmoptimization algorithm in reactive power optimization are given.The improved particle swarm algorithm is applied in the IEEE-14 and IEEE-30 node standard test systems,and they are programmed using Matalb language to simulate them.The purpose of the algorithm is to verify the effectiveness and applicability of the algorithm,and to compare the simulation results of the improved particle swarm optimization and standard particle swarm optimization algorithm.The results show that the improved particle swarm optimization algorithm in reactive power optimization has better results and can be effective.Improve the system’s voltage quality while reducing the system’s active network loss.Whether in terms of convergence speed or global convergence,the improved particle swarm algorithm has significant advantages.