Application Of Improved Particle Swarm Optimization With Muti-strategy Integration In Reactive Power Optimization Of Power System

My country has entered the development stage of the "14th Five-Year Plan".Society,economy and life will also enter a period of high-quality development.As a pillar industry of the national economy,the electric power industry must not only have extremely high reliability of power supply,but also control the balance of power quality,economic benefits,and energy conservation and efficiency.Therefore,it is of great practical significance to study the problem of reactive power optimization of power system.Because the problem of reactive power optimization of power system has the characteristics of multi-variable,multi-constraint,nonlinear,etc.The limitation of traditional optimization algorithms cannot solve such problems well,which leaves space for people to explore and study to solve this problem.With the progress and development of artificial intelligence algorithms,it provides a good solution to such problems.The particle swarm optimization is a method to solve this scheme.It not only has the advantages of high solution accuracy,strong robustness,and fast calculation speed,but also shows good results in solving different types of problems.Although the advantages are outstanding,the particle swarm optimization also has the disadvantage of falling into the locally optimal solution.Aiming at the above problems,this paper proposes an improved particle swarm optimization with multi-strategy integration to solve the reactive power optimization problem.In the initial stage of particle swarm,the idea of chaotic mapping and immune concentration is introduced,and a strategy of vector distance immune concentration-Tent chaotic mapping is proposed.Above all,the characteristics of Ten chaotic mapping can improve the diversity of particle swarms.Secondly,using the vector distance concentration in the immune algorithm screen out particles with higher quality,which provides a good initial value for iterative calculation and reduces the subsequent workload,thereby improving the operational efficiency.In order to improve the further search ability of particle swarm optimization in the solution space,it is mainly improved from two aspects.On the one hand,the improvement of its own parameters,including inertia weight and acceleration factor.The inertia weight adopts a nonlinear decreasing strategy and the acceleration factor adopts a sine function.Symmetric nonlinear change strategy to balance local search ability and global search ability.On the other hand,the mutation operator is added.The mutation operator combines the ideas of Gaussian distribution and Cauchy distribution,and adopts Gauss-Cauchy combinatorial mutation strategy to dynamically adaptively mutating particles,so as to improve the diversity and antistagnation ability of the population.From the point of view of economy and practicability,the minimum active power loss is selected as the objective function.The penalty factor and the constraints of various variables are added to build a complete mathematical model of reactive power optimization.It also provides solutions to the discrete variable,convergence conditions,fitness function and power flow calculation problems in reactive power optimization problems.MATLAB software is used for programming and simulation.The improved particle swarm optimization with multistrategy integration and particle swarm optimization are used for test simulation of IEEE-14 node and IEEE-30 node systems.The simulation results show that the improved particle swarm optimization with multi-strategy integration make voltage distribution better and reduce active power loss.It has more advantages and proves the effectiveness and feasibility of the improved particle swarm optimization algorithm with multi-strategy integration.