The Application Research Of Multi-objective Optimization Algorithm In Power System

With the development of society and the progress of technology,the scale of power system is expanding.People are not only pursuing economic benefits,but also paying more attention to other issues like the energy waste and the safety in power system.The multi-objective optimization of power system is how to achieve better results through artificial manipulation,which is a problem worthy of in-depth study.These objectives are basically conflicting,so how to solve the multi-objective optimization of power system is worth studying and has great significance.However,the optimization of power system is difficult to solve because of its complexity,discreteness and non-differentiability.Researchers have found that intelligent optimization algorithm can overcome some limitations of traditional mathematical methods and solve complex optimization problems.This paper is to study the application of intelligent algorithm to solve multi-objective optimization problems of power system.The specific work is as follows:Firstly,the objective of multi-objective problem of power system is determined.Three objectives of fuel cost,active power loss and voltage quality of power system are used as the objective of optimization in this paper.Mathematical models of each objective are established,and constraints,including equality constraints and inequality constraints,are given.The state variables in power system optimization are constrained by penalty function.The NewtonRaphson method of power flow calculation used in this paper is described and implemented in detail.Then,the principles of the three intelligent algorithms,PSO,GA and DE algorithm,are described in detail,and the parameters of the algorithm are improved to a certain extent.The specific steps and flow charts of the algorithm to solve the single-objective optimization of power system are given.The algorithms are programmed on MATLAB,and the simulation tests of the IEEE-30 bus system are carried out.Each algorithm optimizes the three objectives separately.The results of the experiment can be used as a reference for later multi-objective optimization,and the experimental process also lays a foundation for later multi-objective optimization.Finally,this paper analyses and studies the multi-objective optimization of power system,expounds the principle of the multi-objective forms of three intelligent algorithms used in this paper,and studies the distribution of each algorithm,using adaptive grid method and cyclic congestion degree to ensure the diversity of population and control the distribution of Pareto frontier.The NNSDE algorithm is obtained by improving the multi-objective differential evolution algorithm.According to the three multi-objective optimization algorithms mentioned above,a multi-objective optimization algorithm CO-PGDEA based on cooperative operation among multiple populations is proposed.The above algorithm is tested with standard test function.The experiment shows that NNSDE and CO-PGDEA have better convergence and stability.Then the specific steps and flow charts of each multi-objective optimization algorithm for solving multi-objective problems in power system are given.The algorithms are programmed in MATLAB and simulated on the IEEE-30 nodes.The intelligent multi-objective optimization algorithms are applied to optimize the two-dimensional and three-dimensional target space of power system respectively.The experimental results show that the intelligent multi-objective optimization algorithm is effective in solving multi-objective problems of power system.The results of optimization among algorithms show that the NNSDE and COPGDEA are easier to jump out of local optimum,and the Pareto frontier is broader and the convergence accuracy is higher and the stability is better,which verifies the correctness of the improved algorithm.