Research On The Improvement Of Particle Swarm Optimization Algorithm And Its Application

Particle swarm optimization algorithm(PSO)is a meta heuristic algorithm,which was proposed by Dr.Eberhart and Dr.Kennedy in 1995.Inspired by the research on the behavior of birds foraging,a simplified model based on swarm intelligence is established.Based on the observation of the swarm behavior,the information sharing mechanism among the particles in the group is used to make the movement of the whole swarm evolve,so as to obtain the optimal solution.Particle swarm optimization algorithm is an effective global optimization algorithm,which has simple principle,few parameters,fast convergence speed and high search efficiency.However,the disadvantage of the algorithm is that it is easy be trapped in local optimum,which leads to poor search accuracy of the algorithm.In order to improve the search accuracy of particle swarm optimization algorithm,some optimization strategies are studied in this thesis.The main contents are as follows:1.Firstly,in view of the shortcomings of standard particle swarm optimization(PSO)that is easy be trapped in local optimum,this thesis selects the gravity search algorithm(GSA)with strong global search ability to ensure the diversity of solutions,so as to solve the problem of poor search accuracy caused by the phenomenon of "premature" of PSO.By combining the gravity search algorithm and particle swarm optimization algorithm,this thesis proposes a hybrid optimization algorithm(PSOGSA),which not only retains the high search efficiency of PSO algorithm,but also improves the search accuracy of PSO algorithm.2.Secondly,two linear decreasing weight updating strategies are introduced into PSOGSA algorithm.At the beginning of the algorithm,a large weight coefficient is used to ensure the global search ability of the algorithm.In the later stage of the algorithm,a smaller weight coefficient is used to improve the local development ability of the algorithm.The Matlab simulation results show that the improved hybrid algorithm can achieve higher optimization accuracy on multi-dimensional function optimization problem.3.Thirdly,the weight update strategy of PSOGSA algorithm is further studied,and linear increasing and linear decreasing weight selection strategies are introduced into the optimization process respectively.In addition,an improved PSOGSA algorithm withacceleration factor is proposed.The three improved hybrid algorithms are applied to the power system economic load dispatch problem.The Matlab simulation results show that the three improved hybrid algorithms have achieved high optimization accuracy in the economic load dispatch optimization of power system,and can effectively reduce the generation cost of power system,which has a good guiding significance for power system production planning.