2022 Vol.5 Dec.No4

• Title: An improved particle swarm optimization algorithm for parameter optimization
• Name: Deqing Yang
• Company: Zhengzhou Business University, School of information and electromechanical engineering Zhengzhou,451200, China
• Abstract:

  In order to effectively solve the shortcomings of the elementary particle swarm optimization algorithm, such as easy to fall into local optimization, slow convergence speed and low precision in the later stage, a particle swarm optimization algorithm integrating various improvement measures is proposed. On the basis of the elementary particle swarm algorithm, five improvement methods are proposed: (1) The inertia weight is set to a dynamic change mode of nonlinear decreasing of the quadratic function. (2) The particle swarm algorithm introduces a shrinkage factor, and the shrinkage factor is set to a linear decreasing mode. (3) Adjusted according to the degree of symmetry of the particle distribution near the optimal solution. (4) In the late stage of particle swarm algorithm search, an expansion model is introduced to increase the movement step of the particle, so that the particle can easily jump out of the local optimal solution, (5) increase the accuracy of the particle search and positioning according to the size of the particle's fitness function value, and change the position of the moderate particle.This paper selects the Sphere function, Shaffer function, Griewank function, and Rastrigrin function as the adaptability function of the particle swarm algorithm, and uses these four classic test functions to test the performance of the particle swarm algorithm. It is concluded that the improved particle swarm algorithm in this paper has the advantages of fast convergence speed, high solution accuracy, and avoidance of local optimal solution.

  
  

• Keyword: test function; optimization Improved; particle swarm algorithm
• DOI: 10.12250/jpciams2022090508
• Citation form: Deqing Yang.An improved particle swarm optimization algorithm for parameter optimization [J]. Computer Informatization and Mechanical System,2022,Vol.5,pp.35-38

[1] Huang Q R. Application of SVM algorithm for particle swarm optimization in apple image segmentation [J]. School of Electronic and Electrical Engineering, 2015.

[2] LIU Rong. Research on Adaptive Particle Swarm Algorithm and Its Application in Multi-objective Optimization[D]. [Master's Thesis]. Guangzhou:South China University of Technology,2011.]

[3] Pillai A C, Chick J, Johanning L. Offshore wind farm layout optimization using particle swarm optimization[J]. Journal of Ocean Engineering and Marine Energy, 2018, 4: 73-88.

[4] LI Jia. Research and application of fuzzy self-tuning PID control in superheated vapor temperature control system[D]. Harbin:Harbin University of Science and Technology,2012.]

[5] YANG Jinglong. Research on evacuation model based on Multi-Agent[D]. [Master's Thesis]. Beijing:Beijing University of Technology,2017.]

[6] Zhao J H, Wang X, Zhang H M. Side scan sonar image segmentation based on neutrosophic set and quantum-behaved particle swarm optimization algorithm[J]. 2016, 37(3): 229-241.

[7] HU Jie. Improvement and Application of Bacterial Foraging Optimization Algorithm[D]. [PhD Thesis]. Wuhan:Wuhan University of Technology,2012.]

[8] GAN Fei. Research on Efficient Identification Algorithm and System of Parameters of Steam Turbine and Its Regulation System[D]. [Master's Thesis]. Chongqing:Chongqing University,2016.]