Particle Swarm Optimization Algorithm Based on Natural Selection and Simulated Annealing for PID Controller Parameters
| Autor: | Wu Ying, Minlan Jiang, Lan Jiang, Fei Li |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
Computer science Computer Science::Neural and Evolutionary Computation MathematicsofComputing_NUMERICALANALYSIS PID controller Particle swarm optimization 02 engineering and technology Servomechanism law.invention 020901 industrial engineering & automation Local optimum Control theory law Simulated annealing Genetic algorithm 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Algorithm Premature convergence |
| Zdroj: | Simulation Tools and Techniques ISBN: 9783030322151 SimuTools |
| DOI: | 10.1007/978-3-030-32216-8_35 |
| Popis: | The values of a PID controller’s parameters determine the controller’s effect. The particle swarm optimization (PSO) algorithm is often used to optimize the controller’s parameters. However, PSO has some inherent defects, such as premature convergence and easily turning into a local optimization. In this paper, an improved particle swarm optimization algorithm based on a natural selection strategy and a simulated annealing mechanism is proposed to optimize the PID controller’s parameters. In the improved PSO algorithm, the natural selection strategy is used to accelerate the rate of convergence, and the simulated annealing mechanism is employed to ensure the accuracy of the search and increase its ability to avoid local optima. The improved algorithm not only guarantees the convergence speed but also has a better ability to jump out of the local optimum trap. To verify the performance of the improved algorithm, four types of algorithms are selected to optimize the PID controller parameters of the Second-order Time-delayed System and the Permanent Magnet Synchronous Motor (PMSM) Servo System. They are the PSO algorithm, the optimization algorithm proposed in this paper (NAPSO), the seeker optimization algorithm (SOA), and the genetic algorithm (GA). The results show that the improved algorithm has a better optimal solution. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|