微特电机 ›› 2018, Vol. 46 ›› Issue (4): 62-65.doi: 1004-7018-46-4-62-65

• 驱动控制 • 上一篇    下一篇

基于蚁群-粒子群的模糊神经网络超声波电动机控制

乔维德1,张本法2   

  1. 1. 无锡开放大学,无锡 214011
    2. 常州开放大学,常州 213001
  • 收稿日期:2016-01-31 出版日期:2018-04-28 发布日期:2018-04-28
  • 作者简介:乔维德(1967—),男,教授,研究方向为电气自动化、智能控制。
  • 基金资助:
    无锡市社会事业领军人才资助项目(WX530/2015021YB);江苏城市职业学院“工学结合人才培养模式研究”教学改革课题项目(13-YB-03)

Fuzzy Neural Network Speed Control For Ultrasonic Motor Based on Ant Colony Algorithm-Particle Swarm Optimization

QIAO Wei-de1, ZHANG Ben-fa2   

  1. 1. Wuxi Open University,Wuxi 214011,China
    2. Changzhou Open University,Changzhou 213001,China
  • Received:2016-01-31 Online:2018-04-28 Published:2018-04-28

摘要:

超声波电动机运行时具有高度非线性、时变性及强耦合性。为有效破解超声波电动机非线性和建模困难的瓶颈,研究蚁群算法和粒子群算法相结合优化模糊神经网络参数的超声波电动机转速控制方案。仿真分析与实验结果表明,相比传统的BP算法训练模糊神经网络控制方法,该系统能实现对超声波电动机速度的自适应跟踪,速度脉动较小,调节精度高,动态性能较好,抗干扰能力强。

关键词: 超声波电动机, 蚁群算法, 粒子群算法, 模糊神经网络, 速度控制

Abstract:

The ultrasonic motor is highly nonlinear, time varying and strong coupling. To effectively break the bottleneck of nonlinear and modeling of ultrasonic motor, the scheme was proposed to optimize the parameters of fuzzy neural network parameters of the ultrasonic motor speed control with ant colony algorithm and particle swarm optimization algorithm. Simulation analysis and experimental results show that, compared with the traditional BP algorithm training of fuzzy neural network control method, the system have the advantages of small response speed pulse,flexible control, adaptability, high control precision and robustness.

Key words: ultrasonic motor (USM), ant colony algorithm (ACO), particle swarm optimization (PSO), fuzzy neural network (FNN), speed control

中图分类号: