基于混沌搜索的AMPSO-WNN在交流伺服系统中的应用  

Application of AMPSO-WNN in AC Servo System Based on Chaotic Search

作　　者：李俊杰 侯远龙[1] 侯润民[1] 李佳恬 Li Junjie;Hou Yuanlong;Hou Runmin;Li Jiatian(School of Mechanical Engineering,Nanjing University of Science&Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学机械工程学院,南京210094

出　　处：《兵工自动化》2020年第3期61-65,86,共6页Ordnance Industry Automation

摘　　要：为解决大功率交流伺服系统存在非线性和参数时变等不确定性的问题,提出一种混沌搜索的自适应变异粒子群优化小波神经网络的预测模型。建立交流伺服电机数学模型,利用不同变异方法使粒子趋近于不同的搜索区域,引入混沌优化算法改进粒子群,采用基于混沌搜索的AMPSO-WNN算法,以提高全局收敛的概率和速度。仿真结果表明:优化后模型的预测精度高于优化前,且改进后算法具有较强的函数逼近能力,网络性能得到了显著提高,局部极小值问题得到了有效解决。In order to solve the problem of nonlinearity and parameter time-varying uncertainty in high-power AC servo system, a predictive model of adaptive mutation particle swarm optimization wavelet neural network for chaotic search is proposed. The mathematical model of AC servo motor is established. Different mutation methods are used to make the particles close to different search areas. Chaos optimization algorithm is introduced to improve the particle swarm. The AMPSO-WNN algorithm based on chaotic search is used to improve the probability and speed of global convergence. The simulation results show that the prediction accuracy of the optimized model is higher than before, and the improved algorithm has strong function approximation ability, the network performance is improved significantly, and the local minimum value problem is effectively solved.

关 键 词：小波神经网络 自适应变异粒子群算法 交流伺服控制 系统辨识 混沌搜索 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]