摘要： 基于混沌变量，提出一种神经网络自适应控制系统的优化设计方案。采用混沌状态变量优化神经网络辨识器和控制器权参数，实现混沌粗搜索和局部细搜索相结合，搜索出控制系统参数的全局最优值,具有全局性、快速性、并行性。仿真实验表明采用该方案对强非线性对象的控制具有精度高、超调小、响应快、调节时间短等优点。
关键词:
前向神经网络,
混沌优化,
自适应控制,
混沌
Abstract: In this paper, the optimization design for self-adaptive control system of feed-forward neural network is proposed based on chaotic variable. The chaotic variable is applied to optimize neural network identifier and controller by means of combination of global and partial chaos searching,it has many advantages of global minimum, fast convergence speed and in parallel. Simulation results show control results to the strong nonlinear process have high precision, small overshoot, fast response and short setting-time.
中图分类号:
李祥飞,邹恩,张泰山. 基于混沌变量的神经网络自适应控制系统的优化设计[J]. , 2003, 29(08): 15-17.
LI Xiangfei1,2,ZOU En1,2,ZHANG Taishan2. Chaos Optimization Design for Self-adaptive Control System of Feed-forward Neural Network[J]. , 2003, 29(08): 15-17.