一种简易的无模型控制算法——改进的同时扰动随机逼近控制  

A simple Algorithm of Model-Free Control with a Reformative Method of Simultaneous Perturbation Stochastic Approximation

作　　者：程志锋[1] 周有训[1] 

机构地区：[1]华南理工大学自动化科学与工程学院

出　　处：《广东自动化与信息工程》2004年第1期1-3,共3页Guangdong Automation & Information Engineering

摘　　要：同时扰动随机逼近控制算法是一种特别适合于应用在工况变化大、非线性强的工业过程的无模型控制算法。但该算法采用了神经网络作为函数逼近器,使得在工程上的实现过于复杂,文章对此作出改进,使其大大简化,更适合于在工程上应用。The algorithm with simultaneous perturbation stochastic approximation method is the model-free control algorithm with stochastic approximation method and without modeling. It is superior to those industrial processes with non-linearity and great variations. But this algorithm usually takes the Neural Network as function approximator, which makes it too complicated to apply to engineering. This paper develops the algorithm and the reformative algorithm is easy to apply to engineering.

关 键 词：无模型控制算法 同时扰动随机逼近控制算法 SPSA 神经网络 函数逼近器 工业控制 废水处理控制系统 鲁棒性 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]