模糊神经网络的混合学习算法及其软测量建模  

Fuzzy Neural Network Based on Hybrid Learning Algorithm and Its Application to Soft Sensor

作　　者：刘瑞兰[1] 苏宏业[2] 褚健[2] 

机构地区：[1]南京邮电大学自动化学院,南京210003 [2]工业控制技术国家重点实验室浙江大学先进控制研究所,杭州310027

出　　处：《系统仿真学报》2005年第12期2878-2881,共4页Journal of System Simulation

基　　金：国家杰出青年科学基金(60025308);国家"十五"科技攻关项目(2004BA204B08)。

摘　　要：提出了一阶TSK模糊神经网络的混合学习算法,算法由三部分组成:基于模糊聚类的网络初始化;基于梯度下降的规则前件的学习算法;基于部分最小二乘的规则后件的学习算法。该混合算法可以根据训练样本的分布自动确定模糊神经网络的初始值,当输入变量个数多时不会出现模糊规则数爆炸现象,训练速度快,模型精度高。将混合学习算法应用到PTA工业过程中4-CBA含量的软测量建模中,取得了令人满意的效果。A new hybrid learning algorithm was proposed to train the fuzzy neural network based on TSK fuzzy model. Firstly, fuzzy c-means algorithm was applied to initialize the parameters of the fuzzy neural network. Secondly, the parameters of the premise part of the fuzzy rule were learned by the gradient descent algorithm. Finally, the parameters of consequent part were learned by the partial least squares algorithm. The proposed hybrid method could automatically give appropriate initial parameters of the fuzzy neural network and prevent the fuzzy rule number from increasing for high-dimensional systems. The results of simulation and industrial application show that the hybrid learning algorithm has properties of fast convergence and high accuracy. The proposed method has been applied to build a soft-sensor for measuring the 4-CBA concentration in the industrial PTA （Purified Terephthalic Acid） oxidation process and the result demonstrates that the proposed method is suitable to practical application.

关 键 词：混合学习 TSK模糊神经网络 软测量 部分最小二乘 模糊聚类 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]