一种基于数据挖掘的模糊模型辨识及参数调整新方法  

A new method for identifying fuzzy model and updating its parameters based on data mining

作　　者：张立权[1] 邵诚[1] 

机构地区：[1]大连理工大学信息与控制中心,辽宁大连116024

出　　处：《大连理工大学学报》2006年第5期741-745,共5页Journal of Dalian University of Technology

基　　金：国家"十五"科技攻关计划资助项目(2001BA204B01);教育部骨干教师计划资助项目(69825106)

摘　　要：基于数据挖掘思想,运用模糊集合理论和改进的梯度下降方法,提出一种通用的、同时辨识模糊模型、调整其参数及确定输出变量空间最优划分的新方法.该方法不仅能够修剪冗余和冲突的初始模糊规则,而且通过引入动态误差传递因子,解决了梯度下降法中存在的收敛速度和振荡之间的冲突问题.用经典的倒车控制问题进行了验证,仿真结果表明了本方法的有效性和准确性.On the basis of data mining, a new method is developed for identifying fuzzy model, updating its parameters and determining optimal division of output space simultaneously by means of fuzzy sets theory and the improved gradient descent method. The method can not only prune the redundant and conflicted initial fuzzy rules, but also resolve the conflicts of convergence speed and oscillation existing in gradient descent method by introducing dynamic error transfer factor. The simulation results show the effectiveness and the accuracy of the proposed method by the verification of the classical truck backer-upper control problem.

关 键 词：数据挖掘 模糊模型 描述模型 梯度下降法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]