摘要
传统的T-S模型辨识方法,先进行前件模糊子集划分再使用优化算法来确定规则后件参数,经常会陷入局部极值点,无法建立全局意义上的最优模型。针对这一问题,设计了一种基于标准PSO的T-S模型辨识算法,将T-S模型的结构参数和规则后件参数编码为PSO粒子的位置向量,并进行同时优化。仿真数据表明,相较于传统的T-S模型辨识方法,该方法能够提高全局最优的搜索能力,并且所建模型的辨识精度更高。
The traditional methods of T-S model identification often determine fuzzy subsets first and then optimize its consequent parameters.Therefore they are usually trapped into the local optima and fail to find the optimal solutions.A new identification algorithm of T-S model based on PSO is introduced to solve this above problem,structure parameters and consequent parameters are encoded into a particle position vector and optimized together.Simulation results demonstrate that the proposed algorithm can enhance the global search ability and yield a superior performance over the traditional identification methods.
出处
《工业控制计算机》
2020年第4期76-77,98,共3页
Industrial Control Computer
基金
国家重点研发计划项目(2018YFC0808500)。
关键词
T-S模型
微粒群算法
系统辨识
非线性系统
模糊建模
T-S model
particle swarm algorithms
system identification
nonlinear system
fuzzy modeling