摘要
为了提高非线性系统的模糊建模精度,提出了一种基于改进的菌群优化算法(IBFO)和递推最小二乘(RLS)算法的模糊建模混合学习算法。该方法采用T-S模糊系统进行函数逼近,首先用改进的菌群优化算法优化模糊模型的前提参数,然后用递推最小二乘算法优化模糊模型的后件参数,实现对模糊模型全局参数的优化。对非线性系统、煤气炉数据和气动加载系统的建模表明,该方法在逼近精度方面优于其他方法。
A hybrid learning fuzzy modeling approach based on the improved bacterial foraging optimization algorithm（IBFO）and the recursive least square（RLS）algorithm is proposed to improve the accuracy of fuzzy modeling for nonlinear system. A T- S type fuzzy system is used as the function approximator. The IBFO is used to optimize the premise parameters of the fuzzy model,and the RLS is applied to update the consequent parameters. This method realizes the global parameters optimization for fuzzy modeling. Simulation results on a nonlinear system,Box-Jenkins gas data and a pneumatic loading system show the superiority of the proposed approach in terms of approximation accuracy.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2014年第2期252-258,共7页
Journal of Nanjing University of Science and Technology
基金
河北省自然科学基金(F2010001320)
关键词
改进的菌群优化算法
递推最小二乘算法
T-S模糊系统
全局优化
improved bacterial foraging optimization algorithm
recursive least square algorithm
T-S fuzzy system
global optimization
