基于混沌DNA遗传算法的模糊递归神经网络建模

Fuzzy recurrent neural network modeling based on chaos DNA genetic algorithm

中文关键词  混沌理论  DNA  遗传算法  模糊神经网络  pH中和过程

英文关键词  chaos theory  DNA  genetic algorithms  fuzzy neural networks  pH neutralization process

中文摘要

      本文受生物DNA分子遗传机制和混沌优化算法的启发, 提出了一种混沌DNA遗传算法, 用于优化T-S模糊递归神经网络(FRNN). 该方法使用碱基序列表示T-S模糊递归神经网络的前件部分参数, 包括模糊规则数, 隶属度函数中心点和宽度; 设计更为复杂的遗传操作算子来改进遗传算法的寻优性能; 利用混沌优化算法优化种群中的较差个体. 同时使用递推最小二乘法(RLS)来辨识T-S模糊递归神经网络的后件部分参数. 最后, 采用基于混沌DNA遗传算法的T-S模糊递归神经网络对一种典型的pH中和过程进行建模。通过与其他建模方法的比较，仿真实验结果表明了所建模型的有效性.

英文摘要

      Inspired by the biological deoxyribonucleic acid’s(DNA) genetic mechanism and the chaos optimization method, we propose a chaos DNA based genetic algorithm(CDNA--GA) for the optimization of the T-S fuzzy recurrent neural network(FRNN) modeling method. In the CDNA--GA, the parameters of the antecedent part of the FRNN, including the fuzzy rule numbers, center points and widths of membership functions, are represented as the nucleotide base sequences; more complicated genetic operators are designed to improve the performance of genetic algorithm, and the chaos optimization method is applied to optimize the inferior individuals in the population. Furthermore, the corresponding parameters in the consequent part of the FRNN are determined with the recursive least-squares(RLS) algorithm. Finally, the proposed FRNN modeling method is applied to model a pH neutralization process, and the simulation results of the experiments show the feasibility of the established model compared with other reported methods.