克隆选择算法在优化模糊Petri网参数中的应用

摘要

如何确定模糊产生式规则的各项参数对模糊Petri网的建立具有重要意义,但一直是尚未解决的难题.首次把克隆选择算法引入到模糊Petri网的参数寻优过程,提出一种基于线程实现技术的参数优化算法,该算法实现不依赖于经验数据,对初始输入无严格要求.仿真实例表明,经克隆选择线程优化算法训练出的参数正确率较高,且所得的模糊Petri网具有较强的泛化能力和自适应功能.%It is significant and being unsolved yet for building a Fuzzy Petri Net(FPN) to determine all parameters of fuzzy production rules. In this paper, Clonal Selection Algorithm(CSA) is originally introduced into the procedure of exploring parameters of FPN.An optimization algoritnm based on the techniques of multithreading is proposed. Realization of the algorithm hasn't depended on experiential data and requirements for primary input are not critical. Simulation experiment shows that the trained parameters gained from above CSA are highly accurate and the resultant FPN model possesses strong generalizing capability and self-adjusting purpose.