模糊Petri网的自适应粒子群优化方法研究

The Study of Fuzzy Petri Nets Based on Adaptive Particle Swarm Optimization Algorithm

模糊Petri网对专家知识库的模糊规则具有较强的表达能力,但其缺乏学习能力使得模型参数的确定较为困难,针对此问题提出一种适合于模糊Petri网参数确定的自适应粒子群优化算法。分析了模糊Petri网对专家知识库的模糊规则建模方法;研究了粒子群优化算法及其各种改进算法;在此基础上,提出适合于模糊Petri网参数确定的自适应粒子群优化算法。最后,以无向网络可靠性估计为例,通过实例分析并与基本粒子群优化算法进行比较,证明了所提方法的有效性和优越性。

Fuzzy Petri Nets(FPN)are a powerful modeling tool for expert knowledge database based on fuzzy production rules.But the lack of learning mechanism makes the parameters determination difficult.For this problem,an adaptive particle swarm optimization algorithm which suits the Fuzzy Petri Nets parameters determination is proposed.Firstly,the modeling method uses FPN for expert knowledge database based on fuzzy rule are analysed.Then,Particle swarm optimization algorithm and various improved algorithms are studied.Based on these,an adaptive particle swarm optimization algorithm are proposed which suits for FPN parameters determination.Finally,an application to the reliability estimate for no-direction network is used as an illustrative example and compared with basic particle swarm optimization algorithm.The result shows that this method is efficient and superiority.