一种基于优选BP神经网络的智能模糊优化算法

建立了以模糊逻辑形式表示的基于优选BP神经网络的模糊优化模型,构造了用于优选:BP神经网络权值估计的遗传算子,重点讨论了加权梯度方向的计算方法及沿加权梯度方向进行变异的算子,确定了遗传算法的基本策略,构造了该算法的流程.针对飞边金属消耗设计准则挖掘问题,阐述了该准则建立的依据,讨论了该准则各变量之间的关系,并利用基于模糊优选BP神经网络的模糊优化算法实现了飞边金属消耗设计准则的挖掘.继而应用该实例对算法的稳定性进行了校验,并将该算法与利用梯度下降法对神经网络进行权值估计的算法进行了比较,结果表明,该算法速度快,精度高,稳定性好.

Intelligent Fuzzy Optimization Algorithm Based on BP Neural Network

A fuzzy optimization model presented in the form of fuzzy logic is established,which is based on optimization BP neural network,genetic operators for the evaluation of the weight value in the neural network are constructed,the calculation method of the weighted gradient direction and the mutation operators along the gradient direction are discussed emphatically,the basic strategies of the genetic algorithm are established and the process of this algorithm is constructed.Aiming at the problem of the design c...