小生境遗传算法优化的BP神经网络模型

针对传统BP神经网络模型收敛速度慢、易陷入局部极小点、网络结构不稳定等缺陷,提出一个小生境遗传算法优化的BP神经网络模型.该模型充分利用小生境遗传算法的搜索能力和BP神经网络的非线性映射和学习联想能力,通过小生境遗传算法的选择、交叉、变异及小生境淘汰等操作,优化BP神经网络的初始权值和阈值,并采用BP算法对网络进行训练,有效解决网络初值不合理的问题,提高网络收敛速度、稳定性.实验证明:与传统方法相比,该模型具有很强的可行性和有效性.

Optimized BP neural network model based on niche genetic algorithm

Aimed at the shortcomings of BP neural network model such as slow convergence speed,entrapment in local optimum,unstable network structure etc.,an improved BP neural network model was presented based on niche genetic algorithm.The proposed model made full use of the searching ability of niche genetic algorithm as well as the nonlinear mapping ability and the associated learning ability of BP neural network to optimize the initial weights and thresholds of the BP neural network by means of several operations such as selection crossover,mutation,and niche pass.Then the BP algorithm was adopted to train the network,which could effectively solve the problem of unreasonable BP network initial value and improve the convergence speed and network stability.The experiment result showed that this model was more feasible and effective than traditional methods.