论文部分内容阅读
针对BP神经网络收敛速度慢、易陷入局部极小的缺点,将具有全局搜索能力的遗传算法引入到神经网络的权值优化中。遗传算法优化神经网络模型时,参数选取直接关系到模型优化的效率,在给出一种遗传算法的基础上对相关参数进行了研究分析。并采用Matlab软件编程实现算法,把该算法应用到XOR问题求解中,显示出GA-BP算法的优越性,并通过磨机故障诊断实例验证了算法的有效性。
Aiming at the disadvantage that BP neural network converges slowly and is easy to fall into local minima, the genetic algorithm with global search ability is introduced into the weight optimization of neural network. Genetic algorithm optimization neural network model, the parameter selection is directly related to the efficiency of the model optimization, given a genetic algorithm based on the relevant parameters were studied. And the algorithm is programmed by Matlab software. The algorithm is applied to the XOR problem solving, which shows the superiority of the GA-BP algorithm. The effectiveness of the algorithm is verified by mill fault diagnosis examples.