神经网络训练拟合根据寻优函数的特点构建合适的BP神经网络,用非线性函数的输入输出数据训练BP神经网络,训练后的BP神经网络就可以预测函数输出。遗传算法极值寻优

把训练后的BP神经网络预测结果作为个体适应度值,通过选择、交叉和变异操作寻找函数的

全局最优值及对应输入值。

-Neural network training function fitting based optimization features built right on BP neural network, using non-linear function of the input output data trained BP neural network, the trained BP neural network can predict the function output. The genetic algorithm optimization extreme training BP neural network prediction results as the individual fitness value through selection, crossover and mutation find the global optimal value function and the corresponding input value.