BP网络是通过输入层到输出层的计算来完成的。多于一层的隐含层虽然能在速度上提高网络的训练,但是在实际应用中需要较多的训练时间,而训练速度可以用增加隐含层节点个数来实现,因此在应用BP神经网络进行预测时,选取只有一个隐含层的三层BP神经网络就足够了。(The BP network is completed through the calculation of the input layer to the output layer. Although the hidden layer of more than one layer can improve the training of network in speed, more training time is needed in practical application, and the training speed can be realized by increasing the number of hidden layer nodes. Therefore, when the BP neural network is used to predict, the three layer BP neural network with only one hidden layer is sufficient.)