建立BP神经网络， 12个隐层神经元，4个输出神经元
tranferFcn属性 'logsig' 隐层采用Sigmoid传输函数
tranferFcn属性 'logsig' 输出层采用Sigmoid传输函数
trainFcn属性 'traingdx' 自适应调整学习速率附加动量因子梯度下降反向传播算法训练函数
learn属性 'learngdm' 附加动量因子的梯度下降学习函数(The BP neural network is established, with 12 hidden layer neurons, and 4 output neurons
The tranferFcn property'logsig'hidden layer uses the Sigmoid transfer function
The tranferFcn property'logsig'output layer uses the Sigmoid transport function
TrainFcn property'traingdx'adaptively adjusts learning rate, additional momentum factor, gradient descent, back propagation algorithm, training function
The learn property'learngdm'is a gradient descent learning function with additional momentum factors)