where n is the number of nodes in the input layer, l is the number of nodes in the output layer, ˛ is the constant, and1 < ˛ < 10. The number of neurons in the single hidden layer network is 3–15, and the curve of the mean squared error versus the number of neurons in the hidden layer is shown in Fig. 4. We can find that the minimum mean squared error for the neural network appears when m = 10, and thus the number of hidden layer node is set as 10.