ABSTRACT
Multi-layer ANN model is a nonlinear mathematical model that imitates the structure and function of biological neural network. The multi-layer is composed of input layer, hidden layer and output layer, each layer contains multiple neurons, each neuron represents a specific output function (incentive function), and the information spread through neuron nodes and weights step by step, and connection weights can be adjusted through the learning rule. In the proposed model, the sigmoid incentive function is adopted to establish the relationship between nodes; sigmoid incentive function is as follows:
ε μi i
=
1 1 e+ p( )−
(1)
where μi is ith neuron input value, εi is ith neuron output value.
