BP（Back Propagation）网络是1986年由Rumelhart和McCelland为首的科学家小组提出，是一种按误差逆传播算法训练的多层前馈网络，是目前应用最广泛的神经网络模型之一。BP网络能学习和存贮大量的输入-输出模式映射关系，而无需事前揭示描述这种映射关系的数学方程。它的学习规则是使用最速下降法，通过反向传播来不断调整网络的权值和阈值，使网络的误差平方和最小。BP神经网络模型拓扑结构包括输入层（input）、隐层(hide layer)和输出层(output layer)。-BP (Back Propagation) network in 1986 by Rumelhart and McCelland led team of scientists proposed an algorithm by error back propagation trained multilayer feedforward network, is currently the most widely used one neural network model. BP network can learn and store a lot of input- output model mapping, without prior mapping reveals the mathematical descr iption of this equation. Its learning rule is to use the steepest descent method, by back-propagation network to continuously adjust the weights and thresholds, so the network and the minimum sum of squared errors. BP neural network topology, including the input layer (input), hidden layer (hide layer) and output layer (output layer).