This paper discusses backpropagation neural network model and BP algorithm to determine market response functions.
探讨了利用反向传播神经网络和BP算法确定市场响应函数的方法。
Using backpropagation neural network based on Levenberg-Marquardt algorithm, the universal characteristics of engine were investigated.
采用基于L - M算法的BP神经网络对某发动机万有特性进行研究。
Experiment results show that feed-forward backpropagation network achieves the best performance, which reduces average error rate by54.4%.
实验结果表明前馈后向传播网络的性能最好，与基准模型比较平均错误率下降54.4%。
应用推荐