摘要
结合膨胀土的影响指标和神经网络的特性,提出了一种针对膨胀土进行判别和分类的新方法-BP 网络方法。BP 网络是通过将网络输出误差反馈回传,来对网络参数进行修正,从而实现网络的映射能力。结果表明,用神经网络对膨胀土进行判别和分类的准确率很高,它无须在判别因子与预测目标之间建立基于某种理论的经验统计关系。
The article combines the affected index of expansive soil with the character of artificial neural network and presents a new method of artificial neural network (ANN) model. Back-propagation network revises the network factors through the output error back-propagation, and realizes the mapping ability of the network. The results of the research show that such a system is able to achieve a high accuracy in the discrimination and classification of expansive soil without building up an experiential statistical relationship between the predictable factor and the predictable objective. The method has important practical value.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2004年第6期782-784,共3页
Journal of Liaoning Technical University (Natural Science)
关键词
人工神经网络
膨胀土
判别和分类
artificial neural network
expansive soil
discrimination and classification
