摘要
用BP神经网络方法对山坡平均山坡的解法进行分析,以29个小流域样本的水文数据为基础,通过应用人工神经网络反向传播BP(Back Propagation)算法,引入与山坡平均坡度密切相关的流域影响因子,并且通过调整网络结构中的权因子和阈值,建立了山坡平均坡度与流域影响因子之间的BP网络模型.计算结果表明,用拓扑结构为5-12-1的BP网络,经过学习150000次后,随机测试小流域样本的山坡平均坡度其计算结果和测试结果的相对误差不超过5%;证明该ANN模型的拟合能力强,从而为小流域山坡平均坡度的计算提供了一条新途径.
Based on the hydrological data of 29 small river basins samples, have established the model of the average gradient of slope with many relative factors by application of back propagation （BP） of artificial neural network and introduction of river basins influential factors, and then have got a BP model adapting to the average gradient of slope by adjustment of weighting factors in network structure. By the BP network with the structure 5-12-1 and after 150 000 times study, the test results of the network illustrates the error of the random samples does not exceed 5 %. The results of calculation indicated that the simulating capability of the BP model was powerful and applicable to calculate the average gradient of slope, then provided a new way to calculate the average gradient of slope of small basins.
出处
《数学的实践与认识》
CSCD
北大核心
2009年第9期134-138,共5页
Mathematics in Practice and Theory
基金
铁道部重点科研项目(2003G032)
关键词
人工神经网络
山坡平均坡度
水文
桥梁
artificial neural network
average gradient of slope
hydrology
bridge
