基于LM-BP神经网络的黑河龙电渠流量推测研究

The speculating research of flow in the HeiheLongdiancanal based on LM-BP neural network

本文通过分析总结内陆流域黑河水系干流莺落峡龙电渠测验断面水文特点以及流量测验现存问题,提出应用matlab编程软件设计LM-BP神经网络预测模型的方法;采用有导师训练方式,并获取历史实测流量测验资料中点流速和平均流速数据做为样本集,对网络模型的设计、训练和测试做了研究和分析。根据网络仿真和测试结果可以得出,在误差允许范围之内,黑河龙电渠测验断面流量测验采用"五线一点"法的测验方案和LM-BP神经网络模型推测流量的方法,其推测值精度较高,且具有一定的推广应用价值。

In this paper,through the analysis of the hydrological characteristics and current problems of flow.test in the YingluoxiaLongdian canal test section of the Heiheriver of inland river basin,the method of using MATLAB programming software to design the prediction model of LM - BP neural network is put forward.The design,training and test of the network model are studied and analyzed by adopting tutor training and the data of point velocity and average flow velocity in measured flow test as sample sets.According to the network simulation and test results,it can be obtained that the method of using "five line one point"test scheme and LM - BP neural network model to estimate the flow ratehas certain practical value in the test section of the HeiheLongdian canal.The accuracy of its predicted values is high in the allowable range of error.