基于GA-BP神经网络的明渠流速测点优化研究  

Optimization of Open Channel Flow Velocity Measuring Point Based on GA-BP Neural Network

作　　者：王宝贺 苏沛兰[1] 吴建华[1] 张玉胜[1] 吴鑫昊 WANG Baohe;SU Peilan;WU Jianhua;ZHANG Yusheng;WU Xinhao(College of Water Science and Engineering,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学水利科学与工程学院,山西太原030024

出　　处：《人民黄河》2023年第12期117-123,共7页Yellow River

基　　金：山西省基础研究计划(自由探索类)面上项目(20210302123169);江西省水利厅科技项目(202022YBKT09);运城市2022年科技计划项目(YCKJ2022059)。

摘　　要：灌区渠道流量计量大都以流速仪法测量为主,渠道断面布设较多数量的流速仪可有效保证测流精度,但也提高了测流难度。以太园泵站某段供水明渠为研究对象,基于多测线、多测点流速仪实测数据,将断面流量设定为输出、不同位置测点流速作为输入,对GA-BP神经网络模型进行训练,并将验证结果与流速面积法计算结果进行对比。结果表明,GA-BP神经网络仅需4个特定位置处测点流速可得到满足精度的计算结果,实现对明渠测点数量的优化,保证测流精度并为相关模型在渠道测流的应用提供新思路。The flow measurement of irrigation channels is mostly based on the current meter method.A large number of flow velocity meters in the channel section can effectively ensure the accuracy of flow measurement,but it also increases the difficulty of flow measurement.Taking a section of water supply open channel of Taiyuan Pumping Station as the research object,based on the measured data of multiline and multi-point current meter,the cross-section flow rate was set as output,and the flow rate of measuring points at different positions was taken as input to train the GA-BP neural network model,and the verification results were compared with the calculation results of flow rate area method.The results show that the GA-BP neural network only needs the flow velocity of the measuring points at four specific locations to obtain the prediction results that meet the accuracy,effectively optimize the number of measuring points in the open channel,and ensure the accuracy of flow measurement.It provides a new idea for the application of related models in channel flow measurement.

关 键 词：流速仪法 GA-BP神经网络 测点数量优化 渠道测流 

分 类 号：TV131[水利工程—水力学及河流动力学]