基于智慧水力学的长距离调水工程调度参数预测方法研究OACSTPCD
Research on Scheduling Parameter Prediction Method of Long-distance Water Transfer Project Based on Intelligent Hydraulics
为提高长距离调水工程调度参数的计算精度,提出了智慧水力学的概念,基于南水北调中线工程的运行监测数据,采用人工智能方法开展了长距离调水工程调度参数预测方法研究.在对数据清洗的基础上,基于长短时记忆神经网络(LSTM)模型建立了白河、十二里河、东赵河节制闸调度参数的预测模型,采用鲸鱼优化算法(WOA)优化模型的超参数并进行调度参数预测.结果表明:3 个节制闸的闸前水位和流量预测值与其实测值的最大平均绝对误差分别为 0.655 cm和 1.326 m3/s,说明智慧水力学理念和人工智能方法在长距离调水工程调度参数预测中的适用性和精准性.研究内容可为实现工程调度参数预测与精准智能调度提供理论基础.
In order to improve the calculation accuracy of the scheduling parameters of long-distance water transfer projects,the concept of intelligent hydraulics is proposed.Based on the operation monitoring data of the middle route of South-to-North Water Transfer project,artificial intelligence method is used to study the scheduling parameter prediction method of long-distance water transfer project.On the basis of data cleaning,a prediction model for the scheduling parameters of the control sluices of Baihe River,Twelve Mile River,and Dongzhao River was established based on the long and short-term memory neural network(LSTM)model.The Whale Optimization Algorithm(WOA)is used to optimize the hyper-parame-ters of the model and to predict the scheduling parameters.The results are as follows.The maximum mean absolute errors of the predicted pre-gate water levels and flows of the three control gates with their actual measured values were 0.655 cm and 1.326 m3/s,respectively.This demonstrates the applicability and accuracy of intelligent hydraulics concepts and artifi-cial intelligence methods in the prediction of scheduling parameters for long-distance water transfer projects.The research content can provide a theoretical basis for realizing the prediction of engineering scheduling parameters and accurate intelli-gent scheduling.
刘宪亮;许新勇;陈晓楠;罗全胜
华北水利水电大学 水利学院,河南 郑州 450046||中国南水北调集团中线有限公司,北京 100038||南阳师范学院 南水北调学院,河南 南阳 473061华北水利水电大学 水利学院,河南 郑州 450046中国南水北调集团中线有限公司,北京 100038黄河水利职业技术学院,河南 开封 475000
水利科学
智慧水力学长距离调水工程调度参数预测人工智能WOA-LSTM模型
intelligent hydraulicslong distance water transfer projectscheduling parameter predictionartificial intelli-genceWOA-LSTM model
《华北水利水电大学学报(自然科学版)》 2024 (003)
18-25 / 8
国家自然科学基金项目(51979109);水利青年科技英才资助项目(2021-12-01).
评论