基于GA-ELM算法的城市短期用水预测  

Prediction of Short-term Urban Water Consumption Based on GA-ELM Algorithm

作　　者：辛珂 李文竹[1] 刘心[1] XIN Ke;LI Wen-zhu;LIU Xin(Hebei University of Engineering,School of Information and Electrical Engineering,Handan 056038,China)

机构地区：[1]河北工程大学,河北邯郸056038

出　　处：《电脑知识与技术》2020年第13期217-220,共4页Computer Knowledge and Technology

基　　金：国家自然科学基金项目(61440001);教育部新世纪优秀人才支持计划(NCET-13-0770);河北省高等学校高层次人才科学研究项目(GCC2014062)。

摘　　要：准确的短期用水预测是优化供水系统的基础,对城市水资源实时调度和城市供水系统调度有着重要意义。为了克服传统的神经网络预测模型训练时间长、易于陷入局部最优的预测结果,且在少量数据样本情况下预测精确度不足的缺点,本文提出了一种基于遗传算法-极限学习机的城市短期用水预测方法。在引入相关影响因素的基础上,用擅长全局搜索和并行搜索的遗传算法对极限学习机参数进行寻优。结果表明,本模型的预测精度较高,日均绝对百分比误差仅为2.19%,具有较强的实用价值,为未来水资源实时调度提供理论依据。Accurate prediction of short-term water consumption is the basis of optimizing water supply system,which is of great sig⁃nificance to real-time water resources dispatching and urban water supply system dispatching.In order to overcome the shortcom⁃ings of the traditional neural network prediction model,such as long training time,easy to fall into the local optimal prediction re⁃sults,and insufficient prediction accuracy in the case of a small number of data samples.A prediction method for short-term urban water consumption based on the Genetic Algorithm-Extreme learning machine has been proposed.On the basis of introducing rele⁃vant influencing factors,has the advantages of good global search and parallel search genetic algorithm is used to optimize the pa⁃rameters of the Extreme learning machine.It turns out that this prediction model has good prediction effect and high precision,and the average daily absolute percentage error is only 2.19%,The results also show that this model has strong practical value and pro⁃vides theoretical basis for the future real-time water resources scheduling.

关 键 词：遗传算法 极限学习机 短期用水量 预测模型 

分 类 号：TV213[水利工程—水文学及水资源]