钦州市需水量预测模型研究

Study on the Prediction Model of Water Consumption in Qinzhou City

为提高钦州市水资源的需水量预测精度,基于2006—2018年钦州市用水量及相关影响因子,采用灰色预测模型、人工神经网络、定额法三种预测方法,模拟钦州市的需水量,并对3种方法预测得到的结果进行分析。对比可得:BP神经网络模型在预测中误差最小,预测效果好,可以达到预期效果。该模型在钦州市需水量预测中可行性较高,可为该地区需水量预测提供参考。

In order to improve the prediction accuracy of water demand in Qinzhou City,based on the water consumption of Qinzhou City from 2006 to 2018 and the related influencing factors,this paper uses three forecasting methods:grey prediction model,artificial neural network and quota method to simulate the water demand in Qinzhou City,and analyzes the results obtained from the three methods.By comparison,the BP neural network model has the smallest error in prediction,good prediction effect and can achieve the expected effect.The model is feasible in the prediction of water demand in Qinzhou City,which can provide some reference for the prediction of water demand in this area.