城市供水管网漏损时间的预测模型研究

城市供水企业迫切需要加强给水管网的漏损管理，以减少漏损水量和提高经济效益。在对华北某市供水管网漏损数据进行统计和分析的基础上，按照管段实际发生漏损次数分两种情况建立了供水管网漏损时间的预测模型，对漏损次数≤4次的管段采用基于SAS系统的多元线性回归方法，对漏损次数〉4次的管段则采用灰色预测方法。经实例验证，多元线性回归方法预测的平均相对误差为21％，灰色预测方法预测的平均相对误差〈6％，整套模型的精度可满足城市供水管网漏损宏观管理的需要，能够提高管网漏损防治的效率。

Study on Time-prediction Models for Urban Water Supply Network Leakage

Water leakage management is required for urban water supply industry to minimize water loss and yield good economic benefit.In accordance with statistic data of a city's water distribution network leakage in North China region,two different prediction models were established respectively according to the leakage numbers.The prediction model for the pipes with less than four leakages is based on the multi-linear regression in SAS system,and the gray forecasting theories are adopted in the prediction model for the pipes with more than four leakages.The application results indicate that the relative error of multi-linear regression model is 21%;and the relative error of gray forecasting model is less than 6%.The precision of the models can satisfy the requirements of the macro-management on municipal water distribution network leakage,and increase the efficiency of the network leakage prevention.