净水技术2024，Vol.43Issue(z2)：19-28,10.DOI:10.15890/j.cnki.jsjs.2024.s2.003
基于随机森林耦合因子分析的供水管网健康风险评价方法
Health Risk Assessment Model for Water Supply Network Based on Random Forest Coupling Factor Analysis
摘要
Abstract
With the expansion of municipal water supply network,the risk of urban water supply network is increasing,and it is urgent to evaluate its health risk scientifically.Considering the increasing sample size and dimensions of water supply network data in the context of big data,this paper coupled factor analysis and random forest into the same system framework to construct a health risk assessment model for water supply network,and took the HP District water supply network in Shanghai as an example for calculation and analysis.The results showed that the accuracy and recall rates of the coupled model were 87.50％and 94.12％,indicating that the model could effectively predict the health risk probability of water supply network.According to the results of importance analysis and sensitivity analysis,it was found that pipe age and pipe material were the key factors affecting the health risk of pipe network.The health risk assessment method of urban water supply network proposed in this paper combines the feature simplification advantage of factor analysis with the high accuracy prediction ability of random forest,and provides a certain quantitative basis for water supply enterprises to carry out the management,maintenance,renovation and reconstruction of water supply network.关键词
供水管网/因子分析/随机森林/健康风险评价/分类Key words
water supply network/factor analysis/random forest/health risk assessment/classification分类
土木建筑引用本文复制引用
樊祖辉..基于随机森林耦合因子分析的供水管网健康风险评价方法[J].净水技术,2024,43(z2):19-28,10.基金项目
上海市科技创新行动计划项目(22dz1201800) （22dz1201800）
