首页|Tongji University Reports Findings in Machine Learning (The insightful water qua lity analysis and predictive model establishment via machine learning in dual-so urce drinking water distribution system)
Tongji University Reports Findings in Machine Learning (The insightful water qua lity analysis and predictive model establishment via machine learning in dual-so urce drinking water distribution system)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting originating from Shanghai, Pe ople's Republic of China, by NewsRx correspondents, research stated, "Dual-sourc e drinking water distribution systems (DWDS) over single-source water supply sys tems are becoming more practical in providing water for megacities. However, the more complex water supply problems are also generated, especially at the hydrau lic junction." Our news editors obtained a quote from the research from Tongji University, "Her ein, we have sampled for a one-year and analyzed the water quality at the hydrau lic junction of a dual-source DWDS. The results show that visible changes in dri nking water quality, including turbidity, pH, UV, DOC, residual chlorine, and tr ihalomethanes (TMHs), are observed at the sample point between 10 and 12 km to o ne drinking water plant. The average concentration of residual chlorine decrease s from 0.74 ± 0.05 mg/L to 0.31 ± 0.11 mg/L during the water supplied from 0 to 10 km and then increases to 0.75 ± 0.05 mg/L at the end of 22 km. Whereas the TH Ms shows an opposite trend, the concentration reaches to a peak level at hydraul ic junction area (10-12 km). According to parallel factor (PARAFAC) and high-per formance sizeexclusion chromatography (HPSEC) analysis, organic matters vary si gnificantly during water distribution, and tryptophan-like substances and amino acids are closely related to the level of THMs. The hydraulic junction area is c onfirmed to be located at 10-12 km based on the water quality variation. Further more, data-driven models are established by machine learning (ML) with test R2 h igher than 0.8 for THMs prediction. And the SHAP analysis explains the model res ults and identifies the positive (water temperature and water supply distance) a nd negative (residual chlorine and pH) key factors influencing the THMs formatio n." According to the news editors, the research concluded: "This study conducts a de ep understanding of water quality at the hydraulic junction areas and establishe s predictive models for THMs formation in dual-sources DWDS."
ShanghaiPeople's Republic of ChinaAs iaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
