机器学习在水处理系统中的应用  

Application of machine learning in water treatment system

作　　者：皇甫小留 王晶瑞 龙鑫隆 黄瑞星 HUANG FUXiaoliu;WANG Jingrui;LONG Xinlong;HUANG Ruixing(Key Laboratory of Eco-environments in the Three Gorges Reservoir Region,Ministry of Education,College of Environmental and Ecology,Chongqing University,Chongqing 400044,China;StateKey Laboratory of Urban Water Resource and Environment,School of Municipal and Environmental Engineering,Harbin Institute of Technology,Harbin 150090,China)

机构地区：[1]重庆大学环境与生态学院三峡库区环境教育部重点实验室,重庆400044 [2]哈尔滨工业大学环境学院城市水资源与水环境国家重点实验室,哈尔滨150090

出　　处：《给水排水》2022年第11期153-165,共13页Water & Wastewater Engineering

基　　金：国家自然科学基金(52070029、51878092);重庆英才计划“包干制”项目(cstc2021ycjh-bgzxm0154)。

摘　　要：智慧水务是目前水务事业发展的重要方向,机器学习作为实现人工智能的关键技术,在水务智慧化中有巨大的应用前景。从饮用水处理系统、排水处理系统和新技术研发三个方面,对机器学习的应用进行总结与评述。在饮用水处理体系方面,综述了机器学习在水质水量、药剂投加、氯消毒等方面的应用;在污水处理系统方面,总结了处理过程控制、能耗节约、工艺效率提高、膜污染控制、故障诊断等方面的机器学习方法;在新技术研发方面,归纳了机器学习在污染物高效去除的吸附与氧化等技术中的创新研究。最后,系统分析了不同模型的优缺点与使用范围,对智慧水务中机器学习模型的选择和应用有一定的指导意义。Smart water management system is an important direction of the development of water utilities. Machine learning, as a key technology to realize artificial intelligence, has great application prospects in the intelligent water utilities. This paper summarizes and reviews the application of machine learning in drinking water treatment system, drainage treatment system and new technology development. In terms of drinking water treatment system, the application of machine learning in water quality and quantity, drug dosage, chlorine disinfection and other aspects is reviewed. In terms of sewage treatment system, machine learning methods for process control, energy saving, process efficiency improvement, membrane contamination control, fault diagnosis and other aspects are summarized. In terms of research and development of new technologies, the innovative research of machine learning in the adsorption and oxidation technologies for the efficient removal of pollutants is summarized. Finally, the advantages, disadvantages and application scope of different models are systematically analyzed, which has certain guidance for the selection and application of machine learning models in smart water.

关 键 词：机器学习 饮用水处理系统 污水处理系统 吸附 氧化 

分 类 号：TU991[建筑科学—市政工程] TU992