Artificial intelligence for leak monitoring in water distribution networks

用于供水管网泄漏监测的人工智能

• 批准号: 560711-2020
• 负责人: Cascante, Giovanni
• 金额: $ 2.19万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=707018
• 关键词: Artificial; intelligence; leak; monitoring; water

Nearly 25-30% of drinking water is estimated to be lost between treatment and delivery. This is not only a loss of revenue and wasted energy from storage to delivery, but also a loss of valuable resource when nearly 1 in 10 people on our planet lack access to clean drinking water. Conventional leak detection technologies rely on inspection methods which involve cumbersome processes and expert interpretation of data to determine the presence of leaks and their location. Our industrial partner has invested in a patented monitoring system which can be installed on fire hydrants to detect and locate leaks, year-round even in cold climates. However, the decision-support tools are basic and have significant room for advancement using artificial intelligence (AI) and latest developments in communication technologies, including 5G. This project aims to bring latest developments in AI to bear on the problem of leak detection and localization in water distribution systems. Most importantly, through this partnership, new algorithms and processes which have been developed and validated by the academic team in controlled laboratory and field environments will be further developed and translated to real-world complex operational environments. It is anticipated that such advancements will increase detection rates and robustness, while also reducing or eliminating the need for expert user intervention. Through this collaboration, trainees will also gain skills in machine learning and signal processing which are in high demand in Canada and worldwide.

据估计，在治疗和分娩之间，约有25-30%的饮用水流失。这不仅是从储存到交付的收入损失和能源浪费，而且是宝贵资源的损失，因为地球上近十分之一的人无法获得清洁饮用水。传统的泄漏检测技术依赖于检查方法，这些方法涉及繁琐的过程和专家对数据的解释，以确定泄漏的存在及其位置。我们的工业合作伙伴投资了一种专利监控系统，该系统可以安装在消防栓上，即使在寒冷的气候中也可以全年检测和定位泄漏。然而，决策支持工具是基本的，并且使用人工智能（AI）和通信技术（包括5G）的最新发展有很大的进步空间。该项目旨在将人工智能的最新发展应用于供水系统的泄漏检测和定位问题。最重要的是，通过这种伙伴关系，学术团队在受控实验室和现场环境中开发和验证的新算法和流程将进一步开发并转化为现实世界的复杂操作环境。预计这些进步将提高检测率和鲁棒性，同时还减少或消除对专家用户干预的需求。通过这种合作，学员还将获得机器学习和信号处理方面的技能，这些技能在加拿大和世界各地都有很高的需求。