Seoul City Implements 'AI-Based Water Quality Management System' for Proactive Response to Water Quality Incidents

Real-time Collection of Water Quality, Complaints, and Supply Pipeline Information for Big Data Analysis

[Asia Economy Reporter Jo In-kyung] The Seoul Metropolitan Waterworks Authority is launching the construction of an "Artificial Intelligence (AI)-based Water Quality Management System" to detect water quality incidents early and predict vulnerable areas in advance in the supply of tap water. Data consulting necessary for AI construction will be provided by the Seoul Digital Foundation.

The Seoul Metropolitan Waterworks Authority and the Seoul Digital Foundation announced on the 25th that starting with a kickoff meeting on the 26th, they will carry out the project "Design of Data System for Predicting Water Quality Incidents Applying AI" until December. To this end, they plan to enhance synergy by establishing an industry-academia-research collaboration system with Seoul National University of Science and Technology Environmental Technology Research Institute, advisory groups in AI, water quality, and waterworks fields, and private institutions.

Since 2005, the city has been monitoring water quality in real time by installing 299 automatic water quality measuring devices at 214 monitoring points in the raw water of the Han River, the production of Arisu, and the distribution and supply process through the "Automatic Water Quality Monitoring System," but there were limitations in predicting water quality incidents in advance.

To overcome this, the Waterworks Authority will first build an "Integrated Water Quality Database" that systematically manages data necessary for water quality monitoring and management, which had been scattered across various fields, by collecting and storing it in real time.

Through intelligent spatial analysis, vulnerable areas prone to water quality incidents throughout Seoul will be detected, and these detected vulnerable water quality areas will be displayed on a comprehensive monitoring screen (map) to carry out special management such as replacing vulnerable pipelines, cleaning the water network, and installing automatic water quality measuring devices. Also, starting with data diagnosis for AI development by the end of this month, they plan to complete the collection of training data, verify machine learning models next year, and complete AI development for water quality management by 2022.

In particular, since the prediction accuracy of AI depends on the quality of data used for machine learning, they plan to promote successful AI development through data consulting by the Seoul Digital Foundation.

The Seoul Digital Foundation will diagnose the current status of data necessary for the introduction of the AI-based water quality incident prediction system and propose a data collection system. They will also develop a model to verify and improve the alarm accuracy of the currently operating automatic water quality monitoring system.

This content was produced with the assistance of AI translation services.