Kurita Water Industries Ltd. (Nakano-ku, Tokyo) has developed S.sensing TS, a system for automatically monitoring the processing conditions in sedimentation tanks during sedimentation processes for wastewater treatment. The company began selling the system on August 4.
S.sensing TS uses AI to continuously judge the normality or abnormality of the processing status on behalf of human operators and provides an estimated representation of the cause of abnormalities when they occur. This configuration enables rapid identification of causes and appropriate remedial measures, contributing to the stable operation of wastewater treatment and reduction of CO2 emissions and waste through optimal sludge concentration and management. The system uses Toshiba Digital Solutions 'Meister RemoteX Asset IoT cloud service and Toshiba Analytics' AI "SATLYS".
If sludge is not sufficiently concentrated during the sedimentation process, the dehydrator must operate longer due to a larger mud feed rate, increasing power consumption. Most dewatering machines squeeze sludge. However, if the sludge concentration is low, sufficient pressure cannot be obtained, which increases the water content of the dewatered sludge and the amount of waste.
Visual observation and interface sensors are used for proper control of interfaces and concentrations in sedimentation tanks. However, specialist technicians and skilled workers are needed to identify normal and abnormal treatment conditions, identify causes, and take appropriate measures. This requirement makes securing human resources a challenge.
The developed system, which features AI-based functionality for judging abnormalities, was created to solve these problems. S.sensing TS uses ultrasonic sensors to measure the distance from the water surface to the sludge sedimentation layer and the suspended particles in the sedimentation tank. It also continuously analyzes the conditions in the sedimentation tank using AI.
Operators can check the results of analyses remotely via the Web, and notifications are sent via e-mail when anomalies are detected. The AI automatically and continuously monitors the sedimentation process, which reduces labor related to operation and management. Since the causes of abnormalities are automatically analyzed and presented, they can be easily identified, and appropriate measures can be taken even by non-technical or non-expert staff.
The company has improved the accuracy of machine learning based on large amounts of data accumulated at various sites on phenomena such as settling defects and sludge stirring. It has used this data to optimize operational management by increasing the accuracy of the AI's decisions. By stabilizing the sedimentation process and the concentration of sludge, dehydrator operation times are shortened, and the moisture content of the dehydrated sludge can also be reduced, which contributes to reducing CO2 emissions and waste.
This article has been translated by JST with permission from The Science News Ltd.(https://sci-news.co.jp/). Unauthorized reproduction of the article and photographs is prohibited.