MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641119384 A) filed by Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College; Vidhyalakshmi K M. E; Hari Praveen B; Henry P; Kowshik A; and Sunil Kumar S on October 06, 2026, for Iot-Ml Driven Digital Twin For Real Time Hydro Chemical Analytics With Sensor Fault Detection, And Automated Alert Generation.
Inventors include Vidhya Lakshmi K M. E; Han Praveen B; Henry P; Kowshik A; and Sunil Kumar S.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The current state-of-the-art for assessing water quality typically consists in periodic sampling and lab based analysis resulting often in delayed reactions to rapid hydrochemical change. In addition, IoT sensors may produce erroneous readings due to calibration drift, environmental interference, communications errors and sensor degradation. The present invention aims to address such issues by providing an IoT-ML-Driven Digital Twin Framework making use of IoT sensing, edge data acquisition, cloud based processing, Machine Learning (ML) and Digital Twin technologies for real time hydrochemical analytics, sensor fault detection and automated alerts to enable a continuous and reliable water quality monitoring. IoT sensors are used to collect parameters such as pH, turbidity, temperature, dissolved oxygen (DO) and total dissolved solids (TDS). When the .time-series data is collected, it is preprocessed, validated, normalized, and assessed for sensor health before being synchronized with a Digital Twin representing the water environment to be monitored. An ML layer performs anomaly detection, temporal-pattern analysis, sensor fault classification, and multivariate correlation analysis to differentiate between normal, abnormal, and possibly faulty measurements. The hydrochemical analytics module analyzes the parameter relationships and variations in order to determine variations in water conditions; predictive analytics can generate early indications of potential deterioration using historical data. A rule-based and ML-assisted decision and alert engine generates automated notifications for abnormal water conditions and potential sensor faults. It is anticipated that the system will achieve an average analytical accuracy of at least 90 percent. However, the actual level of accuracy, precision, recall, Fl -score and fault detection performance will be experimentally determined. Such a system contributes to achievement of SDG 6 (Clean Water and Sanitation), SDG 3 (Good Health and Well-being) and SDG 9 (Industry, Innovation and Infrastructure) through intelligent, data-driven and continuous monitoring of water resources.
Disclaimer: Curated by HT Syndication.
