MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202521132677 A) filed by Shriram P. Kathar; Shivani S. Bhosle; Bhushan Dinkar Pawar; Anuradha M. Sangwai; Dr. Pradnya L. Awate; and Dr. Prapti D. Deshmukh, Sambhajinagar, Maharashtra, on Dec. 27, 2025, for 'a method for analysis of water contamination in waterbodies through digital twin technology.'
Inventor(s) include Shriram P. Kathar; Shivani S. Bhosle; Bhushan Dinkar Pawar; Anuradha M. Sangwai; Dr. Pradnya L. Awate; and Dr. Prapti D. Deshmukh.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The present embodiment provides a Digital Twin approach which is comprised of 5 10 15 20 hybrid deep learning and residual boosting techniques for prediction of multiple water quality parameters simultaneously. The model uses a deep neural architecture with conv1D and Bidirectional LSTM layers to capture nonlinear and temporal patterns in environmental data, optimally to be robust in the case of outliers. By modifying residual regression, CatBoost-based residual regression is applied post processing to improve prediction accuracy for variable parameters. The framework is evaluated by four significant contaminating measures, of Turbidity, Chemical Oxygen Demand, Conductivity and Biological Oxygen Demand. The system performed well in both low-range and highly heterogeneous contamination models, with RMSE values 0.523 and 4.280 respectively indicating its ability to handle low range and highly heterogeneous contamination parameters. This shows that a hybrid Digital Twin framework captures multi-output relationships and improves predictive performance. This method serves as a scalable and intelligent solution for real-time water contamination assessment in environmental monitoring, public health management and decision-support systems."
Disclaimer: Curated by HT Syndication.
