MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118460 A) filed by Sathishkumar Natesan; D. Sakthivel; Lejeune Ramot Mithilesh M; Vadlamani Sudarsanam; Mohamed Nawfal M; Habin H; and Vimalan M on October 03, 2026, for Artificial Intelligence–assisted Thermal Performance Prediction And Emission Optimization In Internal Combustion Engines Using Multi- Sensor Data Fusion.

Inventors include D. Sakthivel; Lejeune Ramot Mithilesh M; Vadlamani Sudarsanam; Mohamed Nawfal M; Habin H; and Vimalan M.

The application for the patent was published on October 09, 2026, under issue no. 41/2026.

Abstract: The present invention relates to an Artificial Intelligence (AI)-assisted method and system for real-time thermal performance prediction and emission optimization in internal combustion (IC) engines using multi-sensor data fusion. The system integrates data from in-cylinder pressure sensors, exhaust gas temperature (EGT) sensors, lambda sensors, NOx sensors, and fuel injection monitoring units to generate a unified thermal-performance map of the engine. The proposed system comprises: (a) a multi-sensor data acquisition module configured to capture synchronized combustion and exhaust parameters with sampling rates up to 10 kHz; (b) a data fusion engine employing adaptive Kalman filtering and feature-level fusion to create a normalized thermodynamic dataset; and (c) an AI-based predictive analytics model using a deep neural network (DNN) with 10 hidden layers trained on over 50,000 engine operating cycles across varying loads and speeds. The system predicts brake thermal efficiency (BTE) with 97.2% accuracy and reduces NOx and CO emissions by up to 18% through optimized combustion parameter control. Real-time computation time is maintained below 1.2 seconds per cycle. An adaptive learning module continuously updates model parameters based on new engine operating conditions, ensuring robustness under transient driving scenarios. The system can be implemented in embedded ECU hardware or cloud-based fleet monitoring platforms, making it suitable for automotive, marine, and stationary power generation applications compliant with Bharat Stage VI (BS-VI) norms.

Disclaimer: Curated by HT Syndication.