MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202611100494 A) filed by Madan Mohan Malaviya University Of Technology on August 19, 2026, for Predictive Multi-Signal Edge-Ai System For Early Thermal-Runaway Detection And Autonomous Isolation In Battery Packs.
Inventors include Dr. Ninni Singh; Dr. Rakesh Kumar; Dr. Rajkumar; Dr. Manish Kumar Gupta; Sanjay Kumar; Dr. Anuraj; and Dr. Amit Kumar Dwivedi.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A predictive battery-safety system detects and prevents lithium-ion thermal runaway before conventional alarms trigger. For each cell or module group the system acquires a low-amplitude multi-frequency impedance response, an acoustic-emission signature, and a mechanical strain/swell signal together with temperature. An edge processing unit with a neural accelerator extract features and runs a fused temporal machine-learning model that outputs a runaway-risk score and an estimated time-to- runaway several seconds to minutes in advance. A deterministic safety state-machine maps the risk to a graded response - alert, current de-rating, autonomous electrical isolation of the affected group via a contactor or pyro-fuse, and optional cooling - so that cell-to-cell propagation is halted. Because inference runs locally, protection is immediate and independent of network connectivity, while multi-physics corroboration suppresses false alarms. An IoT gateway logs telemetry and enables privacy- preserving federated learning to keep the model accurate as cells age. The module is compact and retrofittable to existing packs.
Disclaimer: Curated by HT Syndication.
