MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621096816 A) filed by Dr. D. Y. Patil Institute Of Technology, Pimpri, Pune- on August 11, 2026, for Ai-Based Smart Thermal Management System For Electric Vehicle Battery Cooling Using Adaptive Heat Sink Geometry.
Inventors include Dr. Rakesh Raushan; Prof. Jeetendra Dhamone; Atharv Patil; Prasad Shinde; and Radhey Shinde.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: An artificial intelligence (AI)-based smart thermal management system for electric vehicle battery packs is disclosed. The system establishes an active, cyber-physical framework that dynamically alters its physical heat exchanger boundaries to eliminate spatial three-dimensional thermal gradients across cell modules. The system architecture comprises a high-voltage traction battery pack housing a plurality of multi-cell battery modules equipped with embedded spatial temperature sensors. A high-conductivity fluid-cooled cold plate is thermally coupled to the base of the battery pack and forms part of an active closed-loop fluid circuit driven by an inline variable-speed coolant pump, an external radiator, and an electric fan. Layered directly onto the outer surface of the cold plate is an adaptive heat sink geometry structure comprising a matrix array of variable fins mechanically coupled to an adaptive geometry mechanism and driven by localized electromechanical geometry actuators. An edge-computed AI-based control unit is hardwired into the vehicle data bus to ingest multi-modal telemetry streams, including localized cell temperatures, transient current draw, traction motor speed, state-of-charge, navigation routing data, and ambient meteorological metrics. The internal neural networks of the control unit continuously map these parallel data streams to generate a real-time spatial thermal topography of the battery matrix. Instead of running the active fluid loop at maximum capacity, the control unit computes localized convective boundary layers and transmits coordinate-specific command signals to the geometry actuators to morph the heat sink. Under high current draw or localized thermal surges, the actuators immediately extend the variable fins and compress their layout pitch to maximize the effective surface area, establishing a maximum heat removal configuration that aggressively pulls heat from targeted hot spots. Conversely, during low thermal loads, high-speed highway cruising, or sub-zero cell warming cycles, the actuators collapse the variable fins into a completely flat, minimum drag configuration, minimizing fluidic boundary friction layers and eliminating parasitic power draw from the high-voltage bus. Furthermore, upon detecting early-stage cell failure signatures, the control unit drives the local fins adjacent to the affected zone into a hyper-dense, fully extended configuration, creating an active structural thermal shield that halts the initiation vectors of cascading structural thermal runaway.
Disclaimer: Curated by HT Syndication.
