Design and Performance of Battery Temperature Management System with Machine Learning

This research paper illuminates the transition from Internal Combustion Engines (IC Engines) to Electric Vehicles (EVs), driven by environmental awareness and technological advancements. It spotlights the specific challenge faced by electric two-wheelers-battery temperature management systems (BTMS). The discussion underscores the vulnerability of batteries to temperature fluctuations and the imperative role of effective thermal management. Noteworthy are the engineering challenges in designing compact BTMS for two-wheelers, balancing temperature control within space and weight constraints. The proposed solution introduces a Battery Temperature Management Controller utilizing a microcontroller linked to a temperature sensor. When the temperature surpasses a threshold, the controller activates a cooling system while halting battery charging to prevent overheating. Conversely, in colder temperatures, an external voltage powers a PTC Heater, warming the battery coolant. The integration of Machine Learning algorithm, Random Forest for temperature prediction based on load C rates adds a layer of sophistication. This innovative approach addresses critical safety concerns, preventing battery explosions, and enhances battery life in cold climates by mitigating internal resistance. Tailored for electric two-wheelers, especially in colder environments, the proposed BTMS is a practical solution. The paper details the model, simulation, and performance evaluation of this system, offering a comprehensive exploration of its potential benefits and applications. The battery without BTMS had a temperature of 376K but after the instalment of the BTMS it is reduced to 305K.