Smart Charging: AI Solutions For Efficient Battery Power Management In Automotive Applications

Abstract

The blossoming development of connected vehicles and the prolific infusion of artificial intelligence, machine learning, and deep learning computation mechanisms in-car electronics are contributing to a modern integrated, and complex vehicle environment, where an increasing number of vehicle components require an electric power supply to perform their specific role and function. The electric power demand growth is weighted in watts, where even the simple light bulb is no longer only a lamp but an electronic part of the vehicle. This increase in electronic vehicle parts has led to an increased demand for batteries/graphite power cells technological evolution, with a concern towards reduced environmental footprint but also towards innovative manufacturing processes to support the exponential increase in battery numbers. Nonetheless, the thermal/electrical market reality has imposed several limitations on batteries, especially in usage scenarios that demand high repetition and/or are adapted to very specific thermal requirements.The thermal capabilities of batteries are always a key point that drives efficiency, effectiveness, and relevant performance indexes. A battery thermal switch will have a major impact on battery thermal comfort and long-term health when cooling and heating the battery. Smart charging can help to expose and enhance battery performance results, relying on different calculations for battery heating and cooling. The simultaneous ability to control the thermal rearrangement of energy in the battery, improving the power performance during charge and/or discharge, becomes increasingly important. Both the small and large-scale systems with on-board batteries can benefit from this knowledge and management, leading to extra energy savings and contributing to the paradigm of energy and power efficiency. Converging computing architecture with smart battery charging leverages and exposes the preferably coexisting power management and power protection considerations.