摘要: The recent big rise in demand for battery electric vehicles (BEVs) has brought attention to the importance of battery management systems (BMS) in ensuring the performance, safety, and longevity of these vehicles. At the same Artif... 展开 The recent big rise in demand for battery electric vehicles (BEVs) has brought attention to the importance of battery management systems (BMS) in ensuring the performance, safety, and longevity of these vehicles. At the same Artificial Intelligence (AI) tech-niques have become a proven way of tackling complex engineering problems that can-not be easily solved by established methods. At Eatron, we develop AI-powered, connected, and safe BMS solutions for the automotive and mobility industry. To unlock significant real-life improvements in BEV performance and battery lifespan, it is neces-sary to bring advanced BMS features such as accurate state-of-charge (SOC)/state-of-health (SOH) estimation and remaining useful life (RUL) prediction into serial pro-duction applications with the power of AI.In this paper, we present several applications of AI in critical BMS functions. Our first example is an edge-to-cloud connected RUL implementation that uses AI techniques to accurately predict SOH trajectories for each cell in a BEV. We present real data to demonstrate the efficacy of this approach. Our second example focuses on creating cell-based anomaly detection using AI to predict events like "lithium plating." The test results demonstrate that the AI function is computationally light and can be easily inte-grated into a production-grade BMS. Our final example showcases how highly accu-rate SOC/SOH functions can be implemented as AI functions in cost competitive BMS applications with the help of specialized AI accelerators. We work with AI accelerator hardware partners to offer SOC/SOH AI functions for electric 2-3 wheelers or energy storage system (ESS) applications.By utilizing AI in production-grade BMS solutions, we can enhance real-life perfor-mance, safety, and longevity of BEVs while reducing costs and improving efficiency. 收起