Comparison of the new hybrid versus water cooling systems for electric vehicle battery: a machine learning approach

Efficient thermal management in battery systems is crucial for maintaining optimal performance, preventing overheating, and prolonging battery life, particularly in electric and hybrid vehicles. The high temperature along cell edges and the significant amount of heat generated by the battery in a short length of time make water cooling, a commonly used and recognized strategy for improving heat dissipation in battery thermal management systems, ineffective. Driven by the need to improve temperature uniformity within battery cells and enhance their safety and efficiency, this study introduces a wavy axial heat conduction channel in place of the conventional straight design and integrates hybrid cooling strategies. Furthermore, the ElasticNet regression model is employed as an effective predictive tool for accurately estimating cell temperature. According to the study’s findings, the hybrid two-block cooling-opt’s maximum temperature ( T max ) and maximum temperature difference (Δ T max ) are 307.78 and 8.74 K, respectively, which are 46.1 and 42.5 K (83.04%) lower than those of the one-block water cooling design. The proposed Elastic Net model outperformed existing models, achieving a correlation higher than 99% across five cells.