Enhancement of Li-Ion Battery Cathode Performance via Bayesian and Genetic Algorithm Optimization: Experimental and Numerical Analysis

The design of electrode parameters is a crucial determinant of the rate and quantity of lithium storage, which directly impacts the energy density and overall performance of lithium-ion batteries (LIBs) in practical applications. Therefore, the optimal design of electrode parameters is essential for enhancing the performance of LIB cells, especially under high-demand operating conditions. In this study, we develop a hybrid optimization framework that combines Bayesian Optimization (BO) with a Genetic Algorithm (GA) to systematically identify optimal design conditions for LIB cathodes. Unlike conventional approaches, we validate our investigation under practical and experimentally aligned conditions to ensure the reliability and applicability of the results. In the study, a pseudo two-dimensional (P2D) model is employed to examine the impact of physical design parameters on the electrochemical performance at varying C-rates. The Bayesian approach is integrated with the P2D model into the Gaussian process to construct a surrogate model, with the aim of optimizing the electrode design parameters to maximize the discharge capacity. Overall, this paper provides valuable insights into the design and optimization of electrode parameters for improving the performance of LIBs, particularly for NCM622 cathodes operating at high C-rates. This approach not only bridges the gap between simulation and practical application but also demonstrates a scalable methodology for future battery design optimization. Moreover, we reveal that BO is an effective technique for designing battery components.