A hybrid model of prairie dog optimization and closed-form continuous-time neural networks for next generation lithium-ion and sodium-ion batteries

Abstract This study aims to enhance the performance and sustainability of Li-ion and Na-ion batteries by developing a hybrid PDO-CCTNN technique focused on optimizing energy density and efficiency for advanced energy storage applications across diverse sectors. The proposed method integrates Prairie Dog Optimization (PDO) to fine-tune operational parameters and Closed-Form Continuous-Time Neural Networks (CCTNN) to model battery dynamics, ensuring effective energy management, prolonged lifespan, and improved charging/discharging performance in MATLAB simulations. Simulation results demonstrate that the PDO-CCTNN method achieves a high energy density of 210 kWh/g and an efficiency of 94.1 %, outperforming conventional methods such as BOA, ICBO, GA, PSO, and BHMCO across multiple evaluation metrics. The PDO-CCTNN technique significantly boosts battery performance and sustainability, making it a promising solution for next-generation Li-ion and Na-ion batteries in energy storage systems, with high accuracy, optimized operation, and better adaptability for real-world energy applications.