Dynamic stability analysis of 18650 cylindrical lithium-ion batteries on elastic foundation: result verification via machine learning algorithm

Abstract

This research aims to improve the electro-mechanical and vibration characteristics of cylindrical-lithium-ion batteries using advanced nanocomposite materials reinforced with graphene platelets (GPL); these are inserted as the reinforcement of the cathode layer of spirally cross-section batteries that yield better mechanical properties, thermal conductivity as a combined group, and vibrational stability when loaded dynamically. Also, the structure is surrounded by an elastic foundation. Analytical modeling is also investigated with the Rayleigh–Ritz technique to examine the structural and vibration behavior of the cylindrical batteries, accounting for the intricate relationships between material properties and geometry. Furthermore, a deep neural network (DNN) algorithm is modeled to serve as a predictive tool for the electro-mechanical and vibration response of the batteries operating under various service and environmental conditions. The DNN framework shows remarkable accuracy and efficiency with reliable predictions and low computational cost. The results show the promise of GPL nanocomposites to significantly increase the stability and life of lithium-ion batteries when exposed to mechanical shocks and thermal excursions. This analytical and computational framework provides a solid protocol for the design of the next volume of cylindrical lithium-ion batteries to improve performance. The results clearly position advanced material incorporation and artificial intelligence (AI)-guided predictive modeling to help push the integration of efficient and sustainable energy storage technologies. This work provided a basis for battery technology advancement that will promote better energy storage solutions in multiple applications.