AI-driven energy material design and battery life improvement methods for wearable sports devices

Abstract

This study examines the design and endurance enhancement of energy materials for AI-driven wearable sports devices by systematically comparing the performance characteristics of lithium batteries, supercapacitors, graphene batteries, and sodium-ion batteries. An energy efficiency optimization model based on graph neural networks is proposed. Experimental results show that the model extends battery life by 45%–75% while restricting energy loss to 70 mW·h, which is nearly 50% lower than that of traditional battery management systems. Reinforcement learning algorithms further extend battery lifespan by up to 35%, albeit at a computational complexity of 10,000 FLOPS. With kinetic energy recovery efficiency at 50%, device operation time increases from 3 h to over 10 h. Energy consumption comparisons reveal that smart bands, with a standby power of 20–30 mW, can operate for more than 200 h, whereas smart sports glasses operating in GPS mode consume 90–120 mW and sustain operation for only about 30 h. By integrating material properties with energy management strategies using AI technologies, this research provides a quantitatively supported pathway for the prolonged operation of wearable devices.