AI-Assisted Self-Powered Flexible Electronic System for Energy Management

Abstract

Flexible electronic systems hold immense potential for applications such as wearable devices, electronic skin, and healthcare monitoring, but their advancement is significantly hindered by the bulky size and mechanical limitations of conventional rigid batteries. Furthermore, existing flexible battery technologies still face a persistent trade-off between energy density and mechanical flexibility. To address the fundamental mismatch between energy supply and demand, this review systematically examines recent progress in AI-enabled energy management strategies for self-powered flexible electronic systems. Facilitated by the development of lightweight models and Edge AI, artificial intelligence can now be deployed on resource-constrained devices to reduce reliance on cloud computing and minimize latency. This paper categorizes AI integration across three key dimensions: at the algorithmic level, predictive models and reinforcement learning enable dynamic power allocation and coordinate multi-source energy harvesting from intermittent sources like triboelectric and thermoelectric generators; at the device and circuit level, AI facilitates adaptive voltage regulation and dynamic compensation for hardware performance degradation over time; and at the system level, AI transforms traditional passive architectures into autonomous, self-optimizing platforms capable of prioritizing and scheduling tasks based on real-time energy conditions. While significant progress has been made, challenges regarding the high computational demands of AI, long-term device reliability, system integration complexity, and data privacy remain. Future advancements will rely on cross-layer co-optimization to fully realize sustainable, intelligent, and autonomous flexible electronic platforms.