Real-time energy-efficient framework for multi-source harvesting and adaptive communication IIoT networks

Abstract

Over the past few years, Industrial Internet of Things (IIoT) devices have proliferated across modern manufacturing ecosystems, facilitating real-time monitoring, predictive analytics, and autonomous control. Nevertheless, maintaining these devices in low-resource contexts, especially in situations where the use of wired power is impossible and battery management is not feasible, is a significant challenge. Here, we introduce a new hybrid system that combines multi-source ambient energy harvesting (vibration, heat, and RF) with a lightweight adaptive communication protocol specifically designed for energy-limited industrial settings. Harnessing real-time energy buffer levels and environmental feedback, the architecture utilizes a central middleware engine to adapt transmission behaviors, routing decisions, and node activity states. We design and test detailed models to measure the performance of the framework on major metrics such as energy usage, communication delay, packet delivery ratio, and network sustainability. We utilize a synthetic industrial case scenario with 45 IoT nodes out-deployed across three zones we show the framework's strong benefits over baseline and partial methods, reaching as high as 40 % improvement in node life, 28 % improvement in sustained throughput, and substantial improvements in node availability and energy fairness. This study provides a basis for the installation of self-sustained, resilient IoT systems in realistic industrial environments, connecting the state of the art for energy autonomy design with the state of the art in the field of communication intelligence.