Real-Time Sensed Data Scheduling in Power Transmission Corridors Using Edge Computing and Reinforcement Learning

Ensuring the safety and reliability of power transmission corridors requires effective real-time monitoring of environmental and operational conditions. However, terminal sensing devices deployed for monitoring, such as meteorological, thermal imaging, and infrared sensors, often face significant resource constraints, limiting their ability to process and analyze the generated data locally. This challenge is compounded by the need for timely data handling to prevent potential threats, such as wildfires or structural failures. To address these challenges, we propose an edge computing-based data offloading strategy that transfers sensed data to nearby edge servers for efficient processing. The offloading decision-making process is optimized using the Proximal Policy Optimization (PPO) algorithm, a reinforcement learning approach designed to minimize latency and enhance data scheduling efficiency. Our proposed method dynamically adapts to varying network and computational conditions, ensuring optimal data handling and real-time responsiveness. Experimental results demonstrate that our approach significantly reduces data processing latency and improves overall system performance compared to traditional data scheduling techniques, validating its efficacy for real-time monitoring in power transmission corridors.