Charge scheduling for wireless rechargeable sensor networks with multiple mobile charges that uses hybrid reinforcement learning is energy-efficient and lifespan-aware

To ensure continuous sensor coverage, Wireless Rechargeable Sensor Networks (WRSNs) utilise Mobile Chargers (MCs) and wireless drones for energy transfer. However, drones alone are unsuitable for large WRSNs due to limited battery capacity. Existing charge scheduling methods face latency, efficiency, and scalability issues. We propose the Energy Efficient Network Lifespan aware Charge Scheduler (EENL-CS), which uses an Improved Grasshopper Optimisation (IGO) for clustering and an Improved Butterfly Optimisation (IBO) to select Cluster Heads. A Self-Healing Deep Reinforcement Learning (SHDRL) model schedules charging. Simulations show EENL-CS improves charging timeliness, efficiency, cost, and overall network lifetime.