The use of unmanned aerial vehicles (UAVs) has grown significantly in various fields, including surveillance, communication, and environmental monitoring. However, their flight endurance is restricted by the limitations of on-board batteries. To address this challenge, UAVs are integrated with reconfigurable intelligent surfaces (RIS) to implement energy harvesting (EH) using simultaneous wireless information and power transfer (SWIPT). This scheme facilitates the simultaneous transfer of both information and power to the UAV’s battery. A reliable deep reinforcement learning (DRL) based method has been used to enhance the performance of the proposed EH-RIS system in dynamic wireless environments. Results from the simulations demonstrated the superiority in the EH efficiency of the DRLbased EH-RIS system, achieving similar performance attained by exhaustive search. Softmax Deep Double Deterministic Policy Gradient (SD3) and Proximal Policy Optimization (PPO) are the different techniques used. Notably, the proposed PPO emerged as standout candidates, demonstrating superior performance in comparison to the others.