Multi-energy synergy in alfalfa drying: Proximal policy optimization for solar-assisted air-source heat pump systems

To address the challenges of energy-matching uncertainty and precise temperature and humidity control in multi-energy drying systems, this study proposes an adaptive control strategy for a solar-assisted air-source heat pump (SAHP) drying system based on the proximal policy optimization (PPO) algorithm. Due to the intermittent and uncertain nature of solar energy, conventional drying processes struggle to ensure consistent drying quality for alfalfa, often leading to uneven moisture retention and quality degradation. This study developed a synergistic optimization mechanism, integrating the Gym-DryNet simulation environment with a deep reinforcement learning framework, to effectively balance energy input fluctuations and drying process stability. This approach achieved a steady-state control accuracy of 0.3 °C for temperature (relative error of 0.92%) and a humidity regulation accuracy of ± 0.2% RH. Experimental results demonstrate the excellent dynamic anti-interference performance of the intelligent control system. In a 5-h drying cycle, the optimized system produced 160 kg of dried alfalfa (10.3% moisture content) from an initial moisture content of 65.2%, evaporating approximately 250 kg of water at an average rate of 50 kg/h. The total electrical energy consumption was only 3.9 kWh, representing merely 2.3% of the latent heat load associated with the evaporated water. Compared to traditional solar or single heat pump drying systems, this system exhibits significantly reduced temperature fluctuations under full load conditions, demonstrating improved stability and drying efficiency. This fully validates the advantages of the multi-energy synergistic control strategy in optimizing drying efficiency and energy consumption, providing an innovative and intelligent control solution for high-precision drying processes in grass product processing.