Abstract
Grain drying is a critical stage in post-harvest processing, directly affecting storage stability, product quality and energy consumption. Conventional drying systems typically operate under fixed parameter regimes and rely on manual or simple feedback control strategies, which often lead to inefficient energy use and inconsistent moisture content in the final product. Recent advances in artificial intelligence, digital transformation technologies, and predictive control algorithms offer promising opportunities to enhance the performance of drying systems.
This paper investigates modern drying technologies and proposes an adaptive Model Predictive Control (MPC) strategy for drum grain dryers. The proposed control framework integrates real-time sensor data, process modeling and predictive optimization to dynamically adjust drying parameters such as air temperature, airflow velocity, and drum rotation speed.
The results demonstrate that MPC-based adaptive control can improve moisture stabilization accuracy, reduce energy consumption, and increase process reliability. The approach is particularly promising for drying agricultural products such as sunflower seeds, where maintaining seed viability and oil quality is essential.