Abstract
Medicinal herbs are commonly used worldwide for their therapeutic, nutritional, and medicinal benefits, with peppermint being one of the most valued aromatic herbs. Despite their widespread use, energy-efficient drying methods for heat-sensitive medicinal herbs, particularly peppermint leaves, remain underexplored, and conventional techniques often lead to quality degradation and high-energy consumption. Therefore, this study examined the thermal and energy consumption of drying peppermint leaves using an integrated infrared-hot air system at airflow rates of 0.3, 0.5, and 1.0 m/s, hot-air temperatures of 35, 45, and 55 °C; and radiation intensities of 0.08, 0.10, and 0.15 W/cm². Approximately 300 g of peppermint leaves were evenly spread on a stainless-steel mesh conveyor, and weight loss during drying was tracked using a 0.01 g precision load cell. Infrared intensity, air temperature, and airflow velocity were measured using pyranometers, thermometers, and anemometers, respectively. Furthermore, 11 different machine learning models were applied to predict the relationships between the input parameters (infrared power, airflow rate, and air temperature) and response variables, including total energy utilization, specific energy consumption, and thermal and drying efficiency. Our findings indicate that increasing the infrared power and air temperature shortened the drying periods, while increasing the airflow led to an extended drying time. The study also revealed that increased air temperature, infrared intensity, and reduced airflow rates enhanced energy indices. Among the 11 machine learning models evaluated, the Kucuk and Midilli models best fit the drying curves, making them most suitable for predicting peppermint drying behavior. The findings showed that moderate infrared levels, lower temperatures, and higher air velocities can optimize energy use and reduce drying time in peppermint leaves, highlighting the potential of advanced heating technologies to improve food drying efficiency.