Application of numerical analysis and machine learning techniques to improve drying performance and energy consumption of microwave-assisted convective dryer

• The drying sector's modeling, and optimization applications are expanding rapidly with AI. • The energy, thermal efficiency, and drying efficiency, are calculated in thermodynamics • Combined technology is a modern okra drying technology with significant research value • Machine learning a new approach for improving food drying techniques This study investigates the drying performance and energy consumption of okra slices using a hybrid microwave-assisted convective dryer. The effects of air temperature (40-60°C), microwave power (100-400 W), and airflow (0.5-1.5 m/s) on drying time, energy use, and thermodynamic performance were evaluated. Thermodynamic parameters, including energy consumption, drying efficiency, and thermal efficiency, were calculated. Machine learning models, specifically Feedforward Backpropagation (FFBP) and Cascade Forward Backpropagation (CFBP) artificial neural networks, were developed to predict drying outcomes based on input variables. Results showed that increasing air temperature and microwave power reduced drying time (minimum 21 minutes) and specific energy consumption (minimum 2.37 kWh/kg), while higher airflow increased drying time. Maximum drying and thermal efficiencies reached 31.5% and 28.9%, respectively. The developed ANN models achieved high predictive accuracy (R² > 0.99). The combined microwave-hot air-drying technology is a modern okra drying method with significant research value and practical production potential. These findings provide a scientific foundation for optimizing the combined microwave-hot air drying of okra, supporting its industrial application for improved efficiency and product quality.