@article{article_1766754, title={AI-Driven Optimization of a 2.45 Ghz Microwave Drying System for Raw Corn: Enhancing Efficiency and Uniformity via Hybrid CNN-RNN Control}, journal={Black Sea Journal of Engineering and Science}, volume={9}, pages={11–12}, year={2025}, DOI={10.34248/bsengineering.1766754}, author={Murat, Caner}, keywords={Yapay Zeka, Mikrodalga Kurutma, 2, 45 GHz, Dalga Kılavuzu, Gıda Kurutma}, abstract={This study integrates Machine Learning (ML) and Deep Learning (DL) approaches into an integrated methodology to model and optimize the microwave drying of raw corn, which is a multi-variable and non-linear process. The electromagnetic behavior of the process was first simulated using CST Studio Suite software; it was found that a multi-microwave source provides more homogeneous and effective heating compared to a single source. In the experimental phase, classical ML models such as Logistic Regression and SVR, and DL models such as ANN, 1D CNN, and LSTM/GRU were trained using data collected under various input powers (200-500 W) and geometric configurations. The results demonstrated that the CNN-RNN model achieved the highest predictive accuracy for moisture content dynamics. Through systematic AI-driven analysis of experimental data, the optimal drying configuration was identified as 500 W microwave power, 8.1 cm waveguide distance, and 26 cm vertical placement. Under these conditions, 100 grams of raw corn was dehydrated to 40 grams in 5 minutes with minimal quality degradation. The ANN model demonstrated impressive performance metrics in this process, including 0.978 R², 0.041 RMSE, and 0.033 MAE. These results demonstrate the potential of physical simulation and artificial intelligence integration to create a powerful decision support system for improving the efficiency and control of complex industrial processes such as microwave drying.}, number={1}, publisher={Karyay Karadeniz Yayımcılık Ve Organizasyon Ticaret Limited Şirketi}