PREDICTING TEMPERATURE DYNAMICS OF INDIRECT SOLAR DRYERS BASED ON FUZZY LOGIC

Abstract

The drying process in indirect solar dryers operating under changing weather conditions represents a complex nonlinear system. Modeling these types of dryers is essential for determining their efficiency, analyzing performance, and predicting future operations. Modeling helps optimize the drying process by accounting for the complex dynamics involved. Predicting the temperature in the drying chamber is considered crucial for improving the efficiency of the drying process and enhancing product quality. Temperature control enables effective reduction of product moisture while preserving its nutritional properties. Therefore, accurate forecasting of temperature values in advance is a required prerequisite for improving the quality of the product during the drying process and increasing its overall efficiency. In this article, a series of experimental tests were conducted on the dryer under various weather conditions, and its static and dynamic characteristics were examined. Based on the experimental findings, membership functions were formed using the Mamdani method, and a rule base for the inference mechanism was developed. When the Mamdani algorithm model was trained, it was accurately predicted that with solar radiation at 700 W/m² and ambient temperature at 46 °C, the drying chamber temperature would be 50.9 °C. Similarly, when solar radiation was 750 W/m² and the ambient temperature was around 50 °C, the drying chamber temperature would reach 52 °C, which fully corresponds to the experimental results.
Accuracy of the proposed model demonstrated that the root mean square error (RMSE) was 0.38 °C, and the percentage of the root mean square error (RMSE %) was 0.82 %. This method paves the way for the future development of automatic control systems for solar dryers and further advancements in drying technologies.