Document Type : Original Article
Subjects
Performance of Machine Learning Model to Predict Thermodynamical Properties of an Active Solar Dryer for Pistachio Nut Dehydration
Mohsen Mokhtarian (MSc)1*, Ahmad Kalbasi-Ashtari (PhD)2, Fatemeh Koushki (MSc)3, Hong-Wei Xiao (PhD)4
1 College of Food Science and Nutritional Engineering, China Agricultural University, Beijing 100085, PR China
2 Biological and Agricultural Engineering Department, Texas A&M University, College Station, Texas, USA
3 Department of Food Science and Technology, Roudehen Branch, Azad University, Roudehen, Iran
4 College of Engineering, China Agricultural University, Beijing 100083, PR China
Received: 11.04.2023 Accepted: 16.06.2023
Abstract
Introduction: Pistachio is one of Iran's strategic products, playing a significant role in providing the country's foreign exchange resources. The quality of the dried product is of high importance. Therefore, solar drying is one of the new and suitable methods for drying this strategic product.
Materials and Methods: In this study, a solar dryer was used to dry fresh pistachios. Fresh samples were dried to a safe moisture content of approximately 5% (wet basis). During the drying process, the thermodynamic parameters of the solar collector (as the heat energy generator for the dryer) were measured, including solar radiation intensity and the temperature of the air exiting the collector. These parameters were then predicted using an intelligent artificial neural network system (multilayer perceptron network).
Results: The results of this research indicated that the single-layer neural network with 8 neurons in the hidden layer provided the best fit in predicting solar radiation intensity (R²=0.988) and the temperature of the air exiting the collector (R²=0.941).
Conclusion: Overall, the findings of this study showed that the artificial neural network has a high capability in predicting the thermodynamic characteristics of the solar dryer.
►Please cite this article as follows:
Mokhtarian M, Kalbasi-Ashtari A, Koushki F, Xiao HW. Performance of Machine Learning Model to Predict Thermodynamical Properties of an Active Solar Dryer for Pistachio Nut Dehydration. Pistachio and Health Journal. 2023;6(1-2):36-45.