Producing non-traditional flour from watermelon rind pomace: Artificial neural network (ANN) modeling of the drying process

• Published in: Journal of environmental management 2021-03, Vol.281, p.111915-111915, Article 111915
• Main Authors: Fabani, María Paula, Capossio, Juan Pablo, Román, María Celia, Zhu, Wenlei, Rodriguez, Rosa, Mazza, Germán
• Format: Article
• Language: English
• Subjects: Citrullus; Desiccation; Drying kinetics; Flour; Genetic algorithm; Modeling; Neural Networks, Computer; Non-traditional flour; Solar and convective drying; Temperature; Watermelon rind
• ISSN: 0301-4797; 1095-8630
• DOI: 10.1016/j.jenvman.2020.111915

An artificial neural network (ANN) model was developed to simulate the convective drying process of watermelon rind pomace used in the fabrication of non-traditional flour. Also, the drying curves obtained experimentally were fitted with eleven different empirical models to compare both modeling approaches. Lastly, to reduce the required fossil fuel in the convective drying process, two types of solar air heaters (SAH) were presented and experimentally evaluated. The optimization of the ANN by a genetic algorithm (GA) resulted in an optimal number of neurons of nine (9) for the first hidden layer and ten (10) for the second hidden layer. Also, the ANN performed better than the best fitted empirical model. Simulations with the trained ANN showed very promising generalization capabilities. The type II SAH showed the best performance and the highest air temperature it reached was 45 °C. The specific energy consumption (SEC) needed to dry the watermelon rind at this temperature and the CO2 emissions were 609 kWh.kg−1 and 318 kg CO2.kWh−1, respectively. Using the type II SAH, this energy amount would be saved without CO2 emissions. To reach higher drying temperatures the combination of the SAH and the electrical convective dryer is possible. [Display omitted] •The convective drying of watermelon rind is performed and modeled by ANNs.•Two types of solar air heaters (SAH) are presented and evaluated.•The model's hyperparameters are optimized using a genetic algorithm.•An algorithm to correct generalization problems is presented.•The performance of such a model is compared with the usual Two-Term model.