Parameter optimization of drying models using Honey Formation Optimization-1 (HFO-1)
Food and Bioproducts Processing, cilt.154, ss.392-401, 2025 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 154
- Basım Tarihi: 2025
- Doi Numarası: 10.1016/j.fbp.2025.09.016
- Dergi Adı: Food and Bioproducts Processing
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC
- Sayfa Sayıları: ss.392-401
- Anahtar Kelimeler: Bitter Gourd, Drying, GA, HFO-1, Modeling, PSO
- Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
- Akdeniz Üniversitesi Adresli: Evet
Özet
Many mathematical studies in recent years has been proposed to model the drying curves for various grains, fruits and vegetables. In this study, one of the recent optimization algorithms, Honey Formation Optimization with single component (HFO-1) and also Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are utilized to model the drying curves of the bitter gourd (Momordica charantia L.). For this purpose, the bitter gourd was sliced in 8 mm slice thickness and dried at 70°C drying air temperature in convective dryer by applying different pre-treatments such as dipping time (10, 20 and 30 h) and different sugar and grape molasses content (pekmez) (50, 60 and 70 Brix). During the drying period, weight measurements are recorded until the products reached equilibrium moisture content. The values are obtained from 14 commonly used thin layer drying models. The results show that the HFO-1 is observed to perform better than GA and PSO in optimizing the parameters of each thin layer drying models. With HFO-1, the Modified Henderson and Pabis I and II are found to provide better results with respect to RMSE, χ2, MAE and EF metrics. The best RMSE, χ2, MAE and EF values are achieved in the Modified Henderson and Pabis I model with a value of 5.74E-04, 4.43E-07, 4.08E-04 and 1.00E+ 00, respectively, for the samples dipped in 70 Brix sugar for 10 h. With respect to the lowest RMSE values achieved, HFO-1 is observed to perform at least 6.84E+ 02 % better than the both PSO and GA. Altogether, this study presents three contributions. First, it introduces the novel use of the recently developed HFO-1 algorithm to optimize thin-layer drying models—a context in which this algorithm has not been previously applied. Second, these 14 drying models, along with their associated pre-treatments, are collectively applied for the first time to bitter gourd fruit.Third, it provides a robustness evaluation of the drying model against noisy data, enhancing their reliability for practical use. The findings of this study enable significant advancements, including more accurate prediction of drying time and shelf life, improved process optimization and control, enhanced design of efficient drying systems, and strengthened efforts toward final product quality and standardization.