مدل سازي فرايند خشك كردن بادمجان توسط سامانه مادون قرمز به روش الگوريتم ژنتيك- شبكه عصبي مصنوعي

Modeling of Eggplant Drying Process by Infrared System using Genetic Algorithm–Artificial Neural Network Method

در اين مطالعه رفتار خشك كردن لايه نازك ورقه هاي بادمجان در يك خشك كن مادون قرمز (IR) بررسي گرديد. اثر توان لامپ مادون قرمز (150، 250 و 375 وات)، فاصله نمونه از لامپ (5، 10 و 15 سانتي متر)، ضخامت نمونه ها (5 و10سانتي متر) و زمان خشك كردن بر خشك شدن ورقه هاي بادمجان موردبررسي قرار گرفت. نتايج خشك كردن بادمجان به روش مادون قرمز نشان داد با افزايش توان لامپ و كاهش فاصله نمونه ها از منبع حرارتي، سرعت خشك كردن افزايش مي يابد. مقدار كاهش وزن، با افزايش توان لامپ از 150 به 375 وات، از 31/08 به 92/44 درصد افزايش يافت. ميزان وزن، با افزايش فاصله لامپ از 5 به 15 سانتي متر، از 92/44 به 31/15 درصد، كاهش يافت. در اين پژوهش مدل سازي فرآيند به روش الگوريتم ژنتيك - شبكه عصبي مصنوعي با 4 ورودي (توان، فاصله لامپ، ضخامت نمونه و زمان خشك كردن) و 1 خروجي جهت پيشگويي كاهش وزن انجام شد. نتايج مدل سازي به روش الگوريتم ژنتيك - شبكه عصبي مصنوعي نشان داد شبكه اي با تعداد 14 نرون در يك لايه پنهان و با استفاده از تابع فعال سازي سيگموئيدي مي توان به خوبي درصد كاهش وزن در طي فرآيند خشك كردن بادمجان به روش مادون قرمز را پيشگويي نمود (99/0R=). نتايج آناليز حساسيت توسط شبكه عصبي بهينه نشان داد كه توان لامپ مادون قرمز به عنوان موثرترين عامل براي كنترل كاهش وزن بادمجان است.

Background and objectives: Due to low energy efficiency and prolonged drying time of agricultural products by conventional methods, application of the new techniques must be investigated. One of the new techniques in food drying is emplying the infrared radiation systems that increase the drying rate, enhance the final product quality, and decrease the process costs. In this study, the thin-layer drying behavior of eggplant slices in an infrared dryer (IR) was investigated. Materials and methods: The effect of infrared lamp power (150, 250 and 375 watt), distance of sample from lamp (5, 10 and 15 cm), samples thickness (0.5 and 1 cm) and drying time on drying of eggplant slices were examined. For measuring the weight of the samples during experimentation without taking them out of the dryer, the tray with samples was suspended on the digital balance. Also, in this study, process modeling was done with the genetic lgorithm– artificial neural network (GA-ANN) method with 4 inputs (power and lamp distance, sample thickness and drying time) and 1 output for prediction of weight reduction. Results: The results of infrared drying of eggplant showed that with increasing in lamp power and decreases in sample distance from the heat source, the drying rate increases. Increases in eggplant slice thickness had significant effect on weight loss during irradiation (p<0.05) and with increases in sample thickness, drying time increased. With increasing in drying time, amount of evaporated moisture was increased. With increase in infrared power from 150 to 375 watts, weight loss increased from 31.08 to 92.44%. With increase in lamp distance from 5 to 15 cm, weight loss decreased from 92.44 to 31.15%. Due to the lower error value obtained by using the sigmoid activation function, this type of function was selected as the activation function on the hidden and output layer. Based on the test and error method, if 20% of the data is used for training, the network is well able to learn the relationships between inputs and outputs. 20% of the data was used for validating the trained network. The remaining 60 percent of the data was also used to testing the network. Conclusion: The GA-ANN modeling results showed a network with 14 neurons in one hidden layer with using sigmoid function can be well predict the weight loss in eggplant drying by infrared system (R=0.99). Sensitivity analysis results by optimum ANN showed the infrared power was the most sensitive factor for controlling the weight loss of samples.