کاربرد شبکه عصبی مصنوعی در پیش‌بینی هدایت الکتریکی شیر بازساخته

نویسندگان
حیدر ناصری 1
عیسی حزباوی 2
فیض اله شهبازی 3
1 دانش آموخته کارشناسی ارشد، گروه مکانیک بیو سیستم، دانشگاه لرستان، خرم آباد، ایران
2 استادیار، گروه مکانیک بیو سیستم، دانشگاه لرستان، خرم آباد، ایران
3 دانشیار، گروه مکانیک بیو سیستم، دانشگاه لرستان، خرم آباد، ایران فیض اله شهبازی
10.29252/fsct.16.96.65
چکیده
در این تحقیق هدایت الکتریکی شیر بازساخته با استفاده از روش شبکه عصبی مصنوعی مل­سازی و پیش­بینی گردید. پروتئین (1، 2، 3و 4%)، لاکتوز (4، 6، 8 و 10%)، چربی (3 و 6%) و دما (50، 55، 60 و 65 درجه سلسیوس) به عنوان پارامترهای مستقل ورودی و هدایت الکتریکی شیر بازساخته به عنوان متغیر وابسته خروجی تعریف شدند. داده های به دست آمده از دستگاه سنجش هدایت الکتریکی به منظور آموزش و آزمون شبکه استفاده گردید. به منظور توسعه مدلهای شبکه عصبی مصنوعی ابتدا داد­ ها به سه بخش آموزشی (70%)، اعتبارسنجی (15%) و آزمون (15%) مدل­ها تقسیم شدند. شبکه ها با ساختار پرسپترون چند لایه به صورت دو، سه و چهار لایه آموزش داده شدند. تعداد لایه های مخفی و تعداد نرون ها در هر لایه به روش سعی و خطا به دست آمد. بهترین الگوریتم آموزشی، لونبرگ- مارکوارت با کمترین میزان میانگین مربعات خطا بود. معیار انتخاب بهترین شبکه، بیشترین ضریب تبیین (R2) و کمترین مقدار متوسط مربع خطا (MSE) بود. در پیش بینی هدایت الکتریکی شیر بازساخته شبکه با ساختار 1-4-4 بهترین نتیجه را داد. این شبکه در لایه پنهان 4 نرون دارد. مقادیر ضریب تبیین و خطای آن به ترتیب 992/0 و 011/0 بود. از این نتایج در کارخانجات فراوری شیر می­توان بهره گرفت. همبستگی میان مقادیر آزمایشی و پیش بینی شده در ساختارهای مطلوب بیشتر از 99% به دست آمد.
کلیدواژه‌ها
شیر بازساخته
هدایت الکتریکی
مدل‌سازی
شبکه عصبی مصنوعی

موضوعات

عنوان مقاله English

Application of Artificial Neural Network in Predicting the Electrical Conductivity of Recombined Milk

نویسندگان English

Haidar Naseri 1
isa hazbavi 2
Feizollah Shahbazi 3
1 Graduate Master, Biosystem Engineering, Lorestan University, Khorramabad, Iran
2 Assistant Professor, Biosystem Engineering, Lorestan University, Khorramabad, Iran
3 Associate Professor, Biosystem Engineering, Lorestan University, Khorramabad, Iran
چکیده English

In this study, the moisture content of kiwifruit in vacuum dryer was predicted using artificial neural networks (ANN) method. The protein (1, 2, 3 and 4%), lactose (4, 6, 8 and 10%), fat (3 and 6%) and temperature (50, 55, 60 and 65ºC) were considered as the independent input parameters and electrical conductivity of recombined milk as the dependent parameter. Experimental data obtained from electrical conductivity meter, were used for training and testing the network. In order to develop neural network firstly experimental data were randomly divided into three sets of training (70%), validating (15%) and testing model (15%). In order to develop ANN models, we used multilayer perceptron with back propagation with momentum algorithm. MLP models trained as two, three and four layers. The total number of hidden layers and the number of neurons in each hidden layer were chosen by trial and error. The best training algorithm was LM with the least MSE value. The highest coefficient of determination (R2) and lowest mean squared error (MSE) were considered as the criterion for selecting the best network. The network having three layers with a topology of 4-4-1 had the best results in predicting the electrical conductivity of recombined milk. This network has two hidden layers with 8 neurons in the first hidden layer and 5 neurons in the second hidden layer. For this network, R2 and MSE were 0.992 and 0.011, respectively. These results can be used in milk processing factories. The correlation between the predicted and experimental values in the optimal topologies was higher than 99%.

کلیدواژه‌ها English

Recombined milk
Electrical conductivity
Modeling
artificial neural network
[1] Luck, H., Screed. D. 2002. The use of hydrojenpeoxide in milk and Dairy products. German research institute for food chemistry, 423-452.
[2] Loveland, J.W. 1986. Conductance and oscillometry. 2nd ed., USA: Allyn and Bacon, p. 122-43.
[3] Mabrook, M. F., Petty, M. C. 2002. Application of electrical admittance measurements to the quality control of milk. Sensors and Actuators B, 84: 136–141.
[4] Maatje, K., Huijsmans, P. J. M., Rossing, W., Hogewerf, P. H. 2002. The efficacy of in-line measurement of quarter milk electrical conductivity, milk yield and milk temperature for the detection of clinical and subclinical mastitis. Livest Prod Sci, 30: 239-249.
[5] Crow, D. R. 1994. Principles and application of electro chemistry. 4th ed. Glasgow: Blackie Academic and Professional. Glasgow, UK.
[6] Gelais, D., Champagne, C. P., Erepmoc, F., Audet, P. 1995. The use of electrical conductivity to follow acidifcation of dairy blends. Int Dairy J, 5: 427 438.
[7] Prentice, J.H. 1962. The conductivity of milk the effect of the volume and degree of dispersion of the fat. J Dairy Res, 2: 131 139.
[8] Mabrook, M., Petty, M. 2003. Effect of composition on the electrical conductivity of milk. J Food Eng, 69 (3): 321-325.
[9] Lampert, I.M. 1978. Modern Dairy Products. 3th ed. CRC.USA, p. 92- 132.
[10] Nielen, M., Deluyker, H., Schukken, Y. H., Brand, A. 1992. Electrical conductivity of milk: measurement, modifers, and meta analysis of mastitis detection performance. J Dairy Sci, 75 (2): 606 614.
[11] Petzer, I. M., Donkin, E. F., Du Preez, E., Karzis, J., Van der Schans, T. J., Watermeyer, J.C., Reenen, R. 2008. Value of tests for evaluating udder health in dairy goats: somatic cell counts, California Milk Cell Test and electrical conductivity. Onderstepoort. J Vet Res, 75: 279– 287.
[12] Norberg, E., Hogeveen, H., Korsgaard, I.R., Friggens, N.C., Lbvendahl, P. 2004. Electrical conductivity of milk: ability to predict mastitis status. J. Dairy Sci, 87: 1099–1107.
[13] Zhuang, W., Zhou, W., Nguyen, M.H., Hourigan, J.A. 1997. Determination of protein content of whey powder using electrical conductivity measurement. Int Dairy J, 7(10): 647 653.
[14] Dejmek, P. 1989. Precision conductometry in milk renneting. J Dairy Res, 56 (1): 69-78.
[15] Paqurt, Y. 2000. Electrical conductivity as a tool for analyzing fermentation processes for production of cheese starters. Int Dairy J, 10: 391-399.
[16] Farkas,I., Remenyi, P. & Biro, A. 2000. Modeling aspects of grain drying with a neural network. Computers and Electronics in Agriculture, 29, 99-113.
[17] Menhaj, M. B. 2005. Fundamentals of Artificial Neural Networks. 3th ed. Amirkabir University of Technology Publishers. 718 p. (In Farsi).
[18] Therdthai, N., Zhou, W. 2001. Artificial neural network modeling of the electrical conductivity property of recombined milk, 15(6 8): 1743 1752.
[19] Shin, J.,Yang, D., Gan, L., Hong, S., Lee, E., Park, S., Lee, K. 2012. Preparation of recombined milk using modified butterfats containing α-linolenic acid. Journal of Food Science, 78(1): 17-24.
[20] Schalkoff, R. J. 1997. Artificial neural networks, McGraw-Hill. 422 p.
[21] Khoshtaghaza, M. H., Hosseinzadeh, B., Fayyazi, E., Amirnejat, H. 2016. Prediction of thin layer drying of edible mushroom moisture content by feed forward artificial neural networks method. Journal of Food Science and Technology, 13 (50), 171-182. (In Farsi).
[22] Dayhoff J. E. 1990. Neural Network Principles. Prentice-Hall International, U.S.A.
[23] Nazghelichi, T., Kianmehr, M. H., Aghbashlo, M. 2011. Prediction of carrot cubes drying kinetics during fluidized bed drying by artificial neural network. Journal of food science and technology. 48(5): 542-550.
[24] Sharifi, M., Rafiee, Sh., Kayhani, A., Omid, M. 2010. Kinetic model simulation of thin-layer drying of orange fruit (var. Thompson) using artificial neural network. Journal of Food Science and Technology, 7 (24): 39-49. (In Farsi).
[25] Fazaeli, M., Emam, Z., Omid, M., Kalbasi, A. 2013. Prediction of the Physicochemical Properties of Spray-Dried Black Mulberry (Morus nigra) Juice using Artificial Neural Networks. Food Bioprocess Technol. 6(2): 585-590.
[26] Islam, M. R., Sablani, S. S., Mujumdar. A. S. 2003. An artificial neural network model for prediction of drying rates. Drying Technology, 21, 9: 1867-1884.
[27] Amiri Chayjan, R. 2006. Smart prediction of paddy drying process for optimization of process. doctorate dissertation. Biosystem Engineering. Tarbiat Modares University. Tehran. (In Farsi).