PREDICTION OF NUSSELT NUMBER OF RECTANGULAR FINS USING ARTIFICIAL NEURAL NETWORK MODEL

Abstract

In the passive heat transfer enhancement methods, there are several parameters which need to be optimized to maximize the heat transfer coefficient and Nusselt number while minimizing the pressure drop. For this purpose, empirical correlations are generated by experimental and numerical studies. In this study, a heat transfer analysis of rectangular fins with experimental data by an artificial neural network approach is performed. Artificial Neural Network method is compared with the classical regression model. Different networks with a different number of neurons in the hidden layer and several training algorithms are tested for the defined problem. The results show that the ANN model is found faster and more accurate than conventional techniques if the optimum architecture is generated and convenient training algorithm is chosen for the specific problem. For this   problem, 10-5-1 network with Bayesian Regularization training algorithm is selected as the best scenario with 7.6 % mean absolute percentage error (MAPE) and 0.029 RMSE value while maximum MAPE value is reached to 56.3 % with Levenberg- Marquardt training algorithm and with 10-12-1 network.

Keywords

• Artificial Neural Network,Heat Transfer,Nusselt Number,Overall Heat Transfer Coefficient

PREDICTION OF NUSSELT NUMBER OF RECTANGULAR FINS USING ARTIFICIAL NEURAL NETWORK MODEL

Abstract

Pasif ısı transferi iyileştirme metodlarında ısı transferi kat sayısı ve Nusselt sayısını maximize ederken, basınç düşümünü minimize eden yaklaşımı tespit edebilmek için bir çok parametrenin optimizasyonunun yapılması gerekmektedir. Bu sebepten ötürü, deneysel ve sayısal çalışmalara bağlı olarak ampirik korelasyonlar elde edilmektedir. Bu çalışmada dikdörtgensel finlerin ısı transferi davranışı deneysel ve yapay sinir ağları metodları ile ortaya konmuştur. Yapay sinir ağları metodolojisi ile elde edilen sonuçlar korelasyon ile kıyaslanmıştır.  Ayrıca, tanımlanan problem için yapay sinir ağı uygulamasında farklı eğitim algoritmalarının ve katman sayısının sonuçlar üzerindeki etkisi araştırılmıştır. Elde edilen sonuçlara göre YSA yöntemi, korelasyon yönteminden daha hızlı ve daha doğru sonuç vermektedir. Diğer yandan YSA yaklaşımının doğruluğunun arttırılması için uygun eğitim algoritmasının seçimi, uygun katman sayısının tespiti yani uygun mimarinin elde edilmesi önem arz etmektedir.  Tanımlanan bu problem için, 10-5-1 ağına sahip Bayesian Regularization algoritması %7.6 ortalama yüzde hata ve 0.029 RMSE ile iyi senaryo olarak belirlenmiştir. Maximum ortalama hata %56.3 ile  Levenberg- Marquardt algoritmasında 10-12-1 ağı ile elde edilmiştir. 

Keywords

• Yapay Sinir Ağları,Isı Transferi,Nusselt Sayısı,Isı Transferi Katsayısı

References

1. 1] M. Sheikholeslami, M.G. Bandpy, D.D: Ganji,, Review of heat transfer enhancement methods: Focus on passive methods using swirl flow devices, Renewable and Sustainable Energy Reviews, (49) (2015) 444-649.
2. [2] A. Kraus, Ab., Azız, , J. Welty, Extended Surface Heat Transfer, USA: John Wıley &Sons, (20011).
3. [3] T.M., Jeng, S.C Tzeng,.,Pressure Drop and Heat Transfer of Square Pin-Fin Arrays in in-Line and Staggered Arrangements, International Journal of Heat and Mass Transfer, (50) (2007) 2364-2375.
4. [4] K.E., Starner, Jr H.N. McManus, An Experimental Investigation of Free Convection Heat Transfer From Rectangular Fin Arrays”, Journal of Heat Transfer, Series C, (85) (1963) 273-278.
5. [5], C.K., Tan, J. Ward, S.J. Wilcox, R. Payne, Artificial Neural Network Modelling of the Performance of a Compact Heat Exchanger, Applied Thermal Engineering, (29) (2009) 3609-3617.
6. [6] C. Turk, S. Aradag, S. Kakac, Experimental Analysis of a mixed-plate gasketed plate heat exchanger and artificial neural net estimations of the performance as an alternative to classicalcorrelations, International Journal of Thermal Sciences, (109) (2016) 263-269.
7. [7 ]Y. Islamoglu, A. Kurt,, Heat Transfer Analysis Using ANNs with Experimental Data for Air Flowing in Corrugated Channels,International Journal of Heat and Mass Transfer, (47) (2004) 1361-1365.
8. [8] Y. Islamoglu, Y., A. Kurt, A. Parmaksızoğlu, Performance Prediction for non-adiabatic Capillary Tube Suction Line Heat Exchanger: An Artificial Neural Network Approach, Energy Conversion and Management, (46) (2005) 223-232.

9. [9] G. Xie, B. Sunden, Q. Wang, L. Tang,, Performance Predictions of Laminar and Turbulent Heat Transfer and Fluid Flow of Heat Exchangers Having Large Tube-Diameter and Large Tube-Row by Artificial Neural Networks, International Journal of Heat and Mass Transfer, (52) (2009), 2484-2497.
10. [10] G.N Xie., Q.W. Wang, M. Zeng, L.Q., Loo, Heat Transfer Analysis for Shell-and- Tube Heat Exchangers with Experimental Data by Artificial Neural Networks Approach, Applied Thermal Engineering, (27) (2007), 1096-1104.
11. [11] S., Chen, J., Mao, F., Chen, P., Hou, Y. Li, Development of ANN Model for Depth Prediction of Vertical Ground Heat Exchanger, International Journal of Heat and Mass Transfer, (117) (2018), 617-626.
12. [12] M.H., Shojaeefard, J., Zare, A Tabatabaei, H. Mohammadbeigi, Evaluating Different Types of Artificial Neural Network Structures for Performance Prediction of Compact Heat Exchanger, Neural Comput & Applic, (28) (2017), 3953-3965.
13. [13] E. Ayli, O. Bayer, S. Aradag, Experimental Investigation and CFD analysis of Rectangular Profile FINS in square channel for Forced Convection Regimes, International Journal of Thermal Sciences, (109) (2016), 279-290.
14. [14] E., Ayli, F Kiyici,. O., Bayer, S., Aradag, Experimental investigation of heat transfer and pressure drop over rectangular profile fins placed in a square channel, Convective Heat and Mass Transfer (CONV 2014), June 2014, Turkey, (2014).
15. [15] B.R., Munson, D.F., Young, T.H., Okiishi, Fundementals of Fluid Mechanics,5th Edition, John Wiley&Sons, (2006).
16. [16] D., Chatterjee, G., Biswas, S., Amiroudine, Numerical investigation of forced convection heat transfer in unsteady flow past a row of square cylinders, International Journal of Heat and Fluid Flow, (30) (2009) 1114-1128.
17. [17] Ç., Elmas, Yapay Sinir Ağları (Kuram, Mimari, Eğitim, Uygulama), 1. Baskı,Seçkin Yayıncılık, Ankara, (2003)
18. [18] C., Turk. Yapay Sinir Ağları Yardımıyla Isı Değiştirgeci Modellemesi, MSc Thesis, TOBB University of Mechanical Engineering, Ankara, Turkey, (2013) [19] W.J.Gang, JB Wang, Predictive ANN Models of Ground Heat Exchanger fort he Control of Hybrid Ground Source Heat Pump Systems, Applied Energy, (112,) (2013) 1146-1153.
19. [20] M.H. Beale , M.T. Hagan, H.B. Demuth, Neural Network Toolbox, User’s Guide, Mathworks, (2018).
20. [21] M, Sen.., K.T., Yang, Application of Artificial Neural Networks and Genetic Algortihms in Thermal Engineering, The CRC Handbook of Thermal Engineering, CRC Press, Boca Raton, FL(2000)