An Application of Nonlinear Autoregressive (NARX) Model to Predict Adsorbent Bed Temperature of Solar Adsorption Refrigeration System

Abstract

In any solar adsorption refrigeration system, there are three major components: a solar collector adsorbent bed, a condenser and an evaporator. All of those components operate at different temperature levels. A solar collector with a tubular adsorbent configuration is proposed and numerically investigated. In this study, a nonlinear auto-regressive model with exogenous input is applied for the prediction of adsorbent bed temperature during the heating and desorption period. The developed neuronal model uses the MATLAB Network toolbox to obtain a better configuration network, applying multilayer feed-forward, the TANSIG transfer function, and the back-propagation learning algorithm. The input parameters are ambient temperature and the uncontrolled natural factor of solar radiation. The output network contains a variable representing the adsorbent bed temperature. The values obtained from the network model were compared with the experimental data, and the prediction performance of the network model was examined using various performance parameters. The mean square error (MSE) and the statistical coefficient of determination (R ²) values are excellent numerical criteria for evaluating the performance of a prediction tool. A well-trained neural network model produces small MSE and higher R ² values. In the current study, the adsorbent bed temperature results obtained from a neural network with a two neuron in hidden layer and the number of the tapped time-delays d = 9 provided a reasonable degree of accuracy: MSE = 1.0121 and R ² = 0.99864 and the index of agreement was 0.9988. This network model, based on a high-performance algorithm, provided reliable and high-precision results concerning the predictable temperature of the adsorbent bed.

Keywords

Artificial neural networks / NARX / solar collector / adsorption refrigerator / adsorbent bed temperature

References

[1] Abdel Aziz A A, Hatab S I, Moawed M, Zohir A E, Berbish N M. Experimental study on the effect of adsorber with three shapes of conductive material on performance of adsorption refrigeration tube using activated carbon/ethanol pair. Applied Thermal Engineering, 2018, 131: 897-909.

[2] Alelyani S M, Bertrand W K, Zhang Z, Phelan P E. Experimental study of an evacuated tube solar adsorption cooling module and its optimal adsorbent bed design. Solar Energy, 2020, 211: 183-191.

[3] Allouhi A, Kousksou T, Jamil A, Zeraouli Y. Modeling of a thermal adsorber powered by solar energy for refrigeration applications. Energy, 2014, 75: 589-596.

[4] Asif Sha A, Baiju V, Rehna R S, Suzuki T, Singh H, Ichiyanagi M. Performance investigations of carbon bsed consolidated composite adsorbents effective for adsorption cooling systems. Applied Thermal Engineering, 2022, 217: 119199.

[5] Bouzeffour F. Artificial neural network-based modeling for the prediction of heat and mass transfer coefficient of the adiabatic liquid desiccant system. Journal of Renewable Energies, 2022, 25(2): 157-167.

[6] Bouzeffour F, Khelidj B, Taharabbes M. Experimental investigation of a solar adsorption refrigeration system working with silicagel/water pair: A case study for Bou-Ismail solar data. Solar Energy, 2016, 131: 165-175.

[7] Bouzeffour F, Khelidj B, Yahi F, Belkacemi D, Taane W. Performance prediction of a liquid desiccant dehumidifier using artificial neural networks approach. Science and Technology for the Built Environment, 2021, 27(2): 211-225.

[8] Bouzeffour F, Belkacemi D (2018). Artificial neural network modeling to predict the moisture removal rate of a desiccant liquid dehumidifier system. 6th International Renewable and Sustainable Energy Conference (IRSEC). Rabat, Morocco, December 05–08, 2018.

[9] Chan K C, Tso C Y, Wu C, Chao C Y H. Enhancing the performance of a zeolite 13X/CaCl2-water adsorption cooling system by improving adsorber design and operation sequence. Energy and Buildings, 2018, 158: 1368-1378.

[10] Chen Q F, Du S W, Yuan Z X, Sun T B, Li Y X. Experimental study on performance change with time of solar adsorption refrigeration system. Applied Thermal Engineering, 2018, 138: 386-393.

[11] Coruh S, Geyikçi F, Kiliç E, Coruh U. The use of NARX neural network for modeling of adsorption of zinc ions using activated almond shell as a potential biosorbent. Bioresource Technology, 2014, 151: 406-410.

[12] Dias J M S, Costa V A F. Evaluating the performance of a coated tube adsorber for adsorption cooling. International Journal of Refrigeration, 2020, 118: 21-30.

[13] Du B, Lund P D, Wang J. Improving the accuracy of predicting the performance of solar collectors through clustering analysis with artificial neural network models. Energy Reports, 2022, 8: 3970-3981.

[14] ElFadar A. Thermal behavior and performance assessment of a solar adsorption cooling system with finned adsorber. Energy, 2015, 83: 674-684.

[15] Eldokaishi A O, Abdelsalam M Y, Kamal M M, Abotaleb H A. Modeling of water-PCM solar thermal storage system for domestic hot water application using artificial neural networks. Applied Thermal Engineering, 2022, 204: 118009.

[16] Elsheniti M B, Abd El-Hamid A T, El-Samni O A, Elsherbiny S M, Elsayed E. Experimental evaluation of a solar two-bed lab-scale adsorption cooling system. Alexandria Engineering Journal, 2021, 60(3): 2747-2757.

[17] Fischer S, Frey P, Drück H. A comparison between state-of-the-art and neural network modelling of solar collectors. Solar Energy, 2012, 86(11): 3268-3277.

[18] Frey P, Fischer S, Drück H. Artificial Neural Network modelling of sorption chillers. Solar Energy, 2014, 108: 525-537.

[19] Gai X, Song J, Wang L, He B. Numerical analysis of heat pipe-assisted finned adsorber with FAM-Z02/water pair for vehicle air conditioning. Applied Thermal Engineering, 2022, 213: 118715.

[20] Ghritlahre H K, Chandrakar P, Ahmad A. Application of ANN model to predict the performance of solar air heater using relevant input parameters. Sustainable Energy Technologies and Assessments, 2020, 40: 100764.

[21] Hagan M T, Demuth H B, Beale M H, De Jesus O (2014). Neural Network Design, Martin Hagan.

[22] Hosoz M, Ertunc H M, Bulgurcu H. Performance prediction of a cooling tower using artificial neural network. Energy Conversion and Management, 2007, 48(4): 1349-1359.

[23] Huang Y, Li C. Accurate heating, ventilation and air conditioning system load prediction for residential buildings using improved ant colony optimization and wavelet neural network. Journal of Building Engineering, 2021, 35: 101972.

[24] Islam M P, Morimoto T. Performance prediction of solar collector adsorber tube temperature using a nonlinear autoregressive model with exogenous input. International Journal of Computer Applications, 2015, 114(12): 24-32.

[25] Kumar A, Kapilan N, P A D, Kasthurirengan S. Experimental studies on solar assisted activated carbon based adsorption refrigeration system. Materials Today: Proceedings, 2022, 62: 5258-5265.

[26] Keddouda A, Ihaddadene R, Boukhari A, Atia A, Arici M, Lebbihiat N, Ihaddadene N. Solar photovoltaic power prediction using artificial neural network and multiple regression considering ambient and operating conditions. Energy Conversion and Management, 2023, 288: 117186.

[27] Khalil A, El-Agouz E S A, El-Samadony Y A F, Sharaf M A. Experimental study of silica gel/water adsorption cooling system using a modified adsorption bed. Science and Technology for the Built Environment, 2016, 22: 41-49.

[28] Lakshmipathy B, Sivakumar K, Senthil Kumar M, Kajavali A, Sivaraman B. Artificial neural network and experimental work of a solar cavity collector. Materials Today: Proceedings, 2021, 47: 5289-5296.

[29] Lashkarbolooki M, Shafipour Z S, Hezave A Z. Trainable cascade-forward back-propagation network modeling of spearmint oil extraction in a packed bed using SC-CO2. The Journal of Supercritical Fluids, 2013, 73: 108-115.

[30] Mellit A, Kalogirou S A. Artificial intelligence techniques for photovoltaic applications: A review. Progress in Energy and Combustion Science, 2008, 34: 574-632.

[31] Moghaddamnia A, Remesan R, Kashani M H, Mohammadi M, Han D, Piri J. Comparison of LLR, MLP, Elman, NNARX and ANFIS Models - With a case study in solar radiation estimation. Journal of Atmospheric and Solar-Terrestrial Physics, 2009, 71(8–9): 975-982.

[32] Mohammed R H, Mesalhy O, Elsayed M L, Chow L C. Performance enhancement of adsorption beds with silica-gel particles packed in aluminum foams. International Journal of Refrigeration, 2019, 104: 201-212.

[33] Montgomery D C, Runger G C. Applied Statistics and Probability for Engineers, 2010, USA: John Wiley and Sons

[34] Mostafa A, Hassanain M, Elgendy E. Transient simulation and design parameters optimization of a cold store utilizes solar assisted adsorption refrigeration system. Case Studies in Thermal Engineering, 2022, 37: 102273.

[35] Mudhafar M A H, Pan H Y. An experimental study and performance evaluation of a small adsorption airconditioning system with FAM Z05 Zeolite and water. International Journal of Refrigeration, 2022, 138: 206-219.

[36] Nabipour M. Prediction of surface tension of binary refrigerant mixtures using artificial neural networks. Fluid Phase Equilibria, 2018, 456: 151-160.

[37] Samani N, Gohari-Moghadam M, Safavi A A. A simple neural network model for the determination of aquifer parameters. Journal of Hydrology, 2007, 340(1–2): 1-11.

[38] Taki M, Farhadi R. Modeling the energy gain reduction due to shadow in flat-plate solar collectors: Application of artificial intelligence. Artificial Intelligence in Agriculture, 2021, 5: 185-195.

[39] Unvar S, Çolak A B, Menlik T. Experimental analysis of the effect of Nano fluid use on power and efficiency enhancement in heat pipe solar collectors and modeling using artificial neural networks. Heat Transfer Research, 2023, 54(13): 1-18.

[40] Wang Y, Li M, Du W, Yu Q, Ji X, Ma X. Performance comparative study of a solar-powered adsorption refrigerator with a CPC collector/adsorbent bed. Energy Conversion and Management, 2018, 173: 499-507.

[41] Willmott C J. On the validation of models. Physical Geography, 1981, 2(2): 184-194.

[42] Yang K T. Artificial neural networks (ANNs): A new paradigm for thermal science and engineering. Journal of Heat Transfer, 2008, 130(9): 093001.

[43] Yu Y, Pan Q W, Wang L W. A small-scale silica gelwater adsorption system for domestic air conditioning and water heating by the recovery of solar energy. Frontiers in Energy, 2020, 14: 328-336.

[44] Zendehboudi A, Tatar A, Li X. A comparative study and prediction of the liquid desiccant dehumidifiers using intelligent models. Renewable Energy, 2017, 114: 1023-1035.