Thermodynamic evaluation and ann optimization of geothermal aided absorbtion refrigeration system for food cooling

In this study, the single stage geothermal energy aided absorption refrigeration system, working with ammonia-water solution, was optimized in the view of different solution concentrations and design parameters. Depending on parametric evaluation, a number of 3360 designs were made. In these frameworks, changes of the system performance by the effect of generator, evaporator, condenser and absorber temperature were investigated. The capacities of generator, evaporator, condenser, absorber, heat exchanger, pump, coefficient of performance (COP) and exergy efficiency were calculated by energy and exergy analysis. In the calculations, different ratios of NH3-H2O solution were taken into account. These ratios are 46% NH3-54%H2O, 40% NH3-60%H2O, 30% NH3-70%H2O, 20%NH3-80%H2O. Thermodynamic values of NH3-H2O solutions were determined by an engineering software (REFPROP). In the next part of the study, cost analysis of refrigeration system was made for each design. Assuming the economic life of the absorption refrigeration systems as 20 years, the economics of the system was investigated by Net Present Value analysis (NPV). Finally, all economically evaluated designs were modelled using Artificial Neural Network (ANN). In this aim, a software called MATLAB was used. In the models used to guess COP, efficiency of exergy and NPVs, a number of 14640 data was formed. 10248 of this data were used to train and the rest of data in a number of 4392 were used to test the network. The network architecture was trained for 1000 iterations. Statistical results of network outputs were separately calculated for the each product. When analytically calculated values of COP, exergy efficiency and NPV compared with the results calculated by ANN modelling, it was decided that it is statistically acceptable. Keywords: Absorption Refrigeration System, Ammonia-Water, Artificial Neural Network, Energy and Exergy Analysis, Geothermal Energy, Optimization.