Modelling the performance parameters of a horizontal falling film absorber with aqueous (lithium, potassium, sodium) nitrate solution using artificial neural networks

An ANN (artificial neural network) model was developed to determine the efficiency parameters of a horizontal falling film absorber at operating conditions of interest for absorption cooling systems. The aqueous nitrate solution LiNO₃ + KNO₃ + NaNO₃ with salt mass percentages of 53%, 28% and 19%, respectively, was used as a working fluid. The authors created the ANN from the database they had compiled with the results of experiments that they had performed in a set-up designed and built for this purpose. The ANN structure consisted of 6 input variables: inlet solution and cooling water temperatures, cooling water and solution mass flow rates, absorber pressure and inlet solution concentration; 4 output variables which facilitated the assessment of the performance of the absorber: heat and mass transfer coefficients, absorption mass flux and the degree of subcooling of the solution leaving the absorber. The hidden layer contained 9 neurons which were determined by training and test procedures. The results showed that the deviation between the experimental data and the estimated values was well adjusted. This indicated that the ANN model was an effective tool for predicting the efficiency parameters of the absorber. The solution flow rate was also observed to be the most significant operating variable which affected the performance of the absorber. - Highlights: • An ANN was developed to predict the efficiency parameters of a falling film absorber. • The ANN was created using a database corresponding to a triple-effect absorption chiller. • The ANN predicts the efficiency parameters of falling film absorbers with r² > 0.95. • The solution flow rate is the variable that most affects the performance of the absorber.