An effectiveness model for an indirect evaporative cooling (IEC) system: Comparison of artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS) and fuzzy inference system (FIS) approach

Designing an optimal air conditioning system needs the knowledge of its performance. Soft computing tools like fuzzy inference system (FIS), artificial neural networks (ANN) and adaptive neuro fuzzy inference (ANFIS) provides simple but powerful way for predicting the performance of an IEC. In this paper both analytical as well as soft computing approach is used in predicting the performance of an IEC. All the models are trained with simulation data and are then compared and validated using experimental data from the literature. It was found that of the three models, ANN model gives the most accurate results using the training algorithm Levenberg–Marquardt (LM). The statistical values i.e. R2, RMS, cov, MSE and AIC using ANN for the prediction of primary air outlet temperature were 0.9999, 0.1830, 0.7811, 0.0335 and −3.38, and for effectiveness were 0.9999, 0.00335, 0.5212, 1.119E−05 and −11.38 respectively. This work shows the advantage of ANN over ANFIS and FIS for modeling IEC.