A statistical modeling approach on the performance prediction of indirect evaporative cooling energy recovery systems

Indirect evaporative cooling is well-recognized as a sustainable air-cooling solution to pursue a high quality indoor thermal environment with less energy consumption. In hot and humid areas, the application of an Indirect Evaporative Cooling Energy Recovery System (ERIEC) can cool fresh air to its dew point temperature or lower through water evaporation by using exhaust air from air-conditioned spaces. In this research, a statistical modeling approach is developed to predict the performance of the ERIEC under varying outdoor climates, taking into account possible latent heat transfer from fresh air condensation. Based on training data extracted from numerical simulation, a decision tree model was built to identify the occurrence of condensation through conditional expressions on inlet air temperature and relative humidity. A 2-level factorial design was performed to derive correlations for ERIEC performance indicators under non-condensation and condensation states, respectively. As results, the proposed practical model was validated by experimental data within the deviation of 9.52% on wet-bulb efficiency (η_wb) and 7.69% on enlargement coefficient (ε). A field measurement conducted in Hong Kong shows that the proposed model allows fast and precise prediction on ERIEC performance, with the measured total energy recovery of 5.85 kWh/m² and the predicted value of 5.40 kWh/m² for 30 days in cooling season. The model developed in this study can be efficiently integrated into simulation tools for the performance prediction of ERIEC assisted air-conditioning system in the building energy assessment.