The objective of this work is to train an artificial neural network (ANN) to predict the performance of gas cooler in carbon dioxide transcritical air-conditioning system. The designed ANN was trained by performance test data under varying conditions. The deviations between the ANN predicted and measured data are basically less than 5%. The well-trained ANN is then used to predict the effects of the five input parameters individually. The predicted results show that for the heat transfer and CO2 pressure drop the most effective factor is the inlet air velocity, then come the inlet CO2 pressure and temperature. The inlet mass flow rate can enhance heat transfer with a much larger CO2 pressure drop penalty. The most unfavorable factor is the increase in the inlet air temperature, leading to the deterioration of heat transfer and severely increase in CO2 pressure drop. (C) 2008 Elsevier Ltd. All rights reserved.