This study investigates the applicability of artificial neural networks (ANNs) to predict various performance parameters of a cascade vapour compression refrigeration system. For this aim, an experimental cascade system was set up and tested in steady-state operating conditions. Then, utilizing some of the experimental data for training, an ANN model for the system based on the standard back propagation algorithm was developed. The ANN was used for predicting the evaporating temperature in the lower-temperature circuit, compressor power for the lower and higher circuits and coefficients of performance for both the lower circuit and the overall cascade system. Afterwards, the performances of the ANN predictions were tested using new experimental data. The ANN predictions usually agreed well with the experimental results with correlation coefficients in the range of 0.953-0.996 and mean relative errors in the range of 0.2-6.0%. Furthermore, the ANN yielded acceptable predictions for the system performance outside the range of the experiments. The results suggest that the ANN approach can alternatively and reliably be used for modelling cascade refrigeration systems. Copyright (c) 2006 John Wiley & Sons, Ltd.