Performance prediction of an auto-cascade refrigeration system using multiple-algorithmic approaches

Wenlian Ye, Yang Liu, Zhongyou Zhou, Lulu Hu and Yingwen Liu

Energy, 2025, vol. 314, issue C

Abstract: This study aims to utilize the Response Surface Methodology (RSM) and six Deep Neural Network algorithms to predict the performance of an Auto-Cascade Refrigeration (ACR) system which is applied in the ultra-temperature field. Firstly, a Box-Behnken experimental design is adopted, and a regression model is proposed to predict the COP, refrigeration capacity, exergy destruction rate, and exergy efficiency with different input parameters. Subsequently, six Deep Neural Network (DNN) algorithms, such as CNN, LSTM, BiLSTM, CNN-biLSTM, CNN-GRU, etc. are employed to compare the predictions of the auto-cascade refrigeration performance parameters. The results indicate that the determination coefficients for all algorithms exceed 0.95, indicating a high degree of consistency between the predicted and actual values. Moreover, the optimal DNN prediction algorithm for each output parameter is identified. Finally, to validate the predictive accuracy, five sets of confirmation parameters are selected to compared the RSM with six other DNN algorithms. The RSM output parameter correlations are 0.99974, 0.99998, 0.99605, and 0.99968. Additionally, for all types of regression evaluation indices, the RSM outperforms the DNN algorithm. The findings presented in this paper offer insights and guidance for forecasting the performance of auto-cascade refrigeration systems.

Keywords: Auto-cascade refrigeration; Response surface methodology; Deep neural network; Performance prediction (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544224039756
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:314:y:2025:i:c:s0360544224039756

DOI: 10.1016/j.energy.2024.134197

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().