What question did this study set out to answer?

This research aims to enhance the prediction accuracy of centrifugal compressor performance using LSTM and transfer learning.

February 11, 2026Open Access

Performance prediction of the centrifugal compressor using a long short-term memory-based model with transfer learning and similarity conversion

Key Points

This research aims to enhance the prediction accuracy of centrifugal compressor performance using LSTM and transfer learning.
Developed a TransLSTM model incorporating LSTM and transfer learning techniques.
Utilized similarity laws to expand the feature domain from limited operational data.
Compared performance with LSTM, Gaussian Process, Gradient Boosting Decision Tree, Random Forest, and Support Vector Machine models.
TransLSTM achieved the highest prediction accuracy with 2.9% deviation for power and 0.971% for compressor ratio.
Demonstrated excellent generalization ability across limited operational data.
Showed significant stability improvements compared to benchmark models.

Abstract

Abstract Centrifugal compressors are commonly employed for pressure compensation in long-distance natural gas pipelines. Accurate prediction of compressor performance is essential for safety and economic considerations. However, due to the invariance of process requirements over time, operational data is usually insufficient to establish a comprehensive model that covers varying operating conditions. This paper presents a prediction method based on the long short-term memory (LSTM) network and transfer learning technique, integrating similarity laws to expand the feature domain and address the issue of limited data. In the case study, the constructed TransLSTM model, a fusion of transfer learning and LSTM, is compared with pure LSTM, Gaussian Process (GP), Gradient Boosting Decision Tree (GBDT), Random Forest (RF), and Support Vector Machine (SVM) models. Results indicate that TransLSTM achieves the highest prediction accuracy, with deviations from actual values of only 2.9% for power and 0.971% for compressor ratio. Verification through comparisons between the training and test sets demonstrates its excellent generalization ability, which is particularly crucial when operational data spans a limited range of working conditions. Additionally, its stability is significantly superior to that of benchmark models.

Ask AI

Helpful

Bookmark

View Full Paper