Interpretability Study on the Fault Diagnosis Model of the Heat pipe/ Vapor Compression Composite Air Conditioning System
Applying data-driven fault diagnosis models to data center air conditioning systems can significantly improve operational reliability. However, such models often lack diagnostic interpretability, limiting their application. This study develops a composite fault diagnosis model based on typical machi...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | Chinese |
| Published: |
Journal of Refrigeration Magazines Agency Co., Ltd.
2025-01-01
|
| Series: | Zhileng xuebao |
| Subjects: | |
| Online Access: | http://www.zhilengxuebao.com/zh/article/117507343/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Applying data-driven fault diagnosis models to data center air conditioning systems can significantly improve operational reliability. However, such models often lack diagnostic interpretability, limiting their application. This study develops a composite fault diagnosis model based on typical machine learning algorithms, compares the diagnostic performance of different models, and finally conducts interpretability research on the diagnostic models using the SHAP method. The results demonstrate that the CNN-based fault diagnosis model achieves optimal performance in both heat pipe and vapor compression modes, with F-1 scores exceeding 0.999 across all classifications. In heat pipe mode, the diagnosis of CNN model primarily relies on condenser fan frequency, outdoor temperature, and refrigerant pump power consumption as key features, whereas in vapor compression mode, the dominant features are outdoor temperature, compressor frequency, and subcooling degree. |
|---|---|
| ISSN: | 0253-4339 |