CNN classification result.
<div><p>Centrifugal compressors are widely used in the oil and natural gas industry for gas compression, reinjection, and transportation. Fault diagnosis and identification of centrifugal compressors are crucial. To promptly monitor abnormal changes in compressor data and trace the cause...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1852023519841353728 |
|---|---|
| author | Yuan Wang (14955) |
| author2 | Shaolin Hu (19835870) |
| author2_role | author |
| author_facet | Yuan Wang (14955) Shaolin Hu (19835870) |
| author_role | author |
| dc.creator.none.fl_str_mv | Yuan Wang (14955) Shaolin Hu (19835870) |
| dc.date.none.fl_str_mv | 2025-01-16T18:57:40Z |
| dc.identifier.none.fl_str_mv | 10.1371/journal.pone.0315917.g022 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/figure/CNN_classification_result_/28224597 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Science Policy Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified natural language processing dynamic knowledge graph convolutional neural network petrochemical big data natural gas industry missed alarms based method effectively overcomes low oil pressure compressor data anomalies data anomalies compressor data method starts gas compression false alarms widely used two types tolerant filtering three parts security monitoring results show paper proposes non starting fault tracing fault diagnosis drive machine different sources causes leading achieves 100 |
| dc.title.none.fl_str_mv | CNN classification result. |
| dc.type.none.fl_str_mv | Image Figure info:eu-repo/semantics/publishedVersion image |
| description | <div><p>Centrifugal compressors are widely used in the oil and natural gas industry for gas compression, reinjection, and transportation. Fault diagnosis and identification of centrifugal compressors are crucial. To promptly monitor abnormal changes in compressor data and trace the causes leading to these data anomalies, this paper proposes a security monitoring and root cause tracing method for compressor data anomalies. Additionally, it presents an intelligent system design method for fault tracing in compressors and localization of faults from different sources. This method starts from petrochemical big data and consists of three parts: fault dynamic knowledge graph construction, instrument data sliding fault-tolerant filtering, and the fusion and reasoning of fault dynamic knowledge graph and instrument data variation monitoring. The results show that this method effectively overcomes the problems of false alarms and missed alarms based on fixed threshold alarm methods, and achieves 100% classification of two types of faults: non starting of the drive machine and low oil pressure by constructing a PCA (Principal Component Analysis)—SPE (Square Prediction Error)—CNN (Convolutional Neural Network) classifier. Combined with dynamic knowledge graph and NLP (Natural Language Processing) inference, it achieves good diagnostic results.</p></div> |
| eu_rights_str_mv | openAccess |
| id | Manara_b7ea480b4ebe8e1efbc564b2fa2cd99f |
| identifier_str_mv | 10.1371/journal.pone.0315917.g022 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/28224597 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | CNN classification result.Yuan Wang (14955)Shaolin Hu (19835870)Science PolicyBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiednatural language processingdynamic knowledge graphconvolutional neural networkpetrochemical big datanatural gas industrymissed alarms basedmethod effectively overcomeslow oil pressurecompressor data anomaliesdata anomaliescompressor datamethod startsgas compressionfalse alarmswidely usedtwo typestolerant filteringthree partssecurity monitoringresults showpaper proposesnon startingfault tracingfault diagnosisdrive machinedifferent sourcescauses leadingachieves 100<div><p>Centrifugal compressors are widely used in the oil and natural gas industry for gas compression, reinjection, and transportation. Fault diagnosis and identification of centrifugal compressors are crucial. To promptly monitor abnormal changes in compressor data and trace the causes leading to these data anomalies, this paper proposes a security monitoring and root cause tracing method for compressor data anomalies. Additionally, it presents an intelligent system design method for fault tracing in compressors and localization of faults from different sources. This method starts from petrochemical big data and consists of three parts: fault dynamic knowledge graph construction, instrument data sliding fault-tolerant filtering, and the fusion and reasoning of fault dynamic knowledge graph and instrument data variation monitoring. The results show that this method effectively overcomes the problems of false alarms and missed alarms based on fixed threshold alarm methods, and achieves 100% classification of two types of faults: non starting of the drive machine and low oil pressure by constructing a PCA (Principal Component Analysis)—SPE (Square Prediction Error)—CNN (Convolutional Neural Network) classifier. Combined with dynamic knowledge graph and NLP (Natural Language Processing) inference, it achieves good diagnostic results.</p></div>2025-01-16T18:57:40ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0315917.g022https://figshare.com/articles/figure/CNN_classification_result_/28224597CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/282245972025-01-16T18:57:40Z |
| spellingShingle | CNN classification result. Yuan Wang (14955) Science Policy Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified natural language processing dynamic knowledge graph convolutional neural network petrochemical big data natural gas industry missed alarms based method effectively overcomes low oil pressure compressor data anomalies data anomalies compressor data method starts gas compression false alarms widely used two types tolerant filtering three parts security monitoring results show paper proposes non starting fault tracing fault diagnosis drive machine different sources causes leading achieves 100 |
| status_str | publishedVersion |
| title | CNN classification result. |
| title_full | CNN classification result. |
| title_fullStr | CNN classification result. |
| title_full_unstemmed | CNN classification result. |
| title_short | CNN classification result. |
| title_sort | CNN classification result. |
| topic | Science Policy Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified natural language processing dynamic knowledge graph convolutional neural network petrochemical big data natural gas industry missed alarms based method effectively overcomes low oil pressure compressor data anomalies data anomalies compressor data method starts gas compression false alarms widely used two types tolerant filtering three parts security monitoring results show paper proposes non starting fault tracing fault diagnosis drive machine different sources causes leading achieves 100 |