Izvorni znanstveni članak

https://doi.org/10.21278/TOF.462032721

A Dimensionless Immune Intelligent Fault Diagnosis System for Rotating Machinery

Longqiu Shao orcid.org/0000-0003-1644-9059 ; School of Automation, Guangdong University of Petrochemical Technology, Maoming, China
Qinghua Zhang ; School of Automation, Guangdong University of Petrochemical Technology, Maoming, China
Gaowei Lei ; School of Automation, Guangdong University of Petrochemical Technology, Maoming, China
Naiquan Su ; School of Automation, Guangdong University of Petrochemical Technology, Maoming, China
Penghui Yuan ; School of Automation, Guangdong University of Petrochemical Technology, Maoming, China

Sažetak

Aiming at the shortcomings of the traditional frequency domain analysis method, such as failure to find early faults, the misjudgement and omission of fault types, and failure to diagnose complex faults, a new approach is developed, which is different from the existing technical route in the field of fault diagnosis, by closely following real-time online, intelligent and accurate requirements in the field of monitoring and fault diagnosis of large rotating machinery. Combining immune mechanism, dimensionless index, support vector machine and other artificial intelligence technologies, linked with the particularity of fault diagnosis problems, a fault diagnosis classification algorithm based on memory sequence is proposed, and an intelligent fault diagnosis system based on a dimensionless immune detector and support vector machine was developed. Finally, the system was applied to a compressor unit in a petrochemical enterprise and good results were achieved.

Ključne riječi

rotating machinery; dimensionless index; on-line monitoring; fault diagnosis

Hrčak ID:

279219

URI

https://hrcak.srce.hr/279219

Datum izdavanja:

15.6.2022.

Posjeta: 1.348 *