|
CSME 2024/06
Volume 45 No.3
:
279-292
|
|
Research On the Fault Detection Method of Centrifugal Pump Rotor Misaligned
Yechun Lia, Lingfeng Tanga and Weiwei Duan
aSchool of Mechanical Engineering, Anhui Polytechnic University, Wuhu, Anhui 241000, China.
|
Abstract:
Aimed at the problems of low detection accuracy and poor adaptability of centrifugal pump rotor misalignment fault, the mechanism and characteristics of centrifugal pump rotor misalignment fault were discussed. The overall scheme design of the centrifugal pump rotor misalignment fault diagnosis was completed. The vibration state of rotor stability and misalignment fault was simulated, and the time domain curve, order spectrum, and axis trajectory diagram under different states were obtained. A frequency band division method based on improved empirical wavelet and scale-space theory, and a centrifugal pump rotor misalignment fault diagnosis and identification method based on SVM support vector machine optimized by the genetic algorithm are proposed. A centrifugal pump rotor vibration simulation test bench was built to simulate the centrifugal pump rotor misalignment fault. The proposed fault diagnosis method was used for fault identification. The experimental results show that the recognition accuracy of the method was 93 %, which can effectively diagnose faults and optimize fault identification. The feasibility and superiority of the method for centrifugal pump rotor misalignment fault identification were verified, and the research on the centrifugal pump rotor misalignment fault detection method was completed.
|
|
Keywords: centrifugal pump; rotor misalignment; fault diagnosis; SVM support vector machine
|
|
|
*Corresponding author; e-mail:
|
©
2024
CSME , ISSN 0257-9731
|