An Adaptive Autogram Approach Based on a CFAR Detector for Incipient Cavitation Detection

Chu, Ning; Wang, Linlin; Yu, Liang; He, Chunlin; Cao, Linlin; Huang, Bin; Wu, Dazhuan

doi:10.3390/s20082303

Cited by 4 publications

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The finding demonstrates that the method improves the accuracy of cavitation recognition in the noise environment [63]. In addition, the artificial immune algorithm proposed by Matloobi and Riahi [64], the transfer learning and bispectral analysis presented by Hajnayeb and Qin [65], and the adaptive cavitation detection method based on the constant false alarm rate (CFAR) criterion suggested by Chu et al [66] can have an accurate detection of cavitation faults, and the accuracy is higher than that of multilayer artificial neural networks and nonlinear support vector machines under the same conditions.…”

Section: Detection Efficiencymentioning

confidence: 87%

A Review of Pump Cavitation Fault Detection Methods Based on Different Signals

et al. 2023

View full text Add to dashboard Cite

As one of the research hotspots in the field of pumps, cavitation detection plays an important role in equipment maintenance and cost-saving. Based on this, this paper analyzes detection methods of cavitation faults based on different signals, including vibration signals, acoustic emission signals, noise signals, and pressure pulsation signals. First, the principle of each detection method is introduced. Then, the research status of the four detection methods is summarized from the aspects of cavitation-induced signal characteristics, signal processing methods, feature extraction, intelligent algorithm identification of cavitation state, detection efficiency, and measurement point distribution position. Among these methods, we focus on the most widely used one, the vibration method. The advantages and disadvantages of various detection methods are analyzed and proposed: acoustic methods including noise and acoustic emission can detect early cavitation very well; the vibration method is usually chosen first due to its universality; the anti-interference ability of the pressure pulsation method is relatively strong. Finally, the development trend of detecting cavitation faults based on signals is given: continue to optimize the existing detection methods; intelligent algorithms such as reinforcement learning and deep reinforcement learning will be gradually integrated into the field of cavitation status identification in the future; detection systems still need to be further improved to accommodate different types of pumps; advanced sensing devices combined with advanced signal processing techniques are one of the effective means to detect cavitation in a timely manner; draw on other fault detection methods such as bearing faults and motor faults.

show abstract