Standard deviation based acoustic emission signal analysis for detecting valve internal leakage

• Published in: Sensors and actuators. A. Physical. Vol. 283; pp. 340 - 347
• Main Authors: Ye, Guo-Yang, Xu, Ke-Jun, Wu, Wen-Kai
• Format: Journal Article
• Language: English
• Published: Lausanne Elsevier B.V 01.11.2018; Elsevier BV
• Subjects: Acoustic emission; Acoustic emission testing; Emission analysis; Equipment costs; Inlet pressure; Internal leakage; Leak detection; Leakage; Least squares method; Signal analysis; Signal processing; Standard deviation; Valve; Valves; Acoustic emission; Valve; Standard deviation; Internal leakage; Signal analysis
• ISSN: 0924-4247; 1873-3069
• DOI: 10.1016/j.sna.2018.09.048

•The theoretical relationship between standard deviation and leakage rate is established.•The experiments of the valve leakage detection are performed.•The characteristic of valve leakage signal is extracted.•The linear relationship between the standard deviations and the leakage rate is obtained.•The influence of inlet pressure on the AE signal is analyzed. The on-line detection technology of valves based on acoustic emission (AE) can effectively prevent production accidents and reduce the cost of equipment maintenance. However, the existing method for inner leakage detection of the valve cannot effectively detect the small leakage of the valve. Considering the fact that the critical leakage rate of the valve is less than 200 ml/min, the quantitative relationship between the characteristic quantity of the AE signal and the valve leakage rate is studied. The AE mechanism and basic characteristics of valve leakage are analyzed, and the theoretical relationship between standard deviation of AE signal of valve leakage and the leakage rate is determined. The experimental setup of valve leakage detection is built to carry out leakage detection experiments for several valves. The AE signal of valve leakage is processed to obtain its standard deviation. The relationship between the standard deviation and the leakage rate is fitted by least-squares method to establish their mathematical model. On this basis, the influence of inlet pressure on the AE signals of valve leakage is analyzed to lay a foundation for on-line detection of valve leakage under different working conditions.