| With the development of the national economy, in all walks of life, the valve is widely used. But as the growth of the service life of, caused by aging, corrosion valve leakage accident to happen from time to time. As a result, the valve leak detection technology research in industrial production is of important theoretical and realistic significance. Now valve detection technologies mainly include air sampling method, temperature detection method and acoustic emission method. But these methods also have sorts of defects. Such as, air sampling method cannot be real-time detection. Temperature detection method and application scope is very narrow. Acoustic emission testing instrument is expensive.This paper presents a method of valve leakage detection using dynamic pressure sensor. When internal fluid leakage occurs in the valve, the pressure difference between upstream and downstream makes the medium from high voltage side to the low side. Medium impact pipe wall will produce pressure fluctuation.For the valve leakage detection, this paper builds a complete test-system. It includes the conveying pipeline for detection of valve, medium, multi-channel parallel data acquisition terminal based on ARM microprocessor sampling for dynamic pressure sensor and transmitter. This paper makes a software which run on the ARM processor data acquisition terminal. It includes experimental data collection, through the TCP/IP protocol data transmission and other functions. And we propose the preparation of running on the monitoring server of the valve leakage testing procedures. It includes communications, and data terminal data storage, data analysis and other functions.Laboratory conditions, the valve leakage simulation experiment was carried out. By analyzing and comparing the valve downstream of the acoustic sensor output signal under different state characteristics, puts forward the acoustic signal feature extraction method based on wavelet package entropy methods for identifying and valve leakage. This paper explores the valve leakage detection in laboratory environment, the minimum detectable leakage, as well as the continuous operating condition of false positives and omission of indicators. |