基于声发射的阀门泄漏检测方法的研究

发布时间：2019-03-03 12:10

【摘要】：在阀门生产和运行中,泄漏检测是其最重要的工作环节之一。针对目前国内很多厂家普遍采用效率低下的气泡检漏法的现状,本文旨在研究能快速测试阀门泄漏率的声发射检测方法。通过理论和实验分析,得出声发射泄漏检测方法是可行的,能对阀门微小泄漏进行定性分析,为实现阀门自动化泄漏检测提供了理论基础。首先,在本文中分析和对比了各种常用的泄漏检测方法的原理和特点,得出了在阀门泄漏率高于10~(-4) Pa·m~3/s下,声发射检测方法更易实现其自动化泄漏检测;并从理论上分析了阀门泄漏射流的速度场和声场,以及通过FLUENT仿真分析,得出了阀门输入压力和泄漏孔大小均对射流速度场分布产生很大的影响,并进而影响射流声场的分布。其次,基于阀门泄漏产生声发射信号的原理,建立了可用于测试阀门泄漏率高于10~(-3) Pa·m~3/s的微小泄漏状态的声发射检测系统,并采用了均匀气泡法对这一微小泄漏进行标定。大量实验的结果显示,该标定方法能准确地测试出阀门的泄漏率。针对阀门泄漏产生的声发射信号在检测中易受环境噪声和电气噪声干扰的问题,设计了信号调理电路,以及采用了非抽样小波变换信号处理等手段来降低各种噪声的影响。再次,通过编制LABVIEW程序,对采集到的声发射信号分别进行了时域分析、谱分析以及小波包分析,得出了均方根、FFT功率谱、FFT功率谱密度以及各频段的能量经归一化处理后的大小等特征量都与阀门泄漏率有很大的关系,可以用它们来表征阀门泄漏的状态。最后,经过大量实验统计,得出了阀门泄漏率与输入压力成指数函数关系,但输入压力只是影响阀门泄漏率的因素之一,并不能利用同一指数函数来表示它们之间的关系;以及得出了泄漏检测位置和输入压力都会对声发射信号的强度、频谱分布产生影响等结论。
[Abstract]:Leakage detection is one of the most important work links in valve production and operation. In view of the present situation of low efficiency bubble leak detection method adopted by many domestic manufacturers, this paper aims to study the acoustic emission detection method which can quickly test the leakage rate of valves. Through theoretical and experimental analysis, it is concluded that the method of acoustic emission leakage detection is feasible and can qualitatively analyze the micro leakage of valves, which provides a theoretical basis for realizing automatic leakage detection of valves. Firstly, the principle and characteristics of various leakage detection methods are analyzed and compared in this paper. It is concluded that when the leakage rate of valve is higher than 10 ~ (- 4) Pa 路m ~ 3 s, the acoustic emission detection method is easier to realize automatic leakage detection. The velocity field and sound field of valve leakage jet are analyzed theoretically, and through FLUENT simulation analysis, it is concluded that the valve input pressure and the size of leakage hole have a great influence on the distribution of jet velocity field, and then affect the distribution of jet sound field. Secondly, based on the principle of acoustic emission signal produced by valve leakage, an acoustic emission detection system which can be used to test the micro leakage state of valve with leakage rate higher than 10 ~ (- 3) Pa 路m ~ 3 is established. The uniform bubble method was used to calibrate the small leakage. A large number of experimental results show that the calibration method can accurately test the leakage rate of the valve. In order to solve the problem that acoustic emission signal produced by valve leakage is easily disturbed by environmental noise and electrical noise, the signal conditioning circuit is designed, and the non-sampling wavelet transform signal processing method is used to reduce the influence of all kinds of noise. Thirdly, through the programming of LABVIEW program, the time domain analysis, spectrum analysis and wavelet packet analysis of the acquired acoustic emission signals are carried out, and the root-mean-square and FFT power spectrum are obtained. The power spectral density of FFT and the normalized energy of each frequency band have a great relationship with the leakage rate of the valve. They can be used to characterize the leakage state of the valve. Finally, through a large number of experimental statistics, the relationship between valve leakage rate and input pressure is obtained as an exponential function, but the input pressure is only one of the factors affecting valve leakage rate, and the relationship between them can not be expressed by the same exponential function. It is concluded that the location of leakage detection and the input pressure will affect the intensity and spectrum distribution of AE signal.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH134

【参考文献】