Elsevier

Measurement

Volume 43, Issue 2, February 2010, Pages 274-282
Measurement

Investigation of the relationship between internal fluid leakage through a valve and the acoustic emission generated from the leakage

https://doi.org/10.1016/j.measurement.2009.10.005Get rights and content

Abstract

The detection of acoustic emission (AE) signals produced by liquid and gas leakage through valves can be related directly to the qualitative leakage rate. This allows for cost estimation of losses in processes for several industries. However, to find out the relationship between qualitative leakage rate and AE signal large amounts of experimental data is needed. This paper presents a theoretical investigation of the acoustic emission to detect the internal leakage rate through a valve and experimental validation. The AE signals generated by internal liquid and gas leakage through valves were characterised. The effect of the influenced factors of leakage rates, inlet pressure levels, valve sizes and valve types, on AE parameter, AERMS, were studied and explained. The results of theoretical and experimental showed that AE signal power computed from the power spectral density (PSD) correlated well with influenced factors of leakage rates. Finally, a novel and inexpensive AE instrument has been invented for predicting qualitative leakage rate using a micro processor and derived relationship.

Introduction

The acoustic emission (AE) technique has been applied to detect internal valve leakage and it is widely used in the oil, gas, chemical and petrochemical industries [1]. This technique stresses determination of the feasibility of a leak detection system that could be used on valves in situ [2]. Various plants require leakage rate knowledge prior to decisions for repair or changing of damaged valves. However, it is difficult and inconvenient to obtain data from thousands of valves within a leakage detection system. From literature, implementation of valves for transforming the gas phase and liquid phase has commonly been used in several industries [3].

Much research has been completed to investigate and estimate leakage rate through acoustic emission [4], [5], [6], [7], [8]. Multiple regression fitting models for multi-independent variables (such as inlet pressure level and leakage rate) were developed in quantitative models. However, one of the most common problems has been training of the algorithm and information required to account for valve leakage rates. This affected many process variables, for instance, inlet pressure level, valve size and type.

Recently, the theoretical relationship between acoustic emission and internal valve leakage rate was analysed to estimate the rate of gas leakage through a valve [9]. The investigation showed a good correlation between high frequency sound powers, Ps, theoretically derived by turbulence jets behind the valve leakage hole and the calculated AE signal powers, AERMS2. A Pearson correlation coefficient (r) of Ps and AERMS2 for each valve size and inlet pressure level was calculated. The correlation coefficients were high, especially at low inlet pressure levels (100 kPa). The value was r = 0.881, 0.805 and 0.947 with valve sizes of 25.4, 50.8 and 76.2 mm, respectively.

As mentioned above, liquid transformation through a valve has frequently been used. Consequently, the main objective of the present paper was to further analyse and implement a theoretical relationship to estimate valve leakage rate of the liquid phase through the empirical variables, valve size and inlet pressure, by using the AE technique. In additional, the effect of valve types and the characteristics of the AE signal in frequency domain were studied for both liquid and gas phases. The effect of various variables was empirically investigated to validate the derived theoretical formula. Finally, an AE instrument based on the theoretical model was created.

Section snippets

AE signal power

Acoustic emission was defined as the transient elastic waves generated by the rapid release of energy from localised sources. It has been found that AE signals can be produced by fluid leakage [10], [11]. In the case of internal leakage in liquid valves, the generation of a pseudo source produced AE signals, which were largely attributed to the turbulence induced by the internal leakage [12].

The AE signal from valve leakage was of a continuous type in time domain. To extract useful information

Experimental set-up

The experiment was set-up as shown in Fig. 2. It consisted of two parts; valve leakage system and AE measurement system. For the valve leakage system, two types of ball and globe valves were selected as the test subjects in the experiment. Ball and globe valve are widely used in industry. Different sizes of soft-valve-seat, 25.4, 50.8 and 76.2 mm inside diameters, were made to produce leakage. The leakage ranged from 1.0 to 6.0 l/min due to incomplete closure. A piston pump was used to generate

Characteristics of AE signal from gas and liquid valve leakage

The typical characteristic of the AE leakage signals of a compressed air valve (representative of gas phase) and fresh water valve (representative of liquid phase) in frequency domain are shown in Fig. 3a and b. The results were shown as the relationship between power spectrum density (PSD) and leakage rate in frequency domain over a frequency range of 0–1000 kHz. Experimental equipment and conditions were: inlet pressure at 300 kPa and ball valve size of 25.4 mm. The magnitude of the PSD,

Novel instrument to detect valve leakage rate

Based on the investigated relationship, an AE instrument used to measure both liquid and gas leakage phases was invented. The instrument was inexpensive owing to the implementation of micro processing instead of data acquisition. The AE sensor still can be piezoelectric (PZT). It was easy to use as a result of the fluid parameters and could be directly installed into the instrument. Leakage rate can be displayed on the panel of the instrument quickly. Consequently, it was not necessary to use

Conclusions

This paper presented an investigation of the relationship between the liquid through valve leakage rate and Acoustic Emission. It was confirmed that the sound power, Ps, theoretically derived was in good agreement with the AE signal power, AERMS2. This parameter extracted during water valve leakage can be used to predict the actual leakage rate qualitatively. The effects of inlet pressure, leakage rate, valve size and valve type on AERMS were determined. In addition, it was found that the

Acknowledgements

The authors would like to acknowledge the support from King Mongkut’s University of Technology Thonburi and would like to thank researchers from ANDT for their assistance.

References (18)

There are more references available in the full text version of this article.

Cited by (98)

View all citing articles on Scopus
View full text