Cavitation detection of butterfly valve using support vector machines

Yang, Bo‐Suk; Hwang, Won-Woo; Ko, Myung-Han; Lee, Soo-Jong

doi:10.1016/j.jsv.2004.10.033

Cited by 68 publications

(31 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The selection of relevant radial basis function (RBF) kernel parameters was carried out through iteration [1] for more accurate classification of healthy and faulty compressors. In a similar application, SVM methods were applied to reciprocating compressor butterfly valves to classify cavitation faults [3]. Comparable research was performed on reciprocating compressor valves to classify faults through vibration signals alone.…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis of reciprocating compressors using revelance vector machines with a genetic algorithm based on vibration data

Ahmed

Smith

et al. 2014

2014 20th International Conference on Automation and Computing

View full text Add to dashboard Cite

Abstract-This paper focuses on the development of an advanced fault classifier for monitoring reciprocating compressors (RC) based on vibration signals. Many feature parameters can be used for fault diagnosis, here the classifier is developed based on a relevance vector machine (RVM) which is optimized with genetic algorithms (GA) so determining a more effective subset of the parameters. Both a one-against-one scheme based RVM and a multiclass multi-kernel relevance vector machine (mRVM) have been evaluated to identify a more effective method for implementing the multiclass fault classification for the compressor. The accuracy of both techniques is discussed correspondingly to determine an optimal fault classifier which can correlate with the physical mechanisms underlying the features. The results show that the models perform well, the classification accuracy rate being up to 97% for both algorithms.

show abstract

Section: Introductionmentioning

confidence: 99%

Fault diagnosis of reciprocating compressors using revelance vector machines with a genetic algorithm based on vibration data

Ahmed

Smith

et al. 2014

2014 20th International Conference on Automation and Computing

View full text Add to dashboard Cite

show abstract

“…The above function can be minimized using Lagrange multipliers [23]. Lagrange multipliers are used to solve optimization problems with linear equality and inequality constraints.…”

Section: B Support Vector Machinesmentioning

confidence: 99%

“…7(a) shows the OES data from both steps. In the plasma, SF 6 dissociates into atomic fluorine and heavy SF + 5 ions [25], as shown in (23). The resulting plasma leads to silicon etching as shown in chemical reaction (24)…”

Section: Pls-svm Epd Modelmentioning

confidence: 99%

Endpoint Detection in Low Open Area TSV Fabrication Using Optical Emission Spectroscopy

Thadesar

Dembla

et al. 2014

IEEE Trans. Compon., Packag. Manufact. Technol.

View full text Add to dashboard Cite

Through-silicon via (TSV) technology is a key enabler for 3-D and 2.5-D integration, which provides low-power and high-bandwidth chip-to-chip communication. During TSV fabrication, over-etching may cause notching at the base of the TSVs, resulting in TSV diameter variations. Endpoint detection (EPD) techniques are critical for controlling TSV diameter, and detecting the endpoint for low open areas presents a serious challenge to process engineers. In this paper, a hybrid partial least squares-support vector machine model for optical emission spectroscopy data are successfully demonstrated for an EPD of low open area TSVs. Accurate EPD results are shown for 120, 80, and 25 µm diameter TSVs. Index Terms-3-D integration, endpoint detection (EPD), optical emission spectroscopy (OES), through-silicon via (TSV).

show abstract

“…Acoustic signatures have been used to identify cavitation from butterfly valves [1,2], but the understanding of the cavitation field could be improved through improved visualization. Chern et a!.…”

Section: Introductionmentioning

confidence: 99%

Cavitation From a Butterfly Valve: Comparing 3D Simulations to 3D X-Ray Computed Tomography Flow Visualization

Brett¹,

Riveland²,

Jensen

et al. 2011

ASME-JSME-KSME 2011 Joint Fluids Engineering Conference: Volume 2, Fora

View full text Add to dashboard Cite

Flow control valves may experience localized cavitation when the local static pressure drops to the liquid vapor pressure. Localized damage to the valve and surrounding area can occur when the vapor cavity collapses. Valve designs that reduce cavitation are based on empirical evidence and accumulated experience, but there are still considerable cavitation problems in industry. Valve designers may use computational fluid dynamics (CFD) to simulate cavitation in flow control valves, but model validation is challenging because there are limited data of local cavitation from the valve surface. Typically, the intensity of cavitation in a control valve is inferred from measurements of observable side effects of cavitation such as valve noise, vibration, or damage to the valve assembly. Such an indirect approach to characterizing cavitation yields little information about the location, degree, and extent of the cavitation flow field that can be used in CFD validation studies. This study uses 3D X-ray computed tomography (CT) imaging to visualize cavitation from a 5.1 cm diameter butterfly valve and compares the resulting vapor cloud to that predicted by CFD simulations. Qualitative comparisons reveal that the resulting cavitation structures are captured by the simulations when a small amount of non-condensable gas is introduced into the fluid and the simulations are completed in a transient mode. Proceedings of the ASME-JSME-KSME ABSTRACT Flow control valves may experience localized cavitation when the local static pressure drops to the liquid vapor pressure. Localized damage to the valve and surrounding area can occur when the vapor cavity collapses. Valve designs that reduce cavitation are based on empirical evidence and accumulated experience, but there are still considerable cavitation problems in industry. Valve designers may use computational fluid dynamics (CFD) to simulate cavitation in flow control valves, but model validation is challenging because there are limited data of local cavitation from the valve surface. Typically, the intensity of cavitation in a control valve is inferred from measurements of observable side effects of cavitation such as valve noise, vibration, or damage to the valve assembly. Such an indirect approach to characterizing cavitation yields little information about the location, degree, and extent of the cavitation flow field that can be used in CFD validation studies. This study uses 3D X-ray computed tomography (CT) imaging to visualize cavitation from a 5.1 em diameter butterfly valve and compares the resulting vapor cloud to that predicted by CFD simulations. Qualitative comparisons reveal that the resulting cavitation structures are captured by the simulations when a small amount of noncondensable gas is introduced into the fluid and the simulations are completed in a transient mode.

show abstract

Cavitation detection of butterfly valve using support vector machines

Cited by 68 publications

References 10 publications

Fault diagnosis of reciprocating compressors using revelance vector machines with a genetic algorithm based on vibration data

Fault diagnosis of reciprocating compressors using revelance vector machines with a genetic algorithm based on vibration data

Endpoint Detection in Low Open Area TSV Fabrication Using Optical Emission Spectroscopy

Cavitation From a Butterfly Valve: Comparing 3D Simulations to 3D X-Ray Computed Tomography Flow Visualization

Contact Info

Product

Resources

About