Two-Phase Air–Water Slug Flow Measurement in Horizontal Pipe Using Conductance Probes and Neural Network

Fan, Shiwei; Yan, T.

doi:10.1109/tim.2013.2280485

Cited by 40 publications

(14 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A continuous wave ultrasound Doppler sensor employed in this study, has recently been implemented for investigating the velocity characteristics of slug body and film in a two-phase gas-liquid slug flow. The results showed velocity characteristics of the slug flow obtained is in good agreement with other experimental methods (Fan et al 2014). The present approach is by recording and processing of ultrasonic Doppler signals on the flow and then features are extracted using both power spectral density and wavelet transforms methods.…”

Section: Introductionsupporting

confidence: 88%

“…Most frequently used training model in classification problems in the back propagation (BP) which is adopted for this investigation and in other works (Fan & Yan 2014;Blaney & Yeung 2008;Arubi 2011). The MLPNN has properties such as the abilities to learn and transform fewer training set requirements and fast processing.…”

Section: Flow Regime Classification Networkmentioning

confidence: 99%

See 1 more Smart Citation

Non-invasive classification of gas–liquid two-phase horizontal flow regimes using an ultrasonic Doppler sensor and a neural network

Abbagoni¹,

Yeung²

2016

Meas. Sci. Technol.

View full text Add to dashboard Cite

The identification of flow pattern is key issue in multiphase flow which encountered in the petrochemical industry. Gas-liquid two-phase flow is difficult to identify the gas-liquid flow regimes objectively. This paper presents a feasibility of a clamp-on instrument for objective flow regime classification of two-phase flow using an ultrasonic Doppler sensor and artificial neural network. It is on recording and processing of the ultrasonic signals reflected from the two-phase flow. Experimental data obtained on a horizontal test rig with total pipe length of 21 m long and 5.08 cm internal diameter carrying air-water two-phase flow under slug, elongated bubble, stratified-wavy and, stratified flow regimes. Multilayer Perceptron Neural Networks (MLPNNs) used for developing the classification model. The classifier requires features as input which is representative of the signals. Ultrasound signal features extracted by applying both power spectral density (PSD) and discrete wavelet transforms (DWT) methods to the flow signals. A classification scheme of "1-of-C coding method for classification" was adopted to classify features extracted into one of four flow regime categories. To improve the performance of the flow regime classifier network, a second level neural network was incorporated by using output of a first level networks features as input features. Addition of the two network models provided a combined neural network models which has achieved higher accuracy than single neural network models. Classification accuracies evaluated in the form of both the PSD and DWT features. The success rates of the two models are: (1) using PSD features, the classifier missed three datasets out of 24 test datasets of the classification and scored 87.5% accuracy. (2) With the DWT features, the network misclassified only one data point and it was able to classify the flow patterns up to 95.8% accuracy. This approach has demonstrated success of a clamp-on ultrasound sensor for flow regime classification and it would be possible in industry practice. It is considerably more promising than other techniques as it uses of non-invasive and nonradioactive sensor.

show abstract

Section: Introductionsupporting

confidence: 88%

Section: Flow Regime Classification Networkmentioning

confidence: 99%

Non-invasive classification of gas–liquid two-phase horizontal flow regimes using an ultrasonic Doppler sensor and a neural network

Abbagoni¹,

Yeung²

2016

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Two pairs of electrodes were used, each having a width of 6 mm and a distance of 270 mm between each electrode pair. The probe is calibrated offline for stratified flow based on the technique employed by Fan and Yan [18]. During calibration, the uncertainty of the probe was found to be ±2% from direct measurements.…”

Section: B Conductance Probesmentioning

confidence: 99%

“…Conductance is an appropriate intrinsic flow regime indicator [8], [16], [18], [31] since it correlates with the fraction of the liquid phase of the flow. Moreover, the instrumentation for collecting conductance data is simple, non-invasive, relatively cheap, and easy to manufacture.…”

Section: Introductionmentioning

confidence: 99%

Development of a Real-Time Objective Gas–Liquid Flow Regime Identifier Using Kernel Methods

Eyo

Pilario

Lao

et al. 2021

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Currently, flow regime identification for closed channels have mainly been direct subjective methods. This presents a challenge when dealing with opaque test sections of the pipe or at gas-liquid flow rates where unclear regime transitions occur. In this paper, we develop a novel real-time objective flow regime identification tool using conductance data and kernel methods. Our experiments involve a flush mounted conductance probe that collects voltage signals across a closed channel. The channel geometry is a horizontal annulus, which is commonly found in many industries. Eight distinct flow regimes were observed at selected gas-liquid flow rate settings. An objective flow regime identifier was then trained by learning a mapping between the probability density function (PDF) of the voltage signals and the observed flow regimes via kernel principal components analysis (KPCA) and multi-class Support Vector Machine (SVM). The objective identifier was then applied in real-time by processing a moving time-window of voltage signals. Our approach has: (a) achieved more than 90% accuracy against visual observations by an expert for static test data; (b) successfully visualized conductance data in 2-dimensional space using virtual flow regime maps, which are useful for tracking flow regime transitions; and, (c) introduced an efficient real-time automatic flow regime identifier, with only conductance data as inputs.

show abstract

“…This problematic outcome is commonly referred to as the vanishing gradient problem that results in reduced network performance on standard neural network models [12]. In the proposed model, Bayesian regulation is used as a training algorithm to adjust the parameters of the network so as to move the equilibrium in a way that will result in an output that is close as possible to the target output [6].…”

Section: Recurrent Networkmentioning

confidence: 99%

Hybridization of Ensemble Kalman Filter and Non-linear Auto-regressive Neural Network for Financial Forecasting

Abdulkadir

Yong

Marimuthu

et al. 2014

Mining Intelligence and Knowledge Exploration

View full text Add to dashboard Cite

Abstract. Financial data is characterized as non-linear, chaotic in nature and volatile thus making the process of forecasting cumbersome. Therefore, a successful forecasting model must be able to capture longterm dependencies from the past chaotic data. In this study, a novel hybrid model, called UKF-NARX, consists of unscented kalman filter and non-linear auto-regressive network with exogenous input trained with bayesian regulation algorithm is modelled for chaotic financial forecasting. The proposed hybrid model is compared with commonly used Elman-NARX and static forecasting model employed by financial analysts. Experimental results on Bursa Malaysia KLCI data show that the proposed hybrid model outperforms the other two commonly used models.

show abstract

Two-Phase Air–Water Slug Flow Measurement in Horizontal Pipe Using Conductance Probes and Neural Network

Cited by 40 publications

References 32 publications

Non-invasive classification of gas–liquid two-phase horizontal flow regimes using an ultrasonic Doppler sensor and a neural network

Non-invasive classification of gas–liquid two-phase horizontal flow regimes using an ultrasonic Doppler sensor and a neural network

Development of a Real-Time Objective Gas–Liquid Flow Regime Identifier Using Kernel Methods

Hybridization of Ensemble Kalman Filter and Non-linear Auto-regressive Neural Network for Financial Forecasting

Contact Info

Product

Resources

About