NLPCA as a diagnostic tool for control valve stiction

Zabiri, Haslinda; Ramasamy, M.

doi:10.1016/j.jprocont.2009.04.010

Cited by 24 publications

(23 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Type Loop Applicability Horch [40] cross-correlation no LC Horch [48] statistics all type Kano et al [29] MV(OP) patterns all (better FC) Yamashita [41] MV(OP) patterns only FC Scali and Ghelardoni [42] MV(OP) patterns only FC Daneshwar and Noh [49] MV(OP) patterns only FC Yamashita [50] statistics only LC Farenzena and Trierweiler [51] waveform shape only LC Choudhury et al [43] NL detection all Thornhill [44] NL detection all Mohammad and Huang [52] NL detection all Rengaswamy et al [53] waveform shape all Srinivasan et al [45] waveform shape all Rossi & Scali [46] waveform shape all He et al [30] waveform shape all Singhal and Salsbury [47] waveform shape no LC Hägglund [54] waveform shape all Zabiri and Ramasamy [55] waveform shape all Ahmed et al [56] waveform shape all Ahammad and Choudhury [57] harmonics based all Zakharov et al [58] algorithms combination all Afterwards, following this line, Cuadros et al [60] proposed an improved algorithm that fits an ellipse just using the most significant points of OP and PV signals. Despite this method being applicable only to flow control loops, the procedure seems to have more precision than the previous approaches.…”

Section: Methods Featuresmentioning

confidence: 99%

“…Zabiri and Ramasamy [55] developed a method that calculates an index based on nonlinear principal components analysis (NLPCA) using the distinctive shapes of the signals caused by stiction and other sources. Together with its coefficient of determination, the index quantifies the degree of nonlinearity and determines the presence of stiction.…”

Section: Farenzena and Trierweilermentioning

confidence: 99%

See 1 more Smart Citation

RETRACTED: Review on Valve Stiction. Part I: From Modeling to Smart Diagnosis

Capaci

Scali

2015

Processes

View full text Add to dashboard Cite

Abstract:Valve stiction is indicated as one of the main problems affecting control loop performance and then product quality. Therefore, it is important to detect this phenomenon as early as possible, distinguish it from other causes, and suggest the correct action to the operator in order to fix it. It is also very desirable to give an estimate of stiction amount, in order to be able to follow its evolution in time to allow the scheduling of valve maintenance or different operations, if necessary. This paper, in two parts, is a review of the state of the art about the phenomenon of stiction from its basic characterization to smart diagnosis, including modeling, detection techniques, quantification, compensation and a description of commercial software packages. In particular, Part I of the study analyzes the most significant works appearing in the recent literature, pointing out analogies and differences among various techniques, showing more appealing features and possible points of weakness. The review also includes an illustration of the main features of performance monitoring systems proposed by major software houses. Finally, the paper gives indications on future research trends and potential advantages for loop diagnosis when additional measurements are available, as in newly designed plants with valve positioners and smart instrumentation. In Part II, performance of some well-established methods for stiction quantification are compared by applications to different industrial datasets. OPEN ACCESSProcesses 2015, 3 423

show abstract

Section: Methods Featuresmentioning

confidence: 99%

Section: Farenzena and Trierweilermentioning

confidence: 99%

RETRACTED: Review on Valve Stiction. Part I: From Modeling to Smart Diagnosis

Capaci

Scali

2015

Processes

View full text Add to dashboard Cite

show abstract

“…The time-series data segments are classified as increasing, decreasing or steady. Then, a stiction index is Horch and Isaksson (2001) Histogram All type Kano et al (2004) Limit cycle pattern All type Yamashita (2006a) Limit cycle pattern Only FC Scali and Ghelardoni (2008) Limit cycle pattern Only FC Daneshwar and Noh (2015) Limit cycle pattern Only FC Bra ´sio et al (2015) Limit cycle pattern Only LC Yamashita (2006b) Limit cycle pattern + Statistics Only LC Rengaswamy et al (2001) Waveform shape + ANN All type Rossi and Scali (2005) Waveform shape All type Srinivasan et al (2005a) Waveform shape All type Singhal and Salsbury (2005) Waveform shape All type except LC He et al (2007) Waveform shape All type Zabiri and Ramasamy (2009) Waveform shape + Nonlinearity detection All type Teh et al (2018) Waveform shape + Nonlinearity detection All type Ahmed et al (2009) Waveform shape All type Stockmann et al (2009) Waveform shape All type except LC Ha ¨gglund (2011) Waveform shape All type Farenzena and Trierweiler (2012b) Waveform shape Only LC Dambros et al (2016) Waveform shape All type Thornhill (2005) Nonlinearity detection All type Choudhury et al (2006) Nonlinearity detection All type Aftab et al (2016Aftab et al ( , 2017 Nonlinearity detection All type Farenzena and Trierweiler (2009) Machine learning (ANN) All type Venceslau et al (2012) Machine learning (ANN) All type Sharma et al (2017) Machine learning (ANN) Only FC Amiruddin et al (2019) Machine learning (ANN) All type Dambros et al (2019) Machine learning (ANN) All type Napoli et al (2019) Machine learning (CNN) Only FC Kamaruddin et al (2020) Machine learning (CNN) All type Henry et al (2020) Machine learning (CNN) All type Choudhury et al (2007) Controller gain change method All type…”

Section: Limit Cycle Pattern-based Methodsmentioning

confidence: 99%

Comparison of current state of control valve stiction detection and quantification techniques

Akavalappil

Radhakrishnan

2021

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

The process control and automation system is one of the significant features in any process plant. The control valve is considered a fundamental part of the process control system. It regulates the fluid flow by changing the size of the flow passage as directed by the controller. If properly maintained, the control valves can help to reduce process variability and improve product quality, which in turn increases the overall efficiency of the plant. But control valve stiction is an enduring concern within process industries which eventually affects the efficiency of plant operation. Hence, it is extremely essential to detect and quantify stiction at the earliest opportunity, to determine the correct maintenance measures. Numerous methods related to stiction detection and estimation have been carried out so far by researchers. A literature study of the latest publications in stiction detection and quantification discloses that research in the conventional methods of stiction detection and quantification such as limit cycle pattern-based, waveform shape-based, nonlinearity-based, and model-based methods is gradually declining. At the same time, most stiction detection/quantification methods reported in the recent literature are based on artificial neural network and convolutional neural network. This implies that future stiction detection and quantification research will be focused towards machine learning (ML) algorithms. Taking this into consideration, this paper aims to provide a literature review of stiction detection and quantification methods by focusing on the latest research ideas and innovative approaches. This review also aims to compare recent techniques based on ML algorithms with conventional methods by pointing out the relationship among different methods, differences, and possible points of weakness.

show abstract

“…In this study, the Choudhury stiction model developed by [16] was applied. The simulated data were generated using a simple single input single output (SISO) first order transfer function of the feedback control system adapted from [5]. A total number of 10,000 samples of PV and OP data were collected from each case at a sampling rate of 1 s. The initial 500 data points were discarded to ensure the time series had stabilized; leaving 9,500 samples for training.…”

Section: Generation Of Simulated Datamentioning

confidence: 99%

“…Most of the detection methods proposed are based on the relationship between process variables (PV) and controller outputs (OP) due to the difficulties in observing manipulated variable (MV). Various detection methods based on neural networks have been proposed, such as NLPCA [5], ANN [6], NLPCA-AC [7], and SDN [8]. These methods consider the time series input directly as 1D signal.…”

Section: Introductionmentioning

confidence: 99%

Control valve stiction detection by use of AlexNet and transfer learning

2021

Self Cite

View full text Add to dashboard Cite

Control valve stiction is a common problem faced by the process industries, which can have a strong adverse effect on the profitable operation of plants. Although various stiction detection methods based on neural networks have been proposed, few of these studies have considered the performance of stiction detection based on the use of 2D representations of the process signals. In this paper, such an approach is proposed, based on the use of a pretrained convolutional neural network, AlexNet. The proposed convolutional neural network stiction detection (CNN-SD) method showed highly satisfactory performance, which can be further applied on real industrial data.

show abstract

NLPCA as a diagnostic tool for control valve stiction

Cited by 24 publications

References 10 publications

RETRACTED: Review on Valve Stiction. Part I: From Modeling to Smart Diagnosis

RETRACTED: Review on Valve Stiction. Part I: From Modeling to Smart Diagnosis

Comparison of current state of control valve stiction detection and quantification techniques

Control valve stiction detection by use of AlexNet and transfer learning

Contact Info

Product

Resources

About