Application of Multivariate Statistics for Efficient Alarm Generation

Kondaveeti, Sandeep R.; Shah, Sirish L.; Izadi, Iman

doi:10.3182/20090630-4-es-2003.00109

Cited by 23 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For IDV 3, the SPE control chart detects the fault instantly (Figure 7A), while the T 2 control chart takes a little longer and reports an abnormal situation at the 162nd sample (Figure 7B). As this fault is detected earlier by the SPE control chart, the SPE contribution plot is generated (Figure 7C), and it can be seen that XMV (9), XMEAS (21), XMEAS (18), XMEAS (11), XMEAS (16), and XMV(1) are the six variables that have the highest contributions to this fault. It is noteworthy to mention that the 41 measured variables are sequentially placed first in the contribution plots as XMEAS(1)−XMEAS(41), followed by the 11 manipulated variables that are placed as XMV(1)−XMV (11).…”

Section: Industrial and Engineering Chemistry Researchmentioning

confidence: 99%

“…Moreover, these charts are not adaptable to a change in the operating conditions; this may result in a false detection or miss alarm. 11 Multivariate statistical process monitoring (MSPM) tools reduced some complexities associated with the univariate monitoring tools and became more popular for FDD in recent decades. 7,12,13 According to a bibliometric review by Alauddin, Khan, Imtiaz, and Ahmed, 14 principal component analysis (PCA), partial least-squares (PLS), independent component analysis (ICA), Gaussian mixture model (GMM), and their derivates are the most widely used MSPM tools in the process industries.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Process Monitoring Methodology for Detection and Diagnosis of Unobservable Faults

Amin

Khan

Imtiaz

et al. 2019

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

This paper presents a new integrated methodology for fault detection and diagnosis. The methodology is built using the multivariate exponentially weighted moving average principal component analysis (MEWMA-PCA) and the Bayesian network (BN) model. The fault detection is carried out using the MEWMA-PCA; diagnosis is completed utilizing the BN models. A novel supervisory learning-based methodology has been proposed to develop the BNs from the historical fault symptoms. Although the algorithm has been extensively applied to the Tennessee Eastman (TE) chemical process, monitoring of three specific (difficult to observe) faults, IDV 3, IDV 9, and IDV 15, has been demonstrated in this article. Most of the existing data-based methods have faced the challenge to detect these faults with a good detection rate (DR). Hence, these faults have been reported as either unobservable or strenuous to detect. Overall, fault detection performance of the squared prediction error (SPE) statistics combined with the MEWMA-PCA was found to be better than the T 2-based monitoring model. Although the cumulative sum (CUSUM) PCA-based approaches have demonstrated successful detection and diagnosis to these specific faults, the comparative studies suggest that the proposed methodology can outperform the CUSUM PCA approach.

show abstract

Section: Industrial and Engineering Chemistry Researchmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Robust Process Monitoring Methodology for Detection and Diagnosis of Unobservable Faults

Amin

Khan

Imtiaz

et al. 2019

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…This paper introduced exponentially weighted moving average (EWMA) filter. Another technique with receiver operating characteristic-based curve for designing deadbands or thresholds to reduce FAR and MAR has been proposed in [13].…”

Section: A Brief Literature Review On Performance Assessment Of Alarmmentioning

confidence: 99%

Performance evaluation and design for variable threshold alarm systems through semi-Markov process

et al. 2020

View full text Add to dashboard Cite

In large industrial systems, alarm management is one of the most important issues to improve the safety and efficiency of systems in practice. Operators of such systems often have to deal with a numerous number of simultaneous alarms. Different kinds of thresholding or filtration are applied to decrease alarm nuisance and improve performance indices, such as Averaged Alarm Delay (ADD), Missed Alarm and False Alarm Rates (MAR and FAR). Among thresholdbased approaches, variable thresholding methods are well-known for reducing the alarm nuisance and improving the performance of the alarm system. However, the literature suffers from the lack of an appropriate method to assess performance parameters of Variable Threshold Alarm Systems (VTASs). This study introduces two types of variable thresholding and proposes a novel approach for performance assessment of VTASs using Priority-AND gate and semi-Markov process. Application of semi-Markov process allows the proposed approach to consider industrial measurements with non-Gaussian distributions. In addition, the paper provides a genetic algorithm based optimized design process for optimal parameter setting to improve performance indices. The effectiveness of the proposed approach is illustrated via three numerical examples and through a comparison with previous studies.

show abstract

“…Brooks et al put forward a geometric process control method to determine dynamic alarm trippoints from multivariate best operating space. Kondaveeti et al and Izadi et al . employed multivariate statistics to generate alarms efficiently.…”

Section: Introductionmentioning

confidence: 99%

“…Brooks et al 16 put forward a geometric process control method to determine dynamic alarm trippoints from multivariate best operating space. Kondaveeti et al 17 and Izadi et al 18 employed multivariate statistics to generate alarms efficiently. Yang et al 19 applied correlation analysis into the design of bivariate alarm trippoints to reduce the number of false and missed alarms.…”

Section: Introductionmentioning

confidence: 99%

Design of Multivariate Alarm Trippoints for Industrial Processes Based on Causal Model

Gao

Chen

Huang

et al. 2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Alarm trippoints are crucial parameters of alarm systems in industrial processes. Poorly designed alarm trippoints will probably bring about alarm overloading which is detrimental to the performance of alarm systems. In particular, the alarm design in recent alarm systems is mostly static and isolated from other related variables, resulting in the occurrence of nuisance alarms, including false alarms and missed alarms. A remedy is to design the multivariate alarm trippoints by taking the intricate correlation into account. This work proposes a novel framework to design the multivariate alarm trippoints based on a causal model. First, on the basis of our proposed Spatial Interpretative Structural Model (SISM), the alarm propagation paths under one specific fault are recognized. Next, on the basis of the proposed causal-sequential optimization (CSO) approach, the alarm trippoints of the selected key process variables on the paths are designed sequentially according to the causality. Then, two kinds of Kiviat diagram, namely the 3-D and 2-D time-oriented Kiviat diagrams, are introduced in this work to visualize the time trend and normal operation range determined by the optimal multivariate high and low alarm trippoints. Finally, the key benefits of our proposed framework are illustrated through the case studies of the Tennessee Eastman chemical plant.

show abstract

Application of Multivariate Statistics for Efficient Alarm Generation

Cited by 23 publications

References 8 publications

Robust Process Monitoring Methodology for Detection and Diagnosis of Unobservable Faults

Robust Process Monitoring Methodology for Detection and Diagnosis of Unobservable Faults

Performance evaluation and design for variable threshold alarm systems through semi-Markov process

Design of Multivariate Alarm Trippoints for Industrial Processes Based on Causal Model

Contact Info

Product

Resources

About