Concurrent monitoring of operating condition deviations and process dynamics anomalies with slow feature analysis

Chen, Shang; Yang, Fan; Gao, Xueshan; Huang, Xiaolin; Suykens, Johan A. K.; Huang, Dexian

doi:10.1002/aic.14888

Cited by 239 publications

(96 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike other feature extract methods, it can not only guarantee the global optimal solution but also obtain a series of features with varying rate from small to large. Owing to an input signal x ( t ) = [ x 1 ( t ), x 2 ( t ), ⋯ x m ( t )] T with m dimension, the SFA aims to determine a function g ( x ) = [ g 1 ( x ), g 2 ( x ), ⋯ g m ( x )] T , in which the feature s ( t ) = g( x ( t )) varies as slowly as possible, that is, to minimize

normalΔ ((), s_{i}) = min_{g_{i} (() •)} {〈{\dot{s}}_{i}^{2}〉}_{t},

under the constraints

{〈s_{i}〉}_{t} = 0,

{〈s_{i}^{2}〉}_{t} = 1,

\forall i \neq j : {〈s_{i} s_{j}〉}_{t} = 0,

where 〈•〉 t and

\overset{s}{true}

represent mean and the derivative of s with respect to time, respectively. Constraints exclude the constant solution and ensure that the solutions are independent.…”

Section: Preliminariesmentioning

confidence: 99%

“…13 Unlike other feature extract methods, it can not only guarantee the global optimal solution but also obtain a series of features with varying rate from small to large. Owing to an input signal x(t) = [x 1 (t), x 2 (t), ⋯x m (t)] T with m dimension, the SFA aims to determine a function g(x) = [g 1 (x), g 2 (x), ⋯g m (x)] T , in which the feature s(t) = g(x(t)) varies as slowly as possible, 13,14 that is, to minimize Δ s i ð Þ ¼ min…”

Section: Slow Feature Analysismentioning

confidence: 99%

“…The SFA, which was proposed by Wiskott, analyzes the changes in a sample and extracts dynamic information from historical time series data . Therefore, it can not only detect the deviations from operation conditions by monitoring a steady distribution but also identify the process dynamic anomalies in accordance with a temporal distribution . As a relatively new feature extraction technique, various SFA‐based monitoring methods have been developed to address different process characteristics.…”

Section: Introductionmentioning

confidence: 99%

“…SFA, as a new feature extraction method, is different from the general feature extraction method, which extracts slow‐changing components and reflects the dynamic information of process data . In addition, Shang et al established a prediction model via SFA to monitor quality‐relevant fault, and the results showed that slow feature regression (SFR) is superior to the traditional PLS method .…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Quality‐relevant dynamic process monitoring based on mutual information multiblock slow feature analysis

Zheng

Jiang

Yan

2019

Journal of Chemometrics

View full text Add to dashboard Cite

Slow feature analysis (SFA) is an efficient technique in exploring process dynamic information and is suitable for quality‐relevant process monitoring. However, involving quality‐irrelevant variables or features may introduce redundant information and degrade the monitoring performance. A novel multiblock monitoring scheme based on mutual information (MI) and SFA is proposed to detect an efficient quality‐relevant fault for dynamic processes. First, all process variables are divided into two blocks in accordance with their MI values with quality variables. Second, slow features (SFs) of two blocks are extracted via the SFA. The SFs, which are extracted from the quality‐relevant variables, are not all related to the quality variable. Thus, these SFs are further partitioned into two blocks in accordance with their MI values with quality variables. The SFs from three blocks are obtained, and monitoring statistics are constructed. Two simulation studies, including a numerical example and Tennessee Eastman process, demonstrate that the proposed method outperforms conventional methods.

show abstract

normalΔ ((), s_{i}) = min_{g_{i} (() •)} {〈{\dot{s}}_{i}^{2}〉}_{t},

under the constraints

{〈s_{i}〉}_{t} = 0,

{〈s_{i}^{2}〉}_{t} = 1,

\forall i \neq j : {〈s_{i} s_{j}〉}_{t} = 0,

where 〈•〉 t and

\overset{s}{true}

represent mean and the derivative of s with respect to time, respectively. Constraints exclude the constant solution and ensure that the solutions are independent.…”

Section: Preliminariesmentioning

confidence: 99%

Section: Slow Feature Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Quality‐relevant dynamic process monitoring based on mutual information multiblock slow feature analysis

Zheng

Jiang

Yan

2019

Journal of Chemometrics

View full text Add to dashboard Cite

show abstract

“…This method incorporates the preference of slowness in formulating the latent features 40 and works effectively in modeling process data. 41 While adopting above structure to enhance the ability in noise modeling, 42 the transition matrix A for slow features is fixed as diagonal and constrained with the covariance matrix Q…”

Section: An Overview Of Dynamic Feature Extractionmentioning

confidence: 99%

Extracting dynamic features with switching models for process data analytics and application in soft sensing

Huang

2018

AIChE Journal

View full text Add to dashboard Cite

In recent decades, soft sensors have been profoundly studied and successfully applied to predict critical process variables in real-time. While dealing with various application scenarios, data-driven methods with representation learning possess great potentials. Latent features are formulated in these approaches to predict outputs from correlated input variables. In this study, a probabilistic framework of feature extraction is proposed in the context of process data analysis. To address switching behaviors in industrial processes, multiple emission models are utilized to construct latent space. To address temporal correlations from continuously operating processes, a dynamic model is implemented in latent space. Bayesian learning strategy is then developed for parameters estimation, where modeling preferences and uncertainties from multiple models are considered. The effectiveness and practicability of the proposed feature extraction algorithm are illustrated through numerical simulations, as well as an industrial case study.with t 5 T case and hU ðkÞ t I ðkÞ t i Estimate S t 5S ðf Þ t from (38) to (37)

show abstract

Probabilistic slow feature analysis‐based representation learning from massive process data for soft sensor modeling

et al. 2015

View full text Add to dashboard Cite

Latent variable (LV) models provide explicit representations of underlying driving forces of process variations and retain the dominant information of process data. In this study, slow features (SFs) as temporally correlated LVs are derived using probabilistic SF analysis. SFs evolving in a state‐space form effectively represent nominal variations of processes, some of which are potentially correlated to quality variables and hence help improving the prediction performance of soft sensors. An efficient expectation maximum algorithm is proposed to estimate parameters of the probabilistic model, which turns out to be suitable for analyzing massive process data. Two criteria are also proposed to select quality‐relevant SFs. The validity and advantages of the proposed method are demonstrated via two case studies. © 2015 American Institute of Chemical Engineers AIChE J, 61: 4126–4139, 2015

show abstract

Concurrent monitoring of operating condition deviations and process dynamics anomalies with slow feature analysis

Cited by 239 publications

References 40 publications

Quality‐relevant dynamic process monitoring based on mutual information multiblock slow feature analysis

Quality‐relevant dynamic process monitoring based on mutual information multiblock slow feature analysis

Extracting dynamic features with switching models for process data analytics and application in soft sensing

Probabilistic slow feature analysis‐based representation learning from massive process data for soft sensor modeling

Contact Info

Product

Resources

About