Identification of Multiple Sensor Disturbances during Process Monitoring

Stork, Chris L.; Veltkamp, and David J.; Kowalski, Bruce R.

doi:10.1021/ac9704366

Cited by 23 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A slurry of nuclear waste and glassforming chemicals were fed into the LFCM, and glass was periodically poured from the melter, resulting in time-dependent variations in the glass level. The LFCM was monitored with 10 thermocouples located in two thermo-wells within the glass pool (Stork et al, 1997). Raw data set was treated to have 20% missing at random.…”

Section: Simulation and Resultsmentioning

confidence: 99%

Calibration, Prediction and Process Monitoring Model Based on Factor Analysis for Incomplete Process Data

Kim

Jung

et al. 2005

J. Chem. Eng. Japan

View full text Add to dashboard Cite

Unspecific missing of values in real chemical and biological industries have been found. Regardless of the incompleteness of the measured sample, a monitoring system should be designed to tackle the missing data problem and be applied to on-line systems immediately. A calibration method of a factor analysis (FA) model for incomplete data sets is proposed. And a prediction method based on the calibrated model is suggested in order to estimate missing values in incomplete calibration sets and incomplete test sets. An expectation and maximization (EM) algorithm is used to calibrate the model and expectation of conditional density is used to predict the model result. The proposed method is compared with the wellknown iterative singular values decomposition (iSVD) method, i.e. a principal component analysis (PCA) based method; and a simple data set is tested as an illustrative example. The proposed method gives better estimation results for the missing values than the well-known PCA based method. There are several advantages of the proposed method over the PCA based projection methods: (1) data pretreatment is not an essential step since the FA model is scale invariant whereas the PCA model is not, (2) since the proposed method utilizes probability information of all variables directly, to apply it as a statistical process monitoring technique is preferable to others, and (3) the single model can be extended to a mixture of such models by the virtue of the EM algorithm.

show abstract

Section: Simulation and Resultsmentioning

confidence: 99%

Calibration, Prediction and Process Monitoring Model Based on Factor Analysis for Incomplete Process Data

Kim

Jung

et al. 2005

J. Chem. Eng. Japan

View full text Add to dashboard Cite

show abstract

“…This algorithm has been applied to the unmodeled disturbances by means of a Q-ratio. 14 The procedure is applied to the outlying Q samples, in which each sensor is sequentially removed from the m odel and the anomalous sample. The Q 2 j values of the detected sample without sensor j are compared to the corresponding upper limit value Q lim2 j as follows:…”

Section: Th Eoretical Basis H Otelling T 2 and Q Charts 15mentioning

confidence: 99%

“…Likewise, we incorporated an algorithm to detect the variables responsible for the outliers 14 and to help account for why the sample behaves like an outlier.…”

Section: Introductionmentioning

confidence: 99%

Outlier Detection in the Ethylene Content Determination in Propylene Copolymer by Near-Infrared Spectroscopy and Multivariate Calibration

et al. 2001

View full text Add to dashboard Cite

In this study, we employed multivariate control techniques to detect outliers in the determination of ethylene in impact polypropylene samples by near-infrared (NIR) spectroscopy and multivariate calibration partial least-squares (PLS). We also applied an algorithm which identifies those spectral variables responsible for the outlier behavior and that can indicate the source of this behavior. The outliers in the prediction step may be due to three possible situations: errors associated with the prediction of analyte concentrations in samples that have the same characteristics as the calibration set, but that are beyond the concentration range; changes in the matrix composition; and instrumental errors. We show that the proposed techniques make it possible to detect whether or not an analyte belongs to the reference set. In addition, we apply an algorithm that identifies the variables that cause outlier behavior and assigns them to a class.

show abstract

“…Another problem is that a fault in one sensor could lead to an increase in the contributions of other sensors, 8 thereby increasing the chance of false identification. To identify multiple-sensor faults, Stork et al 10 proposed a backward elimination sensor identification (BESI) algorithm that requires a time-consuming recalculation of the PCA model every time one sensor is removed from the model. Another drawback of BESI is that T 2 is not considered.…”

Section: Introductionmentioning

confidence: 99%

Reconstruction-Based Fault Identification Using a Combined Index

Yue¹,

Qin

2001

Ind. Eng. Chem. Res.

451

296

View full text Add to dashboard Cite

Process monitoring and fault diagnosis are crucial for efficient and optimal operation of a chemical plant. This paper proposes a reconstruction-based fault identification approach using a combined index for multidimensional fault reconstruction and identification. Fault detection is conducted using a new index that combines the squared prediction error (SPE) and T 2 . Necessary and sufficient conditions for fault detectability are derived. The combined index is used to reconstruct the fault along a given fault direction. Faults are identified by assuming that each fault in a candidate fault set is the true fault and comparing the reconstructed indices with the control limits. Fault reconstructability and identifiability on the basis of the combined index are discussed. A new method to extract fault directions from historical fault data is proposed. The dimension of the fault is determined on the basis of the fault detection indices after fault reconstruction. Several simulation examples and one practical case are presented. The method proposed here is compared with two existing methods in the literature for the identification single-sensor and multiple-sensor faults. We analyze the reasons that the other two methods lead to erroneous identification results. Finally, the proposed approach is applied to a rapid thermal annealing process for fault diagnosis. Fault subspaces of several typical process faults are extracted from the data and then used to identify new faults.

show abstract

Identification of Multiple Sensor Disturbances during Process Monitoring

Cited by 23 publications

References 13 publications

Calibration, Prediction and Process Monitoring Model Based on Factor Analysis for Incomplete Process Data

Calibration, Prediction and Process Monitoring Model Based on Factor Analysis for Incomplete Process Data

Outlier Detection in the Ethylene Content Determination in Propylene Copolymer by Near-Infrared Spectroscopy and Multivariate Calibration

Reconstruction-Based Fault Identification Using a Combined Index

Contact Info

Product

Resources

About