Time series methods for dynamic analysis of multiple controlled variables

Seppala, C.T.; Harris, Thomas J.; Bacon, D. W.

doi:10.1016/s0959-1524(01)00038-5

Cited by 33 publications

(26 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we shall consider an alternative method for the assessment of multi-variate control loop performance without relying on any a priori knowledge of the interactor matrices. There are several interactor matrix-free methods in the literature, mainly based on closed-loop impulse response [12,23,25], and variance of multi-step prediction errors [7,17,23]. Earlier work in using interactor-free approach may be traced back to [1,2].…”

Section: Assessment Of Multi-variate Control Performance Without Any mentioning

confidence: 99%

“…Therefore, this plot of r i versus i, is also an indication of closed-loop performance of a multi-variate controller, which has been used in some commercial software for multi-variate control performance monitoring and also in the literature [12,25]. In [23], the idea of impulse response as a measure of control performance has been extended to individual impulse response of each output to each shock of the disturbances to measure the interaction of variables. They have also proposed the use of forecast error variance decomposition (FEVD) for the measure of interactions.…”

Section: Assessment Of Multi-variate Control Performance Without Any mentioning

confidence: 99%

“…Note that an increase of the potential implies a deteriorated tuning while a decrease of the potential implies an improved tuning. In analogy to FEVD [23], closed-loop potentials can also be defined for the individual output variable and the relative closed-loop potential index for each output can also be derived. To calculate potential of individual variable, the trace operator tr[AE] should be replaced by diagonalization operator diag [AE].…”

Section: Motivated By the Interpretation Ofmentioning

confidence: 99%

“…We therefore recommend the use of the s i plot and the p i plot (to be discussed next). The s i plot or its equivalent has been discussed in [17,23]. The potential plot of p i shown in Fig.…”

Section: Examplementioning

confidence: 99%

“…Although the same rationale cannot be exactly carried over to multi-variate processes due to the relatively complex delay structure for multi-variate processes, multi-step optimal predictions provide useful information about the control performance [23]. The multi-step optimal prediction error is analogous to closed-loop step response of the univariate process from the white noise to the output.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Alternative solutions to multi-variate control performance assessment problems

Huang

Ding

Thornhill

2006

Journal of Process Control

View full text Add to dashboard Cite

Performance assessment of multi-variate control with minimum variance control as the benchmark requires an interactor matrix to filter the closed-loop output. This is to transfer the coordinate of the original variables into a new one in order to identify the control invariant disturbance dynamics from the first few terms of the closed-loop output Markov parameters. There has been a great deal of interest to simplify this approach, in particular, to find methods that do not need the interactor matrix. With this motivation, this paper explores alternative solutions to multi-variate control performance assessment problems. In particular, we will consider two practical scenarios: (1) known time delays between each pair of inputs and outputs, (2) no a priori knowledge about the process model or time delays at all. Solutions to these two scenarios are proposed. Two data-driven algorithms based on subspace approach are derived for the calculation of performance measures. Several examples illustrate the feasibility of the proposed approaches.

show abstract

Section: Assessment Of Multi-variate Control Performance Without Any mentioning

confidence: 99%

Section: Assessment Of Multi-variate Control Performance Without Any mentioning

confidence: 99%

Section: Motivated By the Interpretation Ofmentioning

confidence: 99%

Section: Examplementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Alternative solutions to multi-variate control performance assessment problems

Huang

Ding

Thornhill

2006

Journal of Process Control

View full text Add to dashboard Cite

show abstract

Introduction

2016

Process Control System Fault Diagnosis: A Bayesian Approach

View full text Add to dashboard Cite

Case studies in Bayesian segmentation applied to CD control

Taylor

Duncan

2003

Adaptive Control & Signal

View full text Add to dashboard Cite

Identifying step changes in historical process and controller output variables can lead to improved process understanding and fault resolution in control system performance analysis. This paper describes an application of Bayesian methods in the search for statistically significant temporal segmentations in the data collected by a cross directional (CD) control system in an industrial web forming process. CD control systems give rise to vector observations which are often transformed through orthogonal bases for control and performance analysis. In this paper two models which exploit basis function representations of vector times series data are segmented. The first of these is a power spectrum model based on the asymptotic Chisquared approximation which allows large data sets to be processed. The second approach, more capable of detecting small changes, but as a result is more computationally demanding, is a special case of the multivariate linear model. Given the statistical model of the data, inference regarding the number and location of the change-points is based on numerical Bayesian methods known as Markov chain Monte Carlo (MCMC). The methods are applied to real data and the resulting segmentation relates to real process events.

show abstract

Time series methods for dynamic analysis of multiple controlled variables

Cited by 33 publications

References 24 publications

Alternative solutions to multi-variate control performance assessment problems

Alternative solutions to multi-variate control performance assessment problems

Introduction

Case studies in Bayesian segmentation applied to CD control

Contact Info

Product

Resources

About