A common practice in a system identification exercise is to perturb the system of interest and use the resulting data to build a model. The problem of interest in this contribution is to synthesize an input signal that is maximally informative for generating good quality models while being "plant friendly", i.e., least hostile to plant operation. In this contribution, limits on input move sizes are the plant friendly specifications. The resulting optimization problem is nonlinear and nonconvex. Hence, the original plant friendly input design problem is relaxed which results in a convex optimization problem. We formulate a SemiDefinite Programme using the theory of generalized Tchebysheff inequalities to derive tight bounds on the quality of relaxation. Simulations show that the relaxation results in more plant friendly input signals. © 2006 IEEE.

Raghunathan Rengaswamy

Department Of Chemical Engineering

Sridharakumar Narasimhan

Convex optimization problems

CONVEX RELAXATION

INPUT DESIGN

Input signal

NONCONVEX

Optimization problems

PLANT OPERATIONS

QUALITY MODELS

Semi-definite programming

System identifications

TCHEBYCHEFF INEQUALITIES

Tight bound

Convex optimization

Optimization

Relaxation processes

Design

IIT Madras is a public technical and research university located in Chennai, Tamil Nadu. Founded in 1959, it is recognised as an Institute of National Importance.

IIT Madras has been ranked as the top engineering institute in India for four years in a row by the National Institutional Ranking Framework of the MHRD

It currently offers undergraduate, postgraduate and research degrees across 16 disciplines in Engineering, Sciences, Humanities and Management. About 596 faculty belonging to science and engineering departments and centres of the Institute are engaged in teaching, research and industrial consultancy.

IIT Madras

Plant Friendly Input Design: Convex Relaxation and Quality

IEEE Transactions on Automatic Control

In several industrial processes, measurements are limited due to time or cost constraints. Hence, scheduling of these measurements to obtain maximally informative and accurate estimates of the states becomes important. The focus of this contribution is to determine a schedule of measurements that maximizes the quality of the estimates at the end of a finite time horizon. This work finds applications in batch chemical processes. The estimate covariance matrix as obtained by applying standard Kalman filtering theory is approximated and the original problem is relaxed and reformulated as a cone program. For a certain class of systems, the approximated covariance matrix is parameterized in terms of a point belonging to a moment space induced by a Tchebycheff system. The theory of Tchebycheff systems is used to determine an improved upper bound on the guaranteed minimum number of measurements. A tractable solution methodology to obtain the optimal cost and a parsimonious discrete schedule using the theory of cone programming, generalized barrier functions and Tchebycheff systems is described. © 2014. IEEE.

Optimal sensor scheduling in batch processes using convex relaxations and tchebycheff systems theory

One of the important requirements in finish milling is maintaining consistent product quality. Quality of cement is measured through blaine which is a measure of the specific surface area or the fineness of cement. Since process adjustments are made based on this quality measurement, infrequent sampling may result in production loss and inconsistent product quality. With this motivation, a soft-sensor deploying kernel principal component regression (KPCR) is sought wherein blaine can be predicted at desired intervals to be used in control, which in turn can guarantee consistent quality and improved production. The KPCR is equivalent, but not identical, to working with a linear model on non-linear transformation of the original variables so as to facilitate working with linear regression technique in feature space instead of using a nonlinear regression technique in the original variable space. The model developed using KPCR was tested with data collected from Cemulator (FLSmidth simulation tool) as well as with real-time data collected from the cement ball mill. The blaine predicted via KPCR has the required prediction accuracy necessary for use in ball mill control. © 2014 IFAC.

IFAC Proceedings Volumes (IFAC-PapersOnline)

Developing a soft sensor for fineness in a cement ball mill

Optimal experiment design for system identification involves determining an optimal input that is used to perturb the system so that the resulting input-output data is maximally informative. Plant friendly identification requires that constraints on input move sizes, output sizes or variance and experiment time be respected. The solution to the optimum input design problem depends on the unknown parameters to be estimated which is often approximated by an initial estimate. Use of the estimate is likely to result in loss in performance or violation of the constraints. An alternative is to formulate a robust optimization problem with uncertain parameters. The contribution of this work is to use the uncertainty sets originating from a prior identification exercise to solve a robust plant friendly input design problem. The methodology is derived for a general class of systems illustrated using numerical simulations. Simulations validate the expectation that the constraints are probabilistically more likely to be satisfied using the robust design than a nominal design based on uncertain parameters. © IFAC.

Fulltext

Robust plant friendly optimal input design

Optimal experiment or input design is the scientific exercise of designing informative excitation signals for the identification of a real-life dynamic system. In the least costly input design framework, the input is designed such that the identification cost is minimized while meeting desired specifications on the quality of the identified model. Identification of real-life processes require that the identification be "plant-friendly". These are typically imposed as constraints on experiment time, input and output amplitudes or input move sizes. This work focusses on an LMI based plant friendly input design in the least costly framework. © 2012 IFAC.

Plant friendly input design for system identification in closed loop

The primary objective in solving optimal input design problems is to obtain maximally informative inputs to be used as perturbation signals in system identification experiments. In plant-friendly identification, the designer has to respect constraints on experiment time, input and output amplitudes or input move sizes. This work focuses on plant friendly input design with constraints on input move size and output power. We present a convex relaxation to the problem of designing an informative input subject to input move size and output power constraints. The problem is finitely parametrized using ideas from Tchebycheff systems and reformulated as a SemiDefinite Programme. © 2011 American Chemical Society.

Industrial and Engineering Chemistry Research

Optimal plant friendly input design for system identification

Cambridge University Press

Convex Optimization

A Quasi-Newton Interior Point Method for Low Order H-Infinity Controller Synthesis

Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference

A Primal-Dual method for low order H<inf>&#x221E;</inf> controller synthesis

Journal of Process Control

Constrained multisine input signals for plant-friendly identification of chemical process systems

IFAC Proceedings Volumes

Multiobjective Robust Control with HIFOO 2.0

Plant friendly input design: Convex relaxation and quality

Journal	IEEE Transactions on Automatic Control
ISSN	00189286
Open Access	No