Sachi Dash

Integrated system identification and PID controller tuning by frequency loop-shaping

A systematic design methodology for proportional-integral-derivative (PID) controllers is presented. Starting from data sets, a model of the system and its uncertainty bounds are obtained. The parameters of the controller are tuned by a convex optimization algorithm, minimizing a weighted difference between the actual loop transfer function and a target in an /spl Lscr//sub 2///spl Lscr//sub /spl infin// sense. The target selection is guided by the identified model and its uncertainty. The problem of disjoint data sets and/or different models for the same system is also addressed. The method has proved successful in numerous practical cases showing both expediency in controller design and implementation and improved performance over existing controllers.

Integrated MIMO identification and robust PID controller design through loop shaping

A framework for integrated identification and robust PID control for multivariable processes is presented. The method relies on robust loop shaping bounds computed directly from identification data using the concept of the structured singular value (/spl mu/). These performance bounds are then used to define the nominal loop shape required to tune PID controllers.

Loop-shaping controller design from input-output data: application to a paper machine simulator

The main motivation for this work was the need for a systematic design of high-performance controllers for paper machines. An integrated identification and controller design approach, based on loop-shaping principles, is presented. Starting with time-domain input-output data, a coprime factorization of the plant is identified using linear least-squares techniques. The residual error from the identification is then used to obtain estimates of the coprime factor uncertainty that are translated to sensitivity and complementary sensitivity bounds for the closed-loop system. These bounds serve as a guide for the selection of a target loop and the controller is designed using H-infinity tools. The application of this approach is demonstrated on Honeywells high-fidelity simulator. The simulations demonstrate the suitability of the approach and illustrate that the technique can be used to reliably provide high performance for both servo and regulatory control.

Identification for PID Control

In this study, we address issues arising in system identification-based plant modeling for the purpose of tuning PID controllers. We discuss some of the more celebrated tuning methods and make an attempt to anticipate future directions of identification procedures in the effort to provide reliable answers to more involved problems. The algorithmic progression of the different methods starts with easy-to-estimate, minimal information about the plant, in the form of a single frequency point in the Nyquist plot or a couple of parameters of the step response, aiming to produce the tuning parameters with back-of-the-envelope calculations. High order general models from either first-principles modeling or system identification were then used in a variety of off-line or on-line optimization problems to produce optimal PID tunings, and were often converted to the always elusive set of quick calculations that could tune a PID controller in the field. Highest in complexity, are the most recent methods which involve maximal information about the plant, in the form of one or more nominal models and a description of uncertainty, and aim for a tuning that combines high performance, adequate robustness, and high reliability. The latter appears to be the key qualitative difference between early and late identification and controller tuning procedures, that is, the ability to provide a reliable tuning with minimal trial-and-error. An alternative implementation of the same concepts can be performed as a direct optimization of the PID parameters, that can be readily converted to an attractive on-line tuning application. Finally, employing more complex, min–max identification methods can safeguard against performance deterioration problems due to disturbances or poorly designed excitation, albeit with a significant increase in computational load and identification time.

Adaptive PID control using filter-banks and frequency loop shaping

This paper addresses issues arising in the online adaptation of PID controller parameters. Employing a frequency loop shaping approach to define the control objective, it is possible to adapt the PID parameters directly with standard least squares algorithms. In addition to this, the use of filter-banks in the regressor leads to the approximate minimization of the H-infinity norm of the error operator. The practical implication of this observation is that PIDs can be reliably tuned, even in cases of large mismatch between the target and the feasible loop shapes. Numerical examples are included, illustrating the properties and the implementation of the algorithm.

Integrated identification and robust control for paper machines

An integrated modeling and robust multivariable control approach is presented for paper machines. An application of this approach is demonstrated on a high-fidelity simulator. Modeling relies on input-output data, collected from an identification experiment at the desired operating condition. An estimate of the process model along with the uncertainty bounds that describe the confidence limits of the model, consistent with the robust control theory, is obtained. The results are used to design an multivariable controller based on loop-shaping principles and guided by the estimated uncertainty bounds. The simulations demonstrate the suitability of the approach and illustrate that the technique can be used to provide high bandwidth performance for both servo and regulatory control.

Implementation of controller performance monitoring on a MIMO example

In this paper, we discuss the implementation of various controller performance monitoring criteria on a multivariable benchmark example. These criteria can be based on either time-domain or frequency-domain concepts and have different requirements for information content and algorithm complexity. The simulation results from the benchmark example are analyzed to derive conclusions on the consistency of each criterion to identify poorly tuned controllers. Typically, all methods generate consistent results with suitably designed tests and when the only source of performance deterioration is the controller tuning. However, in the presence of arbitrary external disturbances (e.g., large amplitude, deterministic) and varying levels of excitation, not all techniques yield consistent results. In this respect, criteria based on controller unfalsification concepts appear to be the more promising to yield reliable performance monitoring for operation in industrial environments