Optimal Compensator Design in Quantitative Feedback Theory.

Abstract

The Quantitative Feedback Theory (QFT) technique developed by Isaac Horowitz over a number of years, is perhaps the only controller design methodology that enables a controller to be designed to a given specification in a transparent quantitative manner. By this is meant that there is a definite quantitative measure of the closeness of the design to an optimum. A major advantage of QFT is the fact that the trade-offs between the constraints and the set of design criteria are visible to the designer in a transparent manner at all stages during the actual design process, rather than at the end, as is the case with 'black box' synthesis techniques such as H to infinity or LQC optimal control. The manual QFT method introduced by Horowitz and others in 1972 represented a major breakthrough in the quantitative design of robust controllers. However, the method is extremely labour intensive and the final loop-shaping stage of the design process requires substantial practice and expertise and it is believed that for this reason, the method has not been as widely accepted as it deserves to be. This report details research carried out to develop a computer-based method for optimal loop-shaping in QFT. Although some work has already been done in this area by Gera and Horowitz in 1980, no practical implementation details have been published. We believe that in OptComp we have made good progress in developing a program that enables the engineer to use QFT methods to design a compensator (or controller) iteratively to any desired order, while remaining transparent at all times about what trade-offs are necessary.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 17, 1996
Accession Number
ADA286989

Entities

People

  • A. L. Stevens
  • J. N. Ridley

Organizations

  • University of the Witwatersrand

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Boundaries
  • Coefficients
  • Compensation
  • Compensators
  • Complex Numbers
  • Composite Materials
  • Computers
  • Control Systems
  • Curve Fitting
  • Design Criteria
  • Errors
  • Rational Functions
  • Specifications
  • Standards
  • Transfer Functions

Readers

  • Robotics and Automation.
  • Systems Analysis and Design