1979International Journal of Systems ScienceRequires access

A method for modelling transfer functions using dominant frequency-response data and its applications

L.S. Shieh, M. Datta-Barua, R.E. Yates

Open publisher page 10 citations

Abstract

This paper presents a fundamental method for modelling transfer functions using the basic performance specifications and frequency-response data at the dominant frequencies. A set of non-linear equations is constructed from the definitions of the basic performance specifications, the dominant frequency-response data and the unknown coefficients of a transfer function. A Newton-Raphson multidimensional method is applied to solve the non-linear equations. Four methods are given to construct approximate representations of the desired transfer functions for the estimation of good starting values to ensure rapid convergence of the numerical method. The applications of the proposed method are : (1) developing a standard model and/or a transfer function of a filter or a compensator using the specified dominant frequency-response data ; (2) identifying the transfer function of a system from available experimental frequency-response data ; and (3) reducing high-order transfer functions to low-order models using dominant frequency-response data.

About this research paper

What this paper is about

This paper presents a fundamental method for modelling transfer functions using the basic performance specifications and frequency-response data at the dominant frequencies. A set of non-linear equations is constructed from the definitions of the basic performance specifications, the dominant frequency-response data and the unknown coefficients of a transfer function. A Newton-Raphson multidimensional method is applied to solve the non-linear equations. Four methods are given to construct approximate representations of the desired transfer functions for the estimation of good starting values to ensure rapid convergence of the numerical method. The applications of the proposed method are : (1) developing a standard model and/or a transfer function of a filter or a compensator using the specified dominant frequency-response data ; (2) identifying the transfer function of a system from available experimental frequency-response data ; and (3) reducing high-order transfer functions to low-order models using dominant frequency-response data.

Why it matters

OpenAlex reports 10 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper presents a fundamental method for modelling transfer functions using the basic performance specifications and frequency-response data at the dominant frequencies. A set of non-linear equations is constructed from the definitions of the basic performance specifications, the dominant frequency-response data and the unknown coefficients of a transfer function. A Newton-Raphson multidimensional method is applied to solve the non-linear equations. Four methods are given to construct approximate representations of the desired transfer functions for the estimation of good starting values to ensure rapid convergence of the numerical method. The applications of the proposed method are : (1) developing a standard model and/or a transfer function of a filter or a compensator using the specified dominant frequency-response data ; (2) identifying the transfer function of a system from available experimental frequency-response data ; and (3) reducing high-order transfer functions to low-order models using dominant frequency-response data.

Key concepts: Transfer function, Frequency response, Computer science, Convergence (economics), Filter (signal processing), Function (biology), Step response, Set (abstract data type)

Related papers

Back to paper searchBrowse research topicsOriginal source
A method for modelling transfer functions using dominant frequency-response data and its applications — Research Paper | ScholarLens