Model Structure Determination and Identifiability Problems in System Identification.

Abstract

System identification has become one of the most active areas in system theory and its applications. In many engineering applications where the system to be identified or estimated is governed by differential or difference equations with known structure, the problem is mainly that of parameter estimation. As the authors extend the concept of system identification to those classes of problems where prior knowledge on structure is limited, some basic problems other than parameter estimation become important. System identification consists of three basic sub-problems: (1) pre-estimation, where one is concerned with structure specification, types of data to be collected, and experimental design, (2) estimation, where the main emphasis is on estimation algorithm development, and (3) post-estimation, where one has to test the validity of using the model for a specific objective. The program of studies reported here is on the basic problem in system identification, with more emphasis on pre-estimation and post-estimation problems.

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 1975
Accession Number
ADA009404

Entities

People

  • Edison Tse
  • Narendra Gupta

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Data Science
  • Difference Equations
  • Differential Equations
  • Engineering
  • Equations
  • Experimental Design
  • Identification
  • Information Science
  • Mathematical Analysis
  • Mathematics
  • Real Variables
  • Specifications
  • Statistical Algorithms

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Computational Modeling and Simulation