Other Reports

A Physically Based Method of Enhancement of Experimental Data-Concepts, Formulation, and Application to Identification of Residual Stresses

  • 01
  • Oct
  • 1993
AUTHOR: J. Orkisz, W. Karmowski
SUBJECT: Evaluation
KEYWORDS: Experimental Data, Mathematical Model, Theoretical Model
ABSTRACT: This paper presents the formulation for a new approach to enhancing experimental data. Using simultaneously all the information available for the problem being studied, i.e, physically based approximation, this approach offers potential advantages over the typical methods used for fitting a curve or a surface to experimental data. Those methods are based primarily on the least squares technique or on a similar mathematical method. While their results are usually compared with a relevant theoretical solution, the new combined approach produces results simultaneously from both the various experimental measurements and the theoretical model. In this way only one, i.e., "the best," solution is obtained. The new approach tries to fit best the experimental data and, at the same time, to satisfy, as much as possible, the requirements of the theory involved. In terms of mathematics, the new approach yields constrained optimization problems for a functional composed of experimental and theoretical parts. Presented are the general concept of the approach and, as a particular example, its formulation for strain gage measurements, as well as its application for residual stress analysis.