Proper orthogonal decomposition and radial basis functions algorithm for diagnostic procedure based on inverse analysis

• Published in: FME transactions Vol. 38; no. 3; pp. 129 - 136
• Main Author: Buljak Vladimir
• Format: Journal Article
• Language: English
• Published: University of Belgrade - Faculty of Mechanical Engineering, Belgrade 01.01.2010
• Subjects: instrumented indentation; inverse analyses; material characterization; proper orthogonal decomposition
• ISSN: 1451-2092; 2406-128X

Identification procedures based on instrumented indentation and inverse analysis are traditionally coupled with Finite Element Modelling (FEM) to perform simulation of the test. However, this approach is not suitable for in-situ applications since it is rather time consuming due to material and geometrical nonlinearities required to be taken into account. This paper presents a novel technique for system response prediction based on Proper Orthogonal Decomposition and Radial Basis Functions. The developed technique gives the results of the same accuracy as those computed by FEM in computing times shorter by several orders of magnitude. Presented examples consider two different engineering applications. The first deals with the assessment of material parameters entering into the constitutive models of possibly damaged materials used for industrial plants. The second considers the identification of residual stresses which arise in components after surface treatments.