ONLINE MATERIAL CHARACTERIZATION USING FULL-FIELD STRAIN MEASUREMENT

摘要

This paper presents and reviews an online methodology which characterizes materials using full-field strain measurement. The proposed methodology utilizes the principle of conservation of energy and formulates both the deterministic technique based on the pseudoinverse analysis and the stochastic technique based on the Kalman filter in terms of recursive linear equations. The methodology further describes the derivation the average Frobenius norm and the differential entropy as recursively computable measures enabling the evaluation of the well-posedness of the material characterization problem as well as the uncertainty of the identified constants. Comparative studies have identified that the deterministic identification is a particular case of the stochastic identification, whilst the adequacy and significance of both the average Frobenius norm and the differential entropy was reconfirmed.