Calibration of nonlocal damage model from size effect tests

The calibration of nonlocal models which contain an internal length has been among the major issues conditioning the implementation of this kind of failure models. Direct calibration from uniaxial testing, where the state of strain remains homogeneous throughout the specimen, is impossible. The soft...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:European journal of mechanics, A, Solids A, Solids, 2003, Vol.22 (1), p.33-46
Hauptverfasser: LE BELLEGO, Caroline, DUBE, Jean Francois, PIJAUDIER-CABOT, Gilles, GERARD, Bruno
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The calibration of nonlocal models which contain an internal length has been among the major issues conditioning the implementation of this kind of failure models. Direct calibration from uniaxial testing, where the state of strain remains homogeneous throughout the specimen, is impossible. The softening law is not directly accessible because the strains cannot remain homogeneous during the entire test. In the absence of local information on the displacement field and on micro cracking in the fracture process zone, the calibration has to rely on inverse analysis. This paper presents such a procedure based on three point bend size effect tests on notched specimens. The complete load deflection curves are used for the identification of the constitutive relations. Manual calibration is discussed first. It is emphasised that calibration on the load deflexion curve from a single experiment is not objective. We show that Bazant's size effect law, which is related to peak loads only, may serve as a helpful guide to reach the closest fit. Then, automatic calibration is described. An optimal set of model parameters can be obtained within a reasonable number of iterations.
ISSN:0997-7538
1873-7285
DOI:10.1016/S0997-7538(02)01255-X