Skip to content

Validation of an automatic limit analysis technique for rigid block systems

Research output: Chapter in Book/Report/Conference proceedingConference contribution

A segmental arch constructed from concrete blocks is tested to failure as validation of a discrete automatic limit analysis technique. A mesh description of geometric compatibility, based on Kirchhoff's network laws, provides the model on which collapse is defined. A minimisation of the work rate of the parametric loads provides a prediction of the failure load and configuration of collapse through the use of a Linear Program. However, the predicted strength of the test structure is significantly overestimated. A detailed set of geometric measurements were seen to reveal small imperfections in the concrete blocks which when assembled create a geometrically flawed structure. The flexibility of the procedure allows for such imperfections to be incorporated into the model by manipulation of the limit conditions defined for each of the interfaces between blocks. The re-analysis gives good agreement with the observed results, highlighting the effect imperfections have on the strength of such a geometrically defined structure. The numerical analysis procedure is seen to be practical, and to provide an accurate prediction of strength.

Original languageEnglish
Title of host publicationComputational Methods and Experimental Measurements VIII
PublisherWIT Press
Pages119-128
Number of pages10
ISBN (Print)9781853124631
Publication statusPublished - 1 Jan 1997
EventProceedings of the 1997 8th International Conference on Computational Methods and Experimental Measurements - Rhodes, Greece
Duration: 1 May 19971 May 1997

Publication series

NameWIT Transactions on Modelling and SImulation
PublisherWIT Press
Volume17
ISSN (Electronic)1743-355X

Conference

ConferenceProceedings of the 1997 8th International Conference on Computational Methods and Experimental Measurements
Abbreviated titleCMEM
CountryGreece
CityRhodes
Period1/05/971/05/97

Related information

Relations Get citation (various referencing formats)

ID: 21066841