Failure modeling of the tubular expansion process

Abstract

This thesis focuses on the research aimed at Solid Expandable Tubular Technology (SETT), which is a recent development in the fields of the manufacturing industry as well as petroleum exploration and drilling. Solid expansion of tubes is basically a cold working process where a conical mandrel is pushed axially into the tube to obtain the desired expansion ratio. Since the failure mechanism associated with SETT is ductile failure therefore the effects of stress state on the ductile failure behavior of these tubular and the mechanism leading to their rupture will be studied thoroughly. The ductile failure process essentially comprises of nucleation, growth and coalescence of the micro voids. This study will assist the understanding of ductile failure characterized by stress state in terms of triaxiality T and Lode parameter L. This is achieved by tensile testing on double notched tube specimen of Aluminum 6063-T5 for a wide range of stress triaxiality with L = 0. Different stress triaxiality levels are achieved by different notch configurations. The commercial finite element code ABAQUS is used to perform numerical simulations of these tests. A good correlation was achieved between the experimental results and numerical simulations. On the basis of experimental and numerical results, a failure envelope is developed in terms of plastic equivalent strain and stress triaxiality. After implementing this failure locus into ABAQUS explicit and simulating the quasi-static expansion process with required expansion ratios, it is validated through expanding the Aluminum tubes with holes drilled at different orientations. Since the failure envelope does not cover the entire range of triaxiality and Lode parameter as encountered in SETT, therefore some FEM and experimental results didn't agree with each other. Continuum damage mechanics (CDM) based approach has been used to develop this failure model. The validity of the model in predicting the onset of the failure is demonstrated by comparison of finite element predictions with experimental results which are in good correlation with each other. This validated model will be used to optimize the expansion process. Finally the conclusions will be drawn and recommendations would be suggested for the future work.

Date of Award	2012
Original language	American English
Supervisor	Imad Barsoum (Supervisor)