Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.566001
Title: Prediction of material damage in orthotropic metals for virtual structural testing
Author: Ravindran, S.
Awarding Body: Cranfield University
Current Institution: Cranfield University
Date of Award: 2010
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
Models based on the Continuum Damage Mechanics principle are increasingly used for predicting the initiation and growth of damage in materials. The growing reliance on 3-D finite element (FE) virtual structural testing demands implementation and validation of robust material models that can predict the material behaviour accurately. The use of these models within numerical analyses requires suitable material data. EU aerospace companies along with Cranfield University and other similar research institutions have created the MUSCA (non-linear MUltiSCale Analysis of large aero structures) project to develop virtual structural testing prediction. The MUSCA project focuses on static failure testing of large aircraft components. It aims to reduce laboratory tests using advanced numerical analysis to predict failure in order to save overall cost and development time. This thesis aims to improve the current capability of finite element codes in predicting orthotropic material behaviour, primarily damage. The Chow and Wang damage model has been implemented within ABAQUS as a VUMAT subroutine. This thesis presents the development of a numerical damage prediction model and an experimental study to develop a damage material characterisation process that can easily be performed using standard tensile test specimen and equipment already available in the aerospace industry. The proposed method makes use of Digital Image Correlation (DIC), a non-contact optical strain field measurement technique. Experiments were conducted at Cranfield University material testing facility on aerospace aluminium alloy material AA-2024-T3 and AA-7010-T7651. After thorough literature survey a complete new method was formulated to implement Chow and Wang damage model in Abaqus Explicit numerical code. The damage model was successfully implemented for isotropic and orthotropic behaviour using single element model, multi-element coupon test model and a simple airframe structure. The simulation results were then verified with the similar experimental results by repeating the experimental procedure using simulation for each material type and found matching results. The model is then compared with experimentally determined orthotropic material parameter for AA2024 and AA7010 for validation and found agreeable results for practical use. The material characterisation of damage parameters from standard tensile specimen using DIC technique was also demonstrated and the procedures were established. In this research the combination of experimental work and numerical analysis with clear and simpler calibration strategy for damage model is demonstrated. This is the important contribution of this research work and the streamlined procedures are vital for the industry to utilise the new damage prediction tools. The damage model implementation and test procedures developed through this research provide information and processes involved in fundamentally predicting the ductile damage in metals and metal alloys. The numerical damage model developed using the well-defined verification and validation procedures explained in this research work with new streamlined damage material characterisation using recent contact less DIC technique has wider implication in the material model development for ductile metals in general. The thesis ultimately delivered a fully verified, validated robust damage model numerical simulation code with a new DIC damage characterisation procedure for practical application. The model is now used by the aerospace industry for predicting damage of large aircraft structures.
Supervisor: Campbell, James Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.566001  DOI: Not available
Share: