Change search
ReferencesLink to record
Permanent link

Direct link
Development of an inverse modelling programming system for evaluation of Ti-6Al-4V Gleeble experiments
2003 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This master’s thesis is a beginning part of the larger VINNOVA financed research project MMIP – Material Modelling in Industrial Processes, in cooperation with Volvo Aero Corporation. The research project’s main aim is to develop methods and routines for material modelling in industrial processes. The work conducted here embraces the understanding and development of a programming system for identification of material parameters by inverse modelling. This comprise a number of sections such as stating demands and specification of Gleeble experiments, modelling a proper FE-model of the experiments with a non-linear finite element method, programming of inverse system, conduction of material parameter identification and evaluation of the system and thereby received results. A literature study, examination of microstructures and a study of three different constitutive models have also been performed. The study of titanium and especially the alloy Ti-6Al-4V show that the material has a complicated behaviour. The mechanical properties are strongly influenced by the chemical composition and heat treatment together with the material’s history and thereby the alteration of micro-structure and phase composition. Compression tests of cylindrical specimens of a forging in the alloy were conducted in a so-called Gleeble 1500 machine. The tests show the alloy’s response to deformation and also that the chosen heat treatment has affected the mechanical properties by formation of a brittle oxide layer, so- called alpha case. The microstructure was not visible affected except for the deformation caused by the compression. This resulted in a deformed structure especially at mid-height of the specimen. Three test series were conducted – isothermal tests at different temperatures, tests for different strain rates at 500 degree Celsius and compression tests during continuous cooling. All tests were heat treated for 20 minutes at 950 degree Celsius and with a cooling rate of 25 degree Celsius/second. At the lower temperatures 25 degree Celsius and 300 degree Celsius fracture occurred at which remaining tests at lower temperatures failed to come off. The isothermal tests at 700 degree Celsius and 900 degree Celsius showed a material hardening followed by a flow softening while at 500 degree Celsius a hardening followed by a flatten behaviour was observed. At the lower temperatures an initial hardening were observed followed by specimen cracking. In order to allow the estimation of the material parameters that describes the thermo mechanical behaviour of the alloy an FE-model that mimics the experiment has been set up. The non-linear finite element code MSC. Marc has been used for the analyses. In the FE-model a constitutive model that manage to describe the behaviour of the alloy must be implemented. It is then the parameters in this constitutive model that constitute the design parameters. The suitability of three constitutive models have been examined: Johnson and Cook, Nemat-Nasser and a polynomial description. The isothermal test at 500 degree Celsius constitutes the experimental data for the determination of the constitutive model’s suitability to describe the material behaviour. During the material parameter estimation it turns out that none of the three constitutive models can represent the experiment. At which parameter estimation against a “fake” experiment has been conducted. The constitutive model by Nemat-Nasser was chosen and the estimation was successful. This since the estimated parameters describes the “fake” experimental hardening. It shall be mentioned that different sets of parameters can describe the same material behaviour at which it is important that the error between the experimental response and the response generated by the FE-analysis is small enough. It is also important to exclude substantial local minimums by investigation of different initial guesses. Further it is important to exclude any errors in the user written programming code. The main aim of this project is hereby achieved and the inverse modelling programming system is ready for parameter estimation as soon as a proper constitutive model has been identified.

Place, publisher, year, edition, pages
Keyword [en]
Technology, inverse modelling, Ti-6Al-4V, thermo mechanical properties, Gleeble experiments, finite element analysis, material, parameters
Keyword [sv]
URN: urn:nbn:se:ltu:diva-49957ISRN: LTU-EX--03/108--SELocal ID: 74258e60-cb5f-435f-ae73-8a5062bf1eafOAI: diva2:1023310
Subject / course
Student thesis, at least 30 credits
Educational program
Mechanical Engineering, master's level
Validerat; 20101217 (root)Available from: 2016-10-04 Created: 2016-10-04Bibliographically approved

Open Access in DiVA

No full text

Search outside of DiVA

GoogleGoogle Scholar

Total: 12 hits
ReferencesLink to record
Permanent link

Direct link