Predictive Modelling and Data Development for Optimisation of Complex Structure and Materials

Abstract

Abstract

This thesis is focused on reporting the research and development of an integrated approach for the analysis and data development of complex material processing, design analysis and materials modelling. The main cases presented include experimental analysis and modelling of electrical resistance spot welding processes, the mechanical deformation of spot-welded joints under tensile shear loading and image-based analysis and modelling of complex M7C3 (M=Fe, Cr, metal elements) carbides. In each case, the main general focus and development include the features analysis, the link between CAD and CAE, parametric programs and analysis of the influence of geometric and materials parameters. The research design related to experimental and modelling works were considered within a full product lifecycle from design, manufacturing and testing, to process, control and maintenance. The work is also linked to general consideration and the need for collaborative development, link between data and knowledge to training with long term impact. The main work on electrical resistance spot welding includes the developments of new parametric electrode models in SolidWorks for developing FE models, parametric program development of coupled mechanical-thermal-electrical simulation and data development for analysing the effects of key dimensional parameters and loading conditions. A suite of electrodes model has been developed covering different shapes and design. Two parametric modelling framework have been developed including interactive approach and .inp-file based approaches. A range of cases has been analysed including the effect of electrode shapes, different layers of workpieces, different sample sizes, variable thicknesses, different contact angles and electrode misalignment. The data for temperature distribution and Mises stress at different stages related to critical temperature dependent properties (Stiffness and Yield Strength) is developed. The use of the model for future electrode design and development is also discussed, including the analysis of worn electrodes, electrodes for temperature and phase control. The work on the mechanical deformation modelling of spot welded joints is a typical case of data development for deformation of a complex system. A python based script program in the form .rpy file has been developed which covered the part building, assembly, material and solution. An .inp-file based approach is also developed which allows the parametric model development without the need to use interactive programs and user interface. The overall structure and key procedures of the programs are presented. Some data at critical locations and deformation stages is evaluated to help with the understanding of the material deformation process and it is used to analyse the effect of some key factors, in particular the sheet thickness on the deformation and failure under tensile-shear loading. The results showed that the program can provide data for analysing the fracture process. It shows that both Von Mises and the FEEQ data are important for analysing criterion for fracture initiation. The model was used to analyse data of force and deformation of different types of stainless steels (ferrite, austenite and duplex stainless steels) and the effects of work hardening coefficients. Microstructure modelling and data is an important research area linking modelling to materials, characterisation, testing and manufacturing. The main case presented is welded hardfacings with large M7C3 carbides, including experimental works to study the key features of the material systems and development of robust procedures to process the microstructure with different image processing approaches. Parametric finite element models have been developed to establish the data for a range of carbides, including isolated carbides and carbides in a matrix under different loading conditions. The models were used to establish the data for analysing effects of carbide shapes and internal features on the mechanical behaviour, the stress concentration and other properties (including thermal and electrical properties of the materials and other systems). The integrated experimental-numerical approach was used for other filler-based systems including conductive rubbers, both the experimental works and python programs for predictive modelling of thermal and electrical behaviours and particle distributions are presented.