Using the solid-shell element to model the roll forming of large radii profiles

Roll forming is an incremental bending process for forming metal sheet, strip or coiled stock. Although Finite Element Analysis (FEA) is a standard tool for metal forming simulation, it is only now being increasingly used for the analysis of the roll forming process. This is because of the excessive computational time due to the long strip length and the multiple numbers of stands that have to be modelled. Typically a single solid element is used through the thickness of the sheet for roll forming simulations. Recent investigations have shown that residual stresses introduced during steel processing may affect the roll forming process and therefore need to be included in roll forming simulations. These residual stresses vary in intensity through the thickness and this cannot be accounted for by using only one solid element through the material thickness, in this work a solid-shell element with an arbitrary number of integration points has been used to simulate the roll forming process. The system modelled is that of roll forming a V-channel with dual phase DP780 sheet steel. In addition, the influence of other modelling parameters, such as friction, on CPU time is further investigated. The numerical results are compared to experimental data and a good correlation has been observed. Additionally the numerical results show that the CPU time is reduced in the model without friction and that considering friction does not have a significant effect on springback prediction in the numerical analysis of the roll forming process.

Numerical tools for isogeometric analysis in the nonlinear regime

The present work deals with the development of robust numerical tools for Isogeometric Analysis suitable for problems of solid mechanics in the nonlinear regime. To that end, a new solid-shell element, based on the Assumed Natural Strain method, is proposed for the analysis of thin shell-like structures. The formulation is extensively validated using a set of well-known benchmark problems available in the literature, in both linear and nonlinear (geometric and material) regimes. It is also proposed an alternative formulation which is focused on the alleviation of the volumetric locking pathology in linear elastic problems. In addition, an introductory study in the field of contact mechanics, in the context of Isogeometric Analysis, is also presented, with special focus on the implementation of a the Point-to-Segment algorithm. All the methodologies presented in the current work were implemented in a in-house code, together with several pre- and post-processing tools. In addition, user subroutines for the commercial software Abaqus were also implemented.

Development of an inverse routine to predict residual stresses in the material based on a bending test

Bending and reverse bending are the dominant material deformations in roll forming, and hence property data derived from bend tests could be more relevant than tensile test data for numerical simulation of a roll forming process. Recent investigations have shown that residual stresses change the material behavior close to the yield in a bending test. So, residual stresses introduced during prior steel processing operations may affect the roll forming process, and therefore they need to be included in roll forming simulations to achieve improved model accuracy. Measuring the residual stress profile experimentally is time consuming and has limited accuracy while analytical models that are available require detailed information about the pre-processing conditions that is generally not available for roll forming materials. The main goal of this study is to develop an inverse routine that determines a residual stress profile through the material thickness based on experimental pure bend test data. A numerical model of the skin passing (temper rolling) process is performed to introduce a residual stress profile in DP780 steel sheet. The skin passed strips are used in a pure bending simulation to record moment-curvature data and this data is then applied in an inverse analysis to predict the residual stress profile in the material. Comparison of the residual stress profile predicted by the inverse routine with that calculated by finite element analysis (FEA) indicates an inverse approach combined with pure bend test may present an alternative to predict residual stresses in sheet metals.

Springback investigation in roll forming of a V-section

To have fuel efficient vehicles with a lightweight structure, the use of High Strength Steels (HSS) and Advanced High Strength Steels (AHSS) in the body of automobiles is increasing. Roll forming is used widely to form AHSS materials. Roll forming is a continuous process in which a flat strip is shaped to the desired profile by passing through numerous sets of rolls. Formability and springback are two major concerns in the roll forming of AHSS materials. Previous studies have shown that the elastic modulus (Young's modulus) of AHSS materials can change when the material undergoes plastic deformation and the main goal of this study is to numerically investigate the effect of a change in elastic modulus during forming on springback in roll forming. Experimental loading-unloading tests have been performed to obtain the material properties of TRIP 700 steel and incorporate those in the material model used in the numerical simulation of the roll forming process. The finite element simulations were carried out using MSC-Marc and two different element types, a shell element and a solid-shell element, were investigated. The results show that the elastic modulus diminution due to plastic strain increases the springback angle by about 60% in the simple V-section roll forming analyzed in this study. © (2014) Trans Tech Publications, Switzerland.

The influence of residual stress on a roll forming process

Roll forming is increasingly used in the automotive industry to form High Strength Steel (HSS) and Advanced High Strength Steel (AHSS) for structural components. Because of the large variety of applications of roll forming in the industry, Finite Element Analysis (FEA) is increasingly utilized for roll forming process design. Bending is the dominant deformation mode in roll forming and sheet materials used in the process are often temper rolled (skin passed), roller- or tension-levelled. These processes introduce residual stresses into the material, and recent studies have shown that those affect the material behaviour in bending. A thickness reduction rolling process available at Deakin that leads to material deformation similar to an industrial temper rolling operation was used in this study to introduce residual stresses into a dual phase, DP780, steel strip. The initial and thickness reduced strips were then used in a 5-stand experimental V-section roll forming set-up to identify the effect of residual stress on the final shape. The influence of residual stress and the effect of plastic deformation on the material behaviour in roll forming are separately determined in numerical simulation. The results show that the thickness reduction rolling process decreases the maximum bow height while the springback angle and end flare increase. Comparison with experimental results shows that using material data from the conventional tensile test in a numerical simulation does not allow for the accurate prediction of shape defects in a roll forming process if a residual stress profile exists in the material. On the other hand including the residual stress information leads to improved model accuracy.

Impact of finite element idealisation on the prediction of welded fuselage stiffened panel buckling

Lap joints are widely used in the manufacture of stiffened panels and influence local panel sub-component stability, defining buckling unit dimensions and boundary conditions. Using the Finite Element method it is possible to model joints in great detail and predict panel buckling behaviour with accuracy. However, when modelling large panel structures such detailed analysis becomes computationally expensive. Moreover, the impact of local behaviour on global panel performance may reduce as the scale of the modelled structure increases. Thus this study presents coupled computational and experimental analysis, aimed at developing relationships between modelling fidelity and the size of the modelled structure, when the global static load to cause initial buckling is the required analysis output. Small, medium and large specimens representing welded lap-joined fuselage panel structure are examined. Two element types, shell and solid-shell, are employed to model each specimen, highlighting the impact of idealisation on the prediction of welded stiffened panel initial skin buckling.