**1. Introduction**

86 Recent Trends in Processing and Degradation of Aluminium Alloys

Nandan, R., DebRoy, T. & Bhadeshia H.K.D.H. (2008). Recent advances in friction-stir

Ram, G.D.J, Mitra, T.K., Raju, M.K. & Sundaresan, S. (2000). Use of inoculants to refine weld

Rao, K.P., Ramanaiah, N. & Viswanathan, N. (2008). Partially melted zone cracking in

Rhodes, C.G., Mahoney, M.W., Bingel, W.H., Spurling, R.A. & Bampton C.C. (1997). Effects

Sato, Y., Urata, M. & Kokawa H. (2002). Parameters controlling microstructure and hardness

Sato, Y.S. Kokawa, H., Enmoto, M. & Jogan, S. (1999). Precipitation sequence in friction stir

Seto, A., Yoshida, Y. & Galtier, A. (2004). Fatigue properties of arc-welded lap joints with

Thomas, W.M. & Dolby R.E. (2003). Friction Stir Welding Developments, Proceedings of the Sixth International Trends in Welding Research, Materials Park, OH, USA. Thomas, W.M., Nicholas, E.D., Needham, J.C., Murch M.G., Temple-Smith P. & Dawes

*Metallurgical Materials Transactions A*, Vol. 33, No. 3, pp. 625-635.

Vol. 53, pp. 980-1023.

36, No. 1, pp. 69-75.

Vol. 30, No. 12, pp. 3125-3130.

Vol. 22, No. 12, pp. 1147-115.

*PCT/GB92/02203*.

*Science and Engineering A*, Vol. 276, pp. 48-57.

AA6061 welds. *Materials Design*, Vol. 29, pp 179-186.

welding-process, weldment structure and properties. *Progress in Materials Science*,

solidification structure and improve weldability in type 2090 Al-Li alloy. *Materials* 

of friction stir welding on microstructure of 7075 aluminum. *Scripta Materialia*, Vol.

during friction-stir welding of precipitation-hardenable aluminum alloy 6063.

weld of 6063 aluminum during aging. *Metallurgical and Materials Transactions A*,

weld start and end points. *Fatigue and Fracture Engineering Materials and Structures*,

C.J. (1991). Friction Stir Butt Welding. *International Patent Application No.* 

Tailor-welded blanks (TWB) are blanks with sheets of similar or dissimilar thicknesses, materials, coatings welded in a single plane before forming. Applications of TWB include car door inner panel, deck lids, bumper, side frame rails etc. in automotive sector (Kusuda et al., 1997; Pallet & Lark, 2001). Aluminium TWBs are widely used in automotive industries because of their great benefits in reducing weight and manufacturing costs of automotive components leading to decreased vehicle weight, and reduction in fuel consumption. The general benefits of using TWBs in the automotive sector are: (1) weight reduction and hence savings in fuel consumption, (2) distribution of material thickness and properties resulting in part consolidation which results in cost reduction and better quality, stiffness and tolerances, (3) greater flexibility in component design, (4) re-usage of scrap materials to have new stamped products and, (5) improved corrosion resistance and product quality1. The forming behaviour of TWBs is affected by weld conditions viz., weld properties, weld orientation, weld location, thickness difference and strength difference between the sheets (Bhagwan, Kridli, & Friedman, 2003; Chan, Chan, & Lee, 2003). The weld region in a TWB causes serious concerns in formability because of material discontinuity and additional inhomogeneous property distribution. Above said TWB parameters affect the forming behaviour in a synergistic manner and hence it is difficult to design the TWB conditions that can deliver a good stamped product with similar formability as that of un-welded blank. Designers will be greatly benefited if an expert system is available that can deliver forming behaviour of TWB for varied weld and blank conditions. Artificial neural network (ANN) modelling technique is found to show better prediction of any response variable that is influenced by large number of input parameters. Artificial Neural Networks are relatively crude electronic models based on the neural structure of the brain. The building blocks of the neural networks is the neuron, which are highly interconnected. In the artificial neural networks, the neurons are arranged in layers: an input layer, an output layer, and several hidden layers. The nodes of the input layer receive information as input patterns, and then transform the information through the links to other connected nodes layer by layer to the output nodes. The transformation behavior of the network depends on the structure of the

<sup>1</sup> http://www.ulsab.org

Prediction of Tensile and Deep Drawing Behaviour of Aluminium Tailor-Welded Blanks 89

behind the pin is forged under pressure from the dowel and consolidates to form a bond. In a study (Miles et al., 2004), three aluminium alloys: 5182-O, 5754-O, and 6022-T4 were considered and TWBs were made using gas tungsten arc welding process. All three of these alloys are being used to fabricate stamped automotive parts. The gas tungsten arc welding process has been used to make aluminium TWBs industrially, so results using this process were compared to FSW results. The results of tensile and formability tests suggest that the 5xxx series alloys had similar tensile ductility and formability regardless of the welding process. However, the 6022-T4 sheets joined using FSW had better formability than those joined using gas tungsten arc welding because FSW caused less softening in the heataffected zone. Other welding processes like non-vacuum electron beam (NVEB) and Nd:YAG laser techniques have also been studied for welding aluminium TWBs (Shakeri et al., 2002). In that study, a limiting dome height (LDH) test is used to evaluate formability of the AA5754 sheet TWBs with gauge combinations 2 to 1 mm, 1.6 to 1 mm and 2 to 1.6 mm. Different weld orientations were considered and the failures of TWBs were studied. In general the failure occurs in welds and the thinner gauge. Weld orientation has a predominant effect on the formability of the TWBs. In a study (Stasik & Wagoner, 1998), laser welded 6111-T4 and 5754-O blanks were tensile tested, and longitudinal weld TWB's were formability tested using the OSU formability test. In the study, press formability was found to be much greater than the inherent weld ductility. Both materials had satisfactory TWB formability under longitudinal deformation, but 6111-T4 was severely limited under transverse loading, because of a softer heat-affected zone in the heat-treatable alloy. The effects of welds with transverse and longitudinal orientations on the formability of aluminium alloy 5754-O laser welded blanks using the swift cup test has been reported (Cheng et al., 2005). The results showed that longitudinal TWBs underwent considerable reduction in the forming limit when compared with transverse TWBs, and an un-welded blank. Transverse welded blanks exhibit approximately the same forming limit as that of an un-welded blank. However, the effect may change if the weld is placed at critical locations,

Few research groups have aimed at predicting the formability of aluminium welded blanks by using different necking theories and FE simulations. For example, Jie et al. (2007) studied the forming behaviour of 5754-O Al alloy sheets, where in the forming limit curve (FLC) of welded blanks with thickness ratio of 1:1.3 was experimentally evaluated and predicted using localized necking criterion based on vertex theory. It is found from the analysis that the forming limit of the TWB is more closer to thinner material FLC and the experimental and predicted FLCs correlate well with each other. Davies et al. (2000) investigated the limit strains of aluminium alloy TWB (1:2 mm thickness), where in the FLCs predicted by Marciniak-Kuczynski (M-K) analysis are compared with the experimental results. Here the geometrical heterogeneity, i.e., the initial imperfection level, involved in the welded blank is modelled by using the strain-hardening exponent determined from miniature tensile testing together with the Hosford yield criterion, to determine a level of imperfection that exactly fits an failure limit diagram (FLD) to each experimentally evaluated failure strains. These empirically determined initial imperfection levels were statistically analyzed to determine the probability density functions for the level of imperfection that exists in un-welded and welded blanks. Since the imperfection level is not a single value and follows a statistical distribution, a means of selecting a single value for imperfection is formulated. Two different FLCs – namely, failure FLC and safe FLC – were defined. The first FLC was based upon an imperfection that represents a 50 per cent predicted failure rate and was designated

say, at some offset from the centre-line.

network and the weights of the links. A neural network has to go through two phases: training and application. During the learning, the training of a network is done by exposing the network to a group of input and output pairs. In the recognition phase, after it is trained, the network will recognize an untrained pattern. Application of ANN modelling technique in predicting the formability of TWB will definitely be helpful in understanding and designing the TWB conditions that can deliver a better stamped product.

This chapter describes the tensile and deep drawing forming behaviour of aluminium TWBs and prediction of the same using expert system based on ANN models. Standard tensile testing and square cup deep drawing set up are used to simulate the tensile and deep drawing processes respectively using elastic-plastic finite element (FE) method. The sheet base materials considered for the present work is a formable aluminium alloy. Global TWB tensile behaviour like yield strength, ultimate tensile strength, uniform elongation, strainhardening exponent, strength coefficient, limit strain, failure location, minimum thickness, and strain path, and deep drawing behaviour viz., maximum weld line movement, draw depth, maximum punch force, draw-in profile are simulated for a wide range of thickness and strength combinations, weld properties, orientation, and location. Later, ANN models are developed to predict these tensile and deep drawing behaviour of TWBs. ANN models are developed using data set obtained from simulation trails that can predict the tensile and drawing behaviour of TWB within a chosen range of weld and blank conditions. To optimize the training data and thus the number of FE simulation, techniques from design of experiments (DOE) have been used for systematic analyses. The accuracy of ANN prediction was validated with simulation results for chosen intermediate levels. The results obtained are encouraging with acceptable prediction errors. An 'expert system framework' has been proposed by the authors (Veerababu et al., 2009) for the design of TWBs and the study described in this chapter is part of this framework to predict the formability of aluminium TWBs (Abhishek et al., 2011; Veerababu et al., 2009, 2010).
