Japan’s Kanazawa University researchers have developed a new simulation method to advance the automotive metal sheet stamping process.
The new method can be used to optimize a metal stamping press in its conceptual design stage, which could in turn help manufacturers reduce the costs of physically trialing new designs.
According to the researchers, the new method is not only cost-effective, but also more comprehensive than past simulation methods.
Auto makers have been testing and using lightweight materials to improve the fuel consumption of vehicles.
One possible alternative is high-strength steel. However, when such sheets are stamped into shape to fabricate car parts, they tend to bend, tear, wrinkle, or become too thin.
Simulations have today become an integral part of advancing or optimizing new tools before they can be built or tested.
This is essential because new tools need to be altered over a lengthy and costly trial-and-error period until they can successfully fabricate high-quality parts.
Many components of the tool have an effect on the final product and could therefore be optimized via simulations.
Sadly, current simulations are not comprehensive enough, and rarely consider the shape of the stamping stencil that the metal sheet is punched through to form the desired shape.
Moreover, research in this area has been solely focused on stamping simple bar or U-shapes.
All this is now bound to change thanks to Kanazawa’s new simulation method.
“We simulated the stamping of S-shapes into sheet metal. Unlike U-shapes, the stamping of S-shapes can cause the metal parts to twist out of shape, allowing us to study ways of reducing twisting springback,” says study co-author Ryoto Ishizuki.
The researchers found a way to reduce the twisting of metal sheets by optimizing the shape of the blank shape while also minimizing tearing and wrinkling of the metal sheet.
They also simulated how much force to use to clamp the metal sheet in place in the so-called ‘blank holder’ and how it should be varied during the punching process to achieve the best results.
According to first author Satoshi Kitayama, sequential approximate optimization using a radial basis function network enabled them to efficiently optimize the blank shape and variable blank holder force trajectory.
Image and content: Wikicommons-OSX/Kanazawa University
