MSC Apex Generative Design

Every year a team of young students from the UPBracing Team develops a FormulaStudent racing car to compete in international competitions. Two factors are essential for a successful racing car: light weight and strong components. In addition to the races in which drivers and racing cars prove their performance, there is also an extra rating for the lightweight construction of the racing car. Due to its fourfold use, the wheel carrier is ideal for saving weight while meeting high load-bearing requirements.

Conventional design for milling, due to the very complex shape the parts weighs 515g but produces 15.5kg aluminum waste

Optimized lightweight design in only 8 hours

Smart Generative Design for tailored 3D Printing Structures

In previous years, the teams had tried to design light weight wheel carriers with complex milling designs. The wheel carrier in 2012 with a traditional, already complex, milling design weighed 515g. It was milled from a 16 kg block of material thus producing 15.5 kg of aluminum waste. For the 2018 season, MSC Apex Generative Design technology was applied to create an optimal lightweight design. Design and non-design areas were defined in the model design and the complex load cases were added. The mesh was then automatically generated and optimized. The algorithm generated the finished smoothed result on a workstation with the high-end NVIDIA Tesla P100 graphic card in 6 hours (which would take approximately 14 hours with a solid workstation GPU NVIDIA QUADRO P5000). The result was single wheel carrier of with a weigh of 266g - a reduction of 48% compared to the 2012 equivalent. There was also no significant waste of material because the part was produced by additive manufacturing. This contributed considerably to the overall weight reduction of the race car and gained a lot of attention and a very good rating within the design report, a special category in the overall competition.

A popular example for testing the optimization quality of different algorithms is the "GE Jet Engine Bracket". This bracket of an aircraft turbine was published on the GrabCAD website by General Electric in 2013 as a design challenge for additive manufacturing. The challenge was to redesign the very simple component design using any method, and to develop a component that was as light as possible while taking the given boundary conditions into account. The Challenge attracted a great deal of attention and is still frequently used today to demonstrate the strengths of new algorithms.

Three different designs for the Jet Engine Bracket Optimization by MSC Apex Generative Design created in very short time

Choosing the most promising Design from various Variants
The same component with the same loads was calculated by MSC Apex Generative Design, with the results shown above. They were each calculated on a CAD workstation with two Nvidia P5000 graphics cards in less than an hour. The result reaches from an optimally-shaped, thin, multi-curved shell with thin additional struts for the transmission of the highest loads to a strut only design. Especially the shell design is very stable, particularly light and easy to produce with additive processes because hardly any support structure is required. It can only be achieved with a particularly high calculation resolution as MSC Apex Generative Design is capable of calculating while still being super fast. The automated generative design from MSC Apex Generative Design shown here is even lighter and has lower stress than other solutions. A weight of only about 140g could be achieved with a maximum stress of 680 MPa (target stress 700MPa).