
Auto Generated Wheel Design using Topology Optimization
Shashwat Agarwal, MCEN-5228-013, Automated Mechanical Design, Spring 2018, CU Boulder

ABOUT SKYLINE

The latest Shape memory alloy (Ni-Ti) Alloy based Flexible & Airless wheel design for futuristic Rovers.
Courtesy: Reinventing the Wheel, NASA Glenn Research Centre
Such designs are being pursued by most of the wheel manufacturers all across the world due to their inevitable advantages over their pneumatic counterparts. However, existing CAD & parametric design approaches are not the most efficient ways to model such interwoven designs. Self-generative design approaches that are optimized for load constraints come up with such interesting patterns by itself which is the major motivation for the Automated Design Algorithms
Gone are the days when a product designer has to go through the repeat cycles of design & Analysis to ensure the optimized safety factor without being redundant in cost, strength or material. As the part of the final class project for the new course, Automated Mechanical Design under Prof. Robert MacCurdy, the project to topologically optimize a wheel disc was successfully completed by Shashwat Agarwal. The project aimed at auto-optimization of the simplified Topology of a wheel as per the usual load constraints under various different criteria like Minimum Stiffness, Max Strength & Min. Volume.
Using a novel, shape optimization plugin, Pareto works that was developed by a Prof. Suresh & his research team at the University of Wisconsin Madison, it deploys a SIMP based approach through Topological sensitivity which essentially helps in finding an optimum solution set with the much lesser number of Finite element analysis. It also uses the Tangled FEA to replace the negative Jacobian mesh elements with geometrically convex Quads. Using this Modified SIMP approach the Auto shape optimization results were more precise & a lot quicker. The results were also validated & the prototype was 3d printed.
The entire cycle was repeated with 2 different materials. Properties of TPU were modeled to imitate a soft material whereas ABS was selected to optimize a wheel design made of stiffer/harder material. Attributes like quick processing time, accurate & verifiable results make the proposed process, the most suited scheme of shape optimization for the fast-paced industrial projects. Also, it demonstrates the inevitable advantages of the futuristic flexible airless wheel designs being pursued by the tyre manufacturers worldwide.


Initial Design
Design Validation

Auto-Generated Shape Optimized Design

Printed Prototype: ABS

Printed Prototype: Ninja-Flex
