You are designing a product and trying to improve it. Whether it is minimizing pressure drop, reducing drag, increasing lift or improving the heat transfer rate, you must find small changes to improve performance. Performing shape optimization on any product can be challenging, especially as components get more complex. Understanding exactly how a product works and which parameters you can change to improve it helps, but your insight may only get you so far.
Simulation can reveal shape-optimization opportunities that even experienced engineering analysts can miss. The optimal solution may elude you because you didn’t choose the correct parameters.
Shape optimization can help you find the optimal solution. ANSYS Fluent adjoint solver takes your stated goals and uses them to automatically morph and optimize the geometry. The adjoint solver can optimize the shape of your component and reduce simulation time in many ways, including:
- Finding the best-performing shape
- Automatically morphing the shape
- Performing mesh-morphing quickly and automatically
- Running a minimal number of simulations
- Easily exporting modified mesh back to CAD
Parametric designs are a good way to sample a design space as well, but parametric methods require you to understand the underlying fluid dynamics to select relevant parameters. Adjoint methods can handle more than 1,000 degrees of freedom and don’t require you to specify parameters up front. The chart below compares adjoint methods to parametric designs.
|Adjoint Methods||Parametric Design|
|Typical Use||Find optimal shape for single/multiple
given operating condition
|Find optimal operating condition for
|Shape Optimization||Parameter-free||Define parameters|
|Design Space (Degrees of Freedom)||1,000+||<20|
|Computation Cost||Low to medium||Medium to high|
|Computation Time||Low to medium||Medium to high|