TURBOdesign Shaper

Basic features of Shaper

Shaper optimizes complex geometries with respect to given targets, such as total pressure loss and velocity uniformity. It does so by computing the sensitivities of the geometry itself versus those targets and then modifying it. The sensitivity information comes directly from the flow field so the optimized shape is the one that fits the given flow best. Also, the computations related with such a process are independent from the number of design variables, delivering the sensitivities or optimized shape at minimum cost.

Shaper versus existing design methods

Although several industrially established methods exist for the optimization of relatively simple geometries (i.e. geometries that can be controlled via a very small number of design variables), there is none that can optimize complex geometries. That is because complex geometries need a high number of design variables so as to be controlled and that would be computationally prohibitive with existing methods. Therefore engineers rely on experience and trial-error experiments to modify complex shapes until now, although the potential for performance improvement from such parts is big.

Competitive advantages of Shaper

Sensitivity computation & optimization of complex geometries

Complex geometries can be examined for sensitivity analysis and optimization, providing designs with improved performance.

Computational cost independent from complexity

No matter how complex the geometry to examine is, the cost of the computations will always be independent from the number of design variables.

Straightforward setup of design

No time consuming setup of design parameters. They will be automatically recognized from the boundaries.

Innovative design

With no constraint on the number of design variables, the design space is larger than ever before. New shapes and ideas can be generated in a matter of minutes or hours.

Significant manpower and design time savings

Instead of expensive trial and error, an accurate mathematical approach drives the design to the best shape.

Integration with existing workflows

Create meshes and post process results in any of the major packages in the market.

Examples

Automatic design variables

All mesh nodes laying on boundaries that have been selected for design will automatically be considered as design variables. These could be hundreds or thousands but the cost will remain the same. This approach makes it easy and fast for the user to set up.

Figure 1: All the nodes on the (red) boundary are design variables

Sensitivity maps

One of the functions of Shaper is to compute the sensitivity of the design surfaces for the given cost functions. These can give insightful information to the engineers on what parts of a given geometry should be modified and how. This accelerates the design process significantly. The sensitivities can be presented in scalar or vector form.

Figure 2: Sensitivities in vector form

Optimization

Shaper can also optimize the geometry and provide improvements in just a few iterations. This eliminates manual changes and drives the shape to a better design mathematically. Optimization can be performed using a single or multiple cost functions.