Surrogate modeling to speed up structural optimization

Industry: Aerospace | Product: pSeven Core | Company: Airbus

Structural optimization Airbus by MACROS: Start

Objective

Numerous calculations of mechanical strength criteria, used as constraints in the considered optimization problem, are necessary for local analyses, resulting in a great increase of the time between two iterations. Computed mechanical strength criteria are noisy and discontinuous which causes problems for any optimizer. Structure optimization computational time on average is equal to several days. The objective was to significantly reduce this time using surrogate modeling.

Challenge

Aircraft structural components (wing, fuselage, tail) optimization is very computationally intensive since it requires at each iteration a two-level process:

  1. From the previous iteration, an update step at full component level must be performed in order to take into account internal loads and their sensitivities in the whole structure involved by changes in local geometry.
  2. Numerous local analyses are run on isolated elements (for example, super stiffeners) of structural components in order to calculate mechanical strength criteria and their sensitivities, depending on current internal loads. An optimization step is then performed from combined global-local sensitivities. This bi-level global-local optimization process is then repeated until convergence of load distribution in the whole structure.

Solution

The approach is to apply surrogate-based optimization. DATADVANCE developed a unique technology of surrogate model construction and optimization based on two generic tools (GT Approx and GT Opt). In addition, it is important to say that this technology was applied to construct surrogate models of optimization constraints (mechanical strength criteria) in order to speed up structural optimization process.

Benefit

It turned out that pSeven Core surrogate-based optimization allows obtaining smoother convergence to a reasonable solution in significantly fewer iterations with a smoother distribution of thickness/stringer dimensions and provides the reduction of structure optimization computational time from several days to a few hours.

Structural optimization Airbus by MACROS: Copti-X final