Automated Structural Optimisation

Automated Structural Optimisation

I Stock -155126807

Stand-alone structural analyses or structural optimisations are powerful tools to verify the structural adequacy of a component or end use article against a set of design requirements. In a traditional, siloed engineering chain these requirements are focused on functionality and certification, and do not consider the overall system architecture or programme management. Therefore, if the mission requirements, project budget, or other major aspects change, the engineering teams must complete another design-analysis-certification cycle. Not surprisingly this adds time, cost and computational expenses.

The approach taken by CFMS is to use a Multi-Objective Optimisation (MOO) that combines traditional structural analysis and optimisation with additional elements of the full product development. When using the MOO any number of variables can be introduced, categorised and given priority to maximise one or more aspects of design and manufacturing, but also tangential characteristics, like the factory layout and emissions when using different production methods. The purpose is to find the best compromise between different goals, from simple structural ones (e.g. weight vs stiffness) to complex multi-disciplinary ones (e.g. stiffness vs cost vs weight vs production lead time, etc). 

For example, when designing an aircraft part, the main goal is for the component to be as lightweight as possible. But lightening a structure comes with a financial cost and shaving off a few additional kilogram of weight considerably increases the manufacturing costs, or even the carbon emissions associated with that manufacturing method. So by controlling the range of each variable, an effective compromise between weight and cost can be achieved. When using this process for all variables, however many they are, the ideal sweetspot that satisfies all preferred requirements can be found.

The method is based on Finite Element Analysis and offers extreme flexibility. Used as a Single-Objective Optimisation (SOO) it can perform a design optimisation element by element. So it can optimise structures in a practical and realistic way, based on the manufacturing capabilities available, reducing the overall component’s mass and design rework.

When used in Multi-Objective mode additional disciplines can be introduced in the FE models to evaluate thermal analysis, electromagnetic conduction, manufacturing processes and defects, CFD. At the same time important aspects of initial design dependent on mission profile (obtained with the Design Optioneering methods) and costs can be included, making it a powerful design tool to automatically generate the desired design in a fraction of time compared to traditional methods.

For example, when using this new approach to improve a wing structure design, MOO analysis identifies key parameters to optimise the design. Through an automated sensitivity study the impact of each parameter to reach a given goal can be evaluated. In the picture below the sensitivity analysis shows how the location of two spars in a simplified wing-box impacts the overall displacement. If the goal is to minimise the deformation, the study identifies the optimum region and the ideal spars location . While this process is visually easy to understand in a 2D representation, it can be executed for as many parameters as required. At the bottom of this page a video with a simplified wing-box study shows how each of the 13 parameters is given a range. Once a goal is formalised (e.g. minimum displacement or minimum cost), the ideal value of all parameters is given.

Graph

MOO also improves awareness of the relative sensitivities in the manufacturing process on the end product quality, which will also reduce both time and cost. Using the same example of a wing structure design, MOO highlights the importance of different parameters. The image below shows how the left and right spar positions will affect wing displacement, which can be used to freeze an initial design early on in the project.

These examples show MOO has a significant impact on how quickly we can get zero emission aircraft in the skies. It also has the potential to accelerate the development of optimised engineering solutions in other industries, including both high-value sectors, such as energy or space, and high volume markets, such as transport.

To find out more about how MOO can improve your engineering performance, contact CFMS.

 

Wing Deformation

Sign Up To Our Newsletter and Events