A Paradigm Shift to Assembly-like Finite Element Model Updating

Area of Science:

• Aerospace Engineering
• Computational Mechanics
• Structural Dynamics

Background:

• Finite element models (FEMs) often mismatch real-world structural behavior, a critical issue in aeronautics.
• Increasingly flexible aircraft wings necessitate highly accurate FEMs for design and analysis.
• Traditional iterative FEM updating methods are computationally intensive for complex structures.

Purpose of the Study:

• To introduce and validate a novel, computationally efficient FEM updating approach termed 'assembly-like'.
• To reduce the computational burden associated with iterative FEM updating using surrogate models.
• To compare the efficiency and accuracy of the assembly-like method against the classical global approach.

Main Methods:

• Developed an 'assembly-like' FEM updating framework where models are updated incrementally during assembly.
• Validated the proposed method using experimental data from a flexible aircraft wing.
• Benchmarked the assembly-like approach against the traditional 'one-shot' global FEM updating method.

Main Results:

• The assembly-like method demonstrated approximately 28% lower overall computational effort compared to the global approach.
• Utilized lower-fidelity subassembly models for about 95% of computations, reducing equation counts and memory usage.
• Achieved high fidelity, within 1% of the global approach for natural frequencies and modal shape indices.

Conclusions:

• The assembly-like FEM updating strategy offers significant computational efficiency gains.
• This novel approach maintains the accuracy required for modern aircraft structural analysis.
• The method is particularly beneficial for designs involving complex and flexible structures like advanced aircraft wings.