Quantum-assisted stacking sequence retrieval and laminated composite design
Quantum and quantum-inspired methods are being used to rethink the time-consuming search for optimal stacking sequences for composite materials in aerospace - faster, more efficiently and taking into account complex design and manufacturing requirements.
Laminated composites play a vital role in aerospace structures due to their high specific
strength and design flexibility. A major challenge in their use is the retrieval of stacking
sequences that satisfy mechanical requirements while adhering to complex manufacturing
constraints. This task, typically the second stage of a bi-level optimization process, becomes
computationally demanding due to the combinatorial nature of the design space.
We reformulate stacking sequence retrieval in a quantum computational setting by mapping
discrete stacking configurations onto quantum states. Design objectives are encoded in a cost
Hamiltonian, and manufacturing rules are implemented as constraint penalties. This enables
the use of ground-state search techniques, including variational quantum algorithms such as
the Filtering Variational Quantum Eigensolver (F-VQE), and quantum-inspired methods such as
the Density Matrix Renormalization Group (DMRG).
We demonstrate the approach on benchmark problems with up to 200 plies and show that both
F-VQE and DMRG can produce physically valid stacking sequences under standard aerospace
constraints. The formulation accommodates multiple design objectives, including lamination
parameter matching, buckling resistance, and control over ply-angle clustering. DMRG shows
competitive performance with established classical methods such as LAYLA and Opti-BLESS,
highlighting the potential of quantum and quantum-inspired solvers for scalable, constraint aware composite design.
Presentation languange: ENG
Speakers (1)
