Co-PiLOT optimizes microstructure design with physics-informed latent space search
Co-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design
Artificial IntelligenceComputational Engineering, Finance, and Science
Summary
Designing materials like metal microstructures is hard because the possible shapes and structures are huge and complicated, and testing each one takes a lot of time. The authors created Co-PiLOT, a way to shrink this complexity into a smaller, understandable space using AI that learns to represent valid designs. Then, they search this smaller space using smart methods guided by physics and uncertainty to find designs that meet specific targets more efficiently. They tested their method on magnesium alloy microstructures and showed it beats other approaches while needing fewer expensive simulations.
What this means in practice
- •For materials engineers: Optimize alloy microstructures for desired properties using fewer costly simulation experiments.
- •For manufacturing process developers: Design high-performance microstructures for metal parts by efficiently exploring feasible structural designs.
Authors
Mahish K. Guru, Mayank Nagar, Ayush vyas, Jan Bohlen, Roland Aydin, Noomane Ben Khalifa
Abstract
Inverse design of physical systems (molecules, devices, microstructures) often reduces to optimizing a high-dimensional structure against an expensive black-box simulator. Direct search is difficult because the space is non-Euclidean, feasibility is hard to encode, and each evaluation is expensive. We present Co-PiLOT, a latent optimization approach that maps candidates through a generative encoder-decoder, uses the decoder as a learned validity prior, and searches the latent space with physics-informed black-box optimization. The framework is applied on the inverse design of magnesium alloy microstructure/texture. We develop a vision transformer based-encoder; paired with latent diffusion, diffusion transformer and rectified-flow transformer-based decoders on $\sim80{,}000$ EBSD-derived microstructure dataset to learn a minimal bottleneck, $z$. The ViT-FMDiT model ($z$=$768$) reconstructs high-fidelity microstructure images (FID $27.86$, MS-SSIM $0.178$), which our self-segmenting orientation codec converts into input grids for crystal plasticity solver. Finally, we introduce MERIDIAN, an active latent optimizer driven by deep-kernel Gaussian-process uncertainty, failure-aware feasibility prediction, manifold-aware trust regions, and target-aware acquisition. Within a budget of $160$ simulations, the ViT-FMDiT and MERIDIAN combination yields the best target-driven objective score, reducing the relative target error by $3$--$22\%$ against seven baselines (DANTE, TuRBO, BAxUS, CMA-ES, DDOM, SEIKO, DDPO) on the same decoder.