Abstract: Mechanical systems often involve multi-physics interactions and complex nonlinearities due to which design optimization become challenging. The inherent complexity of the processes, along with limitations in data availability, mandate principled uncertainty estimates in modeling. Budget restrictions pose serious data limitations in the development and testing phases of the product pipeline, particularly in the presence of a hierarchy in the associated multi-physics process models with respect to their fidelity levels.
This talk will focus on some fundamental developments and applications of probabilistic machine learning strategies in design optimization. From the perspective of statistical learning, the advantages of MFM-based optimization over a single high-fidelity surrogate, specifically under complex constraints, will be discussed with benchmark optimization problems involving noisy data. Novel multi-fidelity surrogate modeling and optimization strategies will be discussed with respect to data-driven engineering design optimization problems, for example, process parameter optimization in additive manufacturing. The proposed framework is expected to accelerate design optimization tasks involving expensive simulations particularly in data scarce regimes.