Mechanistic Learning Based Surrogate-Aided Optimisation: Application to Hydrocephalus Shunt Systems

Abstract Generative design combines computational models with numerical optimisation to enhance design performance. In many cases, e.g., in clinical device design, evaluating designs prior experimental testing typically requires high-fidelity multi-physics models that may involve fluid–structure interaction (FSI), multiple spatial and temporal scales, and dependencies on open systems. Such simulations can take hours or days making them impractical for iterative optimisations. Surrogate models that provide rapid assessments are therefore essential. Traditional data-driven surrogates are fast but offer limited interpretability, while reduced-order models capture physical mechanisms but depend on simplifying assumptions and prior knowledge. Mechanistic learning (MxL) addresses these limitations by combining machine learning (ML) with reduced-order modelling to create a surrogate that retains mechanistic insight while achieving high computational efficiency. Here, we propose to integrate the MxL based surrogates into a generative design protocol. We then showcase this methodology in the design of hydrocephalus shunt systems. Starting from a full 3D finite element FSI model, we motivate the use of a fast surrogate model in which fluid stresses on the FSI boundary are combined with ML techniques to predict solid deformation. The resulting MxL surrogate is used in a recently proposed hybrid optimisation algorithm to propose new designs at a fraction of the cost of traditional optimisation methods. The resulting designs are ultimately tested in the full FSI model to confirm their improved efficacy.