Development of Machine Learning Informed Design Optimization for Double Shell Capsule Graded Density Inner Layer Targets

Advances in machine learning (ML) have the ability to reduce design costs and enhance the design process in HED experiments. ML can use a minimal number of expensive simulations to search the design space efficiently for optimal designs of ICF experiments. This work focuses on optimizing graded density inner shells of indirectly driven double shell targets (Phys. Plasmas, 26, 052702 (2019)), while reducing hydrodynamic instability and maintaining high yield. Graded layer inner shell targets have a parameter space that is too large to fully map out, which is why efficient exploration of the design space is not only beneficial but also necessary. ML methods use predictive physics simulations to identify graded layer designs with high predicted performance as well as novel designs with high uncertainty in performance that may hold unexpected promise. We apply Bayesian optimization to the design optimization of double shell graded inner shell targets using physics models of varying fidelities. By applying ML tools to design simulations, we aim to optimize target geometry to mitigate hydrodynamic instability and improve yield. We present our progress on this new ML informed design tool for future application to NIF double shell experiments with graded layer inner shell targets.