| Abstract Scope |
We discuss a new tool that use physics-informed AI to efficiently solve complex inverse problems, and automate the generation of high-fidelity surrogates for a physically-relevant response surface of interest. By placing all constraining information as transforms that restrict the search space to that of only valid solutions, we simplify the optimization problem and facilitate automation. Thus, by incorporating any certification and validation requirements as constraints, we see dramatic increase in the efficiency and automation of the solution of engineering design problems. We recently utilized this approach to automate Rietveld refinement, reducing the time-to-solution for textures from months to minutes, enabling in situ steering of diffraction experiments. We have applied our framework to the real-time engineering design of quantum sensors and the optimal design of mixture models for additive manufacturing, and have seen a similar orders-of-magnitude reductions in the time-to-solution without loss of fidelity. |