Enabling solution verification and efficient, high-accuracy simulations through spline-based adaptivity in MOOSE

The need for reliable, clean energy has driven renewed interest and remarkable innovations in the development of fission-based power generation. Unfortunately, as nuclear energy companies design next-generation reactor technology, they are slowed by the lack of accessible and reliable engineering simulation for engineers who are not computational specialists. The current project has verified through market research the Nuclear Regulatory Commissionĺs assertion that solution-error estimation and adaptivity are key to addressing the accessibility of these tools to non-experts. Such technology will automate one of the most error-prone and confusing parts of finite element analysis for non-experts. This project will develop fully automatic solver-driven adaptivity based on isogeometric analysis and integrate it with the Nuclear Regulatory Commissionĺs leading open-source multi-physics code. This improvement and integration will enable non-expert engineers to achieve verified solutions. In Phase II, three significant technological advances were achieved that set up the Phase IIA project for success. First, a simple, robust method was developed to generate analysis-suitable computer-aided design models for isogeometric analysis. A second technical advance allows adaptive refinement of a three-dimensional body-fitted higher-order smooth spline. Finally, a technique was developed that enables adaptive mesh refinement in immersed methods. The next phase of the project will improve the computational efficiency of the advancements made in Phase II. It will also develop a robust integration between the commercial code at issue and an open-source nuclear simulation framework developed by the federal government. The work proposed here will improve the utility of simulation for nuclear engineers by providing improved workflows and methods for building simulation models and by providing better tools to allow engineers to verify simulation results. The benefits of fully automatic adaptivity include reduced development time and cost, improved designs, better understanding of risk, and quantifiable confidence in analysis results and engineering decisions in the nuclear industry. Broad commercial applications and benefits to other segments of the modeling and simulation industry are also likely. Others industries face many of the same problems as nuclear energy when it comes to quantifying accuracy, particularly in the areas of automotive, defense, and additive manufacturing. They will experience the same benefits of faster development, better decision-making, and higher confidence through the technology developed in this proposal.