This presentation was made at NAFEMS UK Conference 2018, "Taking Engineering Analysis and Simulation to the Next Level".
The NAFEMS UK Conference 2018 brought together all those involved in analysis and simulation from every corner of industry and academia, giving them an opportunity to advance their knowledge, give their organisations a competitive advantage, and a chance to be part of improving the technology itself.
An overview will be given of a python library called simrev, used for Simulation Revision and management. Simrev enables the hierarchy of an engineered system to be mirrored to a python program; mapping parts, subassemblies, and assemblies; to classes, modules, and submodules. Simrev facilitates geometry specification, meshing (with Cubit/Trelis), analysis keyword specification, material properties, boundary conditions, and interface definitions, analysis batch job submission, and results postprocessing.
We will discuss benefits to collaboration and lessons learned from putting such a program under version control (using git), tracking and organizing team members’ contributions on Atlassian BitBucket (similar to github) and organizing agile development for analysis tasks and design objectives with Atlassian Jira.
Building the “digital-twin” is a process of continuous refinement to engineering models; the result of hundreds or thousands of validated analyses using CFD, FEA, and other engineering or physics codes. The simrev processor is a patent pending method that connects high performance computing simulation results (potential order of TB data) to the state of a code repository (usually order MB) in a way that is secure and traceable 1-1 throughout the product lifecycle.
Simrev captures the capital investment going into simulation with a “software-twin.” By applying a wide range of existing enterprise management tools for software development, the software-twin can move freely, efficiently, and automatically through the enterprise instead of remaining trapped in design silos like individual analysis results. Enterprise management tools provide a meeting space for domain experts and management, keeping them “on the same page” as the design is matured, design objectives are achieved, and risk is retired. Effective collaboration is the key to shortening time to market and meeting customer requirements with simulation, especially for systems with tight coupling between a diverse set of physics domains, e.g. aerospace, defense, and electronics.
Simrev has been in use at LLNL for the past several years. It has been used to perform coupled simulations with the implicit thermo-structural code Diablo (with heat flux and pressure applied from CFD analyses for multiple load cases), the implicit FEM code NIKE3D, the explicit FEM code Paradyn, and the implicit/explicit Multiphysics Arbitrary Eulerian-Lagrangian code ALE3D (with domain mesh coupling). (ALE3D is becoming commercially available as a product called ALE3D for Industry, or ALE3D4I).
Simrev is currently configured for use on LLNL HPC systems. However, this talk will propose future workflows involving other concepts from software including; “continuous deployment,” cloud-based HPC analysis spawned by push updates; augmenting project manpower with programmers who are not engineering experts; automated system requirement checks through “unit tests” of subassemblies; Multiphysics “integration tests” of full systems with massively parallel HPC; and automated reporting of physical requirements, e.g. exceeding stress, temperature, or noise constraints, for automated reporting of risk burn-down to decision makers.
Simrev as a code library, and associated IP, is available for license.
This document was prepared as an account of work sponsored by an agency of the United States government. Neither the United States government nor Lawrence Livermore National Security, LLC, nor any of their employees makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States government or Lawrence Livermore National Security, LLC. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States government or Lawrence Livermore National Security, LLC, and shall not be used for advertising or product endorsement purposes.
Lawrence Livermore National Laboratory is operated by Lawrence Livermore National Security, LLC, for the U.S. Department of Energy, National Nuclear Security Administration under Contract DE-AC52-07NA27344.