An isogeometric approach to streamlining high-end structural dynamics simulation for automotive and other applications

A significant problem in automotive, defense, and other industries is the complexity of setting up and running simulations for various aspects of a design to ensure accuracy, assess failings, and predetermine fatal and non-fatal errors. A high-prole and increasingly important type of simulation is the modeling of crash phenomena, where today's best methods require millions of dollars of manual labor to set up each simulation using the error-prone process of translating Computer-Aided Design (CAD) models to Finite Element Analysis (FEA) models. This process is tedious and frustrating even for seasoned veterans, and after the analysis model is constructed, it often fails to perform robustly, producing low fidelity or erroneous results. This limits how much the simulation can actually be leveraged to improve the design. The revolutionary toolset developed in this SBIR promises to streamline the CAD design/FEA simulation workflow, accelerate simulations by up to 500x, and return more accurate results. For larger customers, these advantages translate into savings of millions of dollars per project. This team's extensive domain experience enables them to deliver fundamentally new approaches to both CAD and FEA, wrapped in a novel multi-user cloud computing framework. Leveraging HPC (high-performance computing) hosted in the cloud, the approach will deliver scalable, customized solutions through an intuitive and simple-to-use web interface. In Phase I, the patented U-spline technology was significantly matured, with promising simulation results. Additionally, the team was able to gain insight from potential customers in the automotive, aerospace, and defense industries regarding the most impactful entry point into their workflows, helping to dene Phase II. This entry point centers around the most demanding aspect of crash simulation: capturing high fidelity structural response to large-scale contact/impact events in the presence of material failure. To provide a cutting-edge solution to this problem, the team will mature the patented U-spline technology, develop new simulation technology based on isogeometric analysis, and parallelize the technology to prepare for cloud deployment. Customer-proposed benchmark problems will be run to verify the advantages of the approach. This toolset will reduce time-to-market and result in better designs across multiple industries. The multi-user, cloud-based interface will allow engineers to collaboratively design and prepare models for analysis. The proposed methodology will remove error-prone parts of the simulation workflow and improve simulation fidelity. This will simplify the user experience so that even non-experts can run analyses. The team has started and exited a company in this space previously and is well prepared to bring the products and company to success.