You are here

Automated Geometry and Mesh Updates for Accelerator Shape Optimization

Award Information
Agency: Department of Energy
Branch: N/A
Contract: DE-SC0018715
Agency Tracking Number: 247166
Amount: $999,515.00
Phase: Phase II
Program: SBIR
Solicitation Topic Code: 24b
Solicitation Number: DE-FOA-0001976
Timeline
Solicitation Year: 2019
Award Year: 2019
Award Start Date (Proposal Award Date): 2019-08-19
Award End Date (Contract End Date): 2021-08-18
Small Business Information
10 Executive Park Drive
Clifton Park, NY 12065-5630
United States
DUNS: 012076795
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Ottmar Klaas
 (518) 348-1639
 oklaas@simmetrix.com
Business Contact
 Mark Beall
Phone: (518) 348-1639
Email: mbeall@simmetrix.com
Research Institution
N/A
Abstract

Accelerators are complex scientific instruments requiring significant resources to design and build.At the largest scale, accelerators for use in physics discovery are billion-dollar-class facilities that take decades to develop.Medical accelerators can cost millions of dollars.Other electromagnetic applications include threat detection, antenna design, wireless device design, and cancer treatment.Simulation codes have been developed for these systems which include the ability to determine the optimal shape for components using advanced optimization techniques.However, the current optimization process is not suitable for general usage due to limitations on how it implements geometry and mesh updates needed for the simulation.By removing these limitations, this project will make these powerful design tools available to a wide audience.This project will develop software components that enable the fully automatic geometry optimization of accelerator and other systems.Specific capabilities to be developed will provide the ability to: a) define shape optimization parameters on an existing geometric model of a component and update the geometry based on changes to those parameters, b) automatically update the computational mesh based on changes to the geometric model shape, c) integrate with the simulation codes and optimization process to provide fully automatic shape optimization capabilities.Implementation of additional geometry update functionality to support typical accelerator geometry was done and linked to new procedures for local mesh updates based on mesh motion and local remeshing.The ability to calculated needed derivative information for the optimizer was implemented, interfaced with the solver and optimizer, and demonstrated on an example accelerator geometry.The geometry and meshing capabilities will continue to be advanced to address more complex geometries and to support operation in a parallel computing environment.The integration with the optimizer/solver will be fully implemented and optimized.Highly complex, real world problems will be used to demonstrate the resulting capabilites.The project will provide industry, government labs and universities an effective means to take full advantage of existing parallel simulation software.The development and introduction of these geometry and meshing technologies will allow industry to broadly integrate high performance parallel simulations into their design processes.Truly automatic shape optimization procedures will enable more efficient and higher performance designs to be quickly realized, providing savings in both design and operation.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government