Presentation is loading. Please wait.

Presentation is loading. Please wait.

G. Cheng, R. Rimmer, H. Wang (Jefferson Lab, Newport News, VA, USA)

Similar presentations


Presentation on theme: "G. Cheng, R. Rimmer, H. Wang (Jefferson Lab, Newport News, VA, USA)"— Presentation transcript:

1 G. Cheng, R. Rimmer, H. Wang (Jefferson Lab, Newport News, VA, USA)
THERMAL ANALYSIS OF SCRF CAVITY COUPLERS USING PARALLEL MULTIPHYSICS TOOL TEM3P* V. Akcelik, L-Q Lee, Z. Li, C-K Ng, and K. Ko (SLAC, Menlo Park, CA, USA) G. Cheng, R. Rimmer, H. Wang (Jefferson Lab, Newport News, VA, USA) Abstract: SLAC has developed a multi-physics simulation code TEM3P for simulating integrated effects of electromagnetic, thermal and structural effects. TEM3P shares the same software infrastructure with SLAC’s parallel finite element electromagnetic codes, thus enabling all physics simulations within a single framework. The finite-element approach allows high-fidelity, high-accuracy simulations and the parallel implementation facilitates large-scale computations with fast turn-around times. In this paper, TEM3P is used to analyze thermal loading at the coupler end of the JLAB SCRF cavity. Thermal simulation of JLab HCCM Cavity A multi-physics simulation tool: TEM3P SLAC has developed a parallel multi-physics simulation tool, TEM3P, for design and analysis of thermal, structural and electromagnetic effects such as cavity wall heating, and Lorentz force detuning simulations. TEM3P shares the same software infrastructure with SLAC finite-element electromagnetic codes. TEM3P enables all multi-physics simulations to be done in a single framework, and provides a complete tool for engineering prototyping. The parallel implementation of TEM3P allows large scale computations on massively parallel supercomputers. Metallic walls for thermal analysis CAD Model EM Analysis Vacuum region for EM analysis Electric field distribution for the accelerating mode Thermal Analysis Challenges of thermal simulations for SCRF Cavity Couplers The thermal simulation of SCRF cavity is a large scale problem due to the small features of cavity walls and couplers. Single processor simulations may take days to converge. The thermal simulation for the SCRF cavity is a strongly nonlinear problem. Thermal conductivities, surface resistances and Kapitza conductance cooling depend on the temperature. HCCM cavity has very thin layers of copper coating. These layers are discretized with shell elements. TEM3P uses inexact Newton method to solve the nonlinear problems. Linear equations are solved with iterative methods. Parallel implementation decreases the simulation time from days to minutes. Temperature distribution from TEM3P Temperature distribution around AlMg seal CAD model includes cavity vacuum region and surrounding metallic walls. Integrated EM and thermal analysis is performed with TEM3P. Nonlinear problem is solved with combination of Picard and inexact Newton method. Linear equations are solved with preconditioned Krylov space methods. Verification and validation is under progress. Convergence studies and parallel speedup Summary Field convergence Temperature convergence studies performed using two meshes, with 1.35 M (coarse mesh) and 4.3 M (fine mesh) tetrahedron elements. Simulations use linear and quadratic finite element discretizations. Largest simulation performed has 7.07 M dofs, uses 256 processors (on NERSC’s Franklin), and converges in 25 minutes, with 12 nonlinear iterations. TEM3P, a parallel multi-physics tool including integrated electromagnetic, thermal and structural effects, allows accurate and fast analysis for cavity design and performance. Nonlinear problem such as that arising from convective cooling at cavity surface is solved efficiently. Further code verification and validation will be carried out for the SCRF cavity in collaboration with JLAB scientists . *Work supported by the U.S. DOE ASCR, BES, and HEP Divisions under contract No. DE-AC02-76SF00515


Download ppt "G. Cheng, R. Rimmer, H. Wang (Jefferson Lab, Newport News, VA, USA)"

Similar presentations


Ads by Google