Answers are grounded in this paper and cite their sources.
Purpose The DC bus bar is an essential component in electric vehicles (EVs), as it is used for high-current-density power supply between modules. This study presents a shared memory-based parallel, object-oriented C++ framework for fully coupled thermo-electric multiphysics topology optimization of DC bus bars. Design/methodology/approach A density-based topology optimization approach minimizes thermal compliance while satisfying constraints. The governing thermal and electrical equations are discretized using the finite element method, and the method of moving asymptotes is employed to update the material layout. To address the computational expense associated with fine mesh resolutions and coupled physics, we perform a detailed time analysis of the serial program to identify performance bottlenecks. An OpenMP-based shared-memory parallelization technique has been applied in the program, including element-level matrix computations and the preconditioned conjugate gradient solver. Performance of parallelized code is evaluated using multiple geometric configurations and mesh sizes. Findings The maximum parallel speedup of 10.64 is achieved using 32 threads, and the maximum parallel efficiency of 94% is observed with two threads. Originality/value The framework demonstrates robust scalability and provides a practical tool for high-resolution topology optimization of thermal-electrical power components in EVs.
No takes yet. Share an insight, caveat, or question.
Sinha et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: