9 October 2018 in IEEE Transactions on Antennas and Propagation
An embedded domain decomposition method (DDM) is presented herein for solving electromagnetic problems with complex geometries. In the method, the original computational domain is decomposed into a background subdomain and multiple embedded subdomains. The subdomain problems are easier to solve than the original problem. Furthermore, the shape of the subdomains can be geometrically non-conformal, and the discretizations in the subdomains are allowed to be completely independent. Information exchange between the background and embedded subdomains are addressed in their shared regions, including four ingredients: field continuity, material difference, perfect electrical conductor (PEC), and port. Field continuity on the boundaries of the embedded subdomains is weakly enforced by employing Robin transmission condition. Modified volume sources are introduced to account for material difference between the subdomains. For PEC and port, surface currents on them in embedded subdomains are impressed into the background subdomain in a proper way. The accuracy and flexibility of the proposed method are demonstrated through several numerical examples.
25 October 2018 in IEEE Transactions on Microwave Theory and Techniques
In integrated circuit designs, adjustments and replacements of components are frequently required for analyzing or improving signal integrity. However, the modifications of geometry will cause remeshings and resimulations in the full-wave analysis. Consequently, the design efficiency can be impaired due to repeated discretizations. In this paper, we introduce an embedded domain decomposition approach to address such repetitive simulations. A problem is first decomposed into a simpler background subdomain and multiple embedded subdomains. The geometrical details of interest can be placed into the embedded subdomains while the background subdomain covers the rest. Afterward, the communications between the subdomains are built by imposing coupling sources that represent field continuities, material polarizations, and surface currents on perfect electric conductor and port. The meshes between the subdomains are completely nonconformal, and the modification of an embedded subdomain will not affect the geometries and discretizations in the other subdomains. Such flexibility can be exploited in simulating complicated 3-D printed circuit board structures or investigating different components in an electronic system, as illustrated in the numerical examples.
8 February 2024 in Journal of Computational Physics
21 November 2023 in IEEE Transactions on Microwave Theory and Techniques
Domain decomposition methods (DDMs) provide a powerful discretization framework for analyzing the electromagnetic (EM) phenomena in complex high-speed circuits. However, even with recent advancements in DDMs, achieving iterative convergence remains challenging for sophisticated structures. To facilitate the convergence of DDMs, we develop a preconditioner in this work, which is constructed by forming the approximate inverse of the DDM system matrix. The subdomains are recursively grouped into a multilevel tree, and randomized matrix compressions are adapted to handle their couplings. The approximate inverse preconditioner is then constituted by the inverses of the subdomain matrices with low-rank updates that account for their couplings. Through simulations of practical circuit examples, the strengths of the preconditioner are demonstrated by comparing it to a standard DDM solver. Importantly, a weakly convergent scenario is identified for DDMs when circuit signals propagate parallel to the subdomain interfaces. This situation, which significantly hinders computational efficiency and causes stability issues, can be effectively addressed through the proposed preconditioning approach. Finally, a novel integration is introduced to combine our method with a recently developed subdomain compression scheme.
17 July 2024 in IEEE Transactions on Antennas and Propagation
Nonoverlapping and nonconformal domain decomposition methods (DDMs) have established an effective discretization framework for complex electromagnetic applications, where global iterative solvers are commonly employed to tackle DDM equations. However, DDMs are not free from convergence issues, and excessive iteration numbers can significantly prolong the duration of simulations involving multiple right-hand sides. This study explores the possibility in developing a low-rank direct solver for nonconformal DDMs, with the goal of offering a stable and reliable alternative to DDM iterative approaches. Our direct solver consists of two major components: a subdomain solver and a DDM solver. The subdomain solver is developed by extracting an interface skeleton from subdomains, achieved through recursive partitioning and merging of mesh elements. The DDM solver builds upon the skeletonization, and is constructed through recursive reorderings of interface matrices with the integration of hierarchical matrix techniques and randomized matrix compressions. Both solvers have drawn inspirations from existing mathematical studies but also possess distinct features due to unique matrix structures in DDMs. We validate and demonstrate the effectiveness of our solver through several numerical examples, showcasing its capability to stably handle DDM systems and its efficiency in scenarios involving multiple right-hand sides.
18 June 2024 in IEEE Transactions on Electromagnetic Compatibility
An electromagnetic (EM)-circuit cosimulation approach is presented for evaluating the vulnerability of integrated circuits (ICs) when exposed to external EM waves. The approach comprises frequency-domain simulations of an EM subsystem, time-domain simulations of a circuit subsystem, and an interface scheme that integrates the two subsystems. To address the multiscale geometric complexities in intricate circuit designs, a nonconformal domain decomposition method is employed for efficient mesh generation and numerical computations. For capturing the EM subsystem's behavior over a wide-frequency band, an adaptive interpolation scheme, built upon Loewner matrix technique, is developed for fast frequency sweep. The simulated and interpolated behavior of the EM subsystem is integrated with IC models and subsequently processed with time-domain analyzes. To demonstrate the effectiveness of the method and assess the EM vulnerability of a practical IC design, numerical simulations and laboratory experiments are conducted utilizing near-field antenna and far-field wave sources.