New paradigms for rigorous modeling and fast solution of large-scale EMC problems

自然或故意产生的电磁场（EM）以组件和系统故障或最终故障的形式表现出其不利影响。关键基础设施，特别是电力系统，必须确保在受到不同来源（如地磁暴和闪电回击电流）产生的电磁场影响时连续可靠地运行。雷电流及其相关电磁场（也称为雷电电磁脉冲（LEMP））对电网（电力系统、通信站点和控制中心）的有害影响早已为人们所知。关于地磁场，一个现实的例子是1989年魁北克水电电力系统的停电，这是由于一个大的太阳活动。评估电磁场对电力网络、通信系统、输油管道、电气化运输系统等的不利影响是至关重要的。上述电磁场及其不利影响的准确量化被视为一个涉及复杂传播介质的大规模电磁兼容（EMC）问题，即不均匀、分散的地面和不规则的地形。这种大规模的EMC问题的准确建模是超出了现有的计算包的能力，主要是由于其计算性能的限制和建模工作的关注。该提案的目的是开发严格和高性能的计算方法，用于有效建模和解决严格的大规模EMC问题。计算平台可容纳先进版本的电磁学数值方法，如时域有限差分法和矩量法，为大型多尺度EMC问题设计和定制。因此，拟议的研究计划旨在开发：（1）统一的计算框架，能够分析涉及非均匀和分散地面、非线性、弯曲或粗糙传播地形的实际问题，（2）结合基于硬件的并行化技术使用的高性能计算算法，以及（3）高效的算法异构建模和仿真之间的相互作用传播的电磁场和复杂的网络与任意拓扑结构和元素。 拟议研究计划的长期目标是为科学界和工程师提供先进的，高性能的强大工具，用于不同类型的大规模EMC问题的全波建模，特别是接地系统，地磁干扰，雷电电磁场及其与复杂电气化基础设施的相互作用。

Naturally or intentionally created Electromagnetic (EM) fields manifest their adverse effects in the form of malfunction or definitive failure of components and systems. Critical infrastructures in general and electrical power systems in particular must ensure continuous and reliable operation when subjected to EM fields generated by different sources such as geomagnetic storm and lightning return stroke current. The detrimental impacts of lightning current and its associated electromagnetic fields -also called lightning electromagnetic pulse (LEMP)- on electrical networks (power system, communication sites and control centers) have long been known. Regarding the geomagnetic fields, a realistic example is the blackout in the Hydro-Quebec power system in 1989 which is attributed to a large solar activity. It is always essential to assess adverse effects of EM fields on electrical power networks, communication systems, oil pipelines, electrified transportation systems and etc. The accurate quantification of the above-mentioned EM fields and their adverse effects is viewed as a large-scale Electromagnetic Compatibility (EMC) problem involving complex propagation media, i.e. inhomogeneous, dispersive ground and irregular terrain. Accurate modeling of such large-scale EMC problem is beyond the capability of the existing computational packages mainly due to their computational performance limitations and modelling effort concerns. The objective of this proposal is to develop rigorous and high-performance computational methods for the efficient modeling and solution of stringent large-scale EMC problems. The computational platform accommodates advanced version of numerical methods in electromagnetics such as Finite Difference Time Domain and Method-of-Moments designed and customized for large multiscale EMC problems. The proposed research program therefore aims to develop: (1) a unified computational framework, capable of analyzing realistic problems involving inhomogeneous and dispersive ground, nonlinearities, curved or rough propagation terrain, (2) high-performance computation algorithms used in conjunction with hardware-based parallelization techniques and (3) efficient algorithms for heterogeneous modeling and simulation of interaction among propagated EM fields and complex networks with arbitrary topology and elements. The long-term objective of the proposed research program is to provide the scientific community and engineers with advanced, high performance robust tools for full-wave modeling of different types of large-scale EMC problems with particular reference to grounding systems, geomagnetic disturbances, lightning electromagnetic fields and their interaction with complex electrified infrastructures.