Next generation electromagnetic transient simulation tool for large-scale power systems with detailed equipment models

电力系统是最大的人造地理分布的非线性结构，是 由发电机、变压器、输电线路、用户负荷、 绵延数千公里。使用电磁暂态仿真(EMT)工具分析和 解决电力系统中的重大运行和控制问题。大规模电力系统的电磁兼容仿真 并行计算耗费了大量的计算能力，迫切需要并行编程技术。 当前可用的EMT仿真工具被设计为在单核CPU上运行，因此不能 利用多核CPU和多核图形处理器(GPU)管理计算 装填。海量线程计算是提高EMT计算量的关键技术之一 当处理单元具有足够的硬件核心时，基本上可以提高处理能力。这个项目的目的是 为实现大规模并行大线程EMT模拟器而开发了一系列创新 在GPU和多核CPU硬件架构上为系统供电。这部作品的主要特点是 为各种电力系统和电力电子元件设计和制造大螺纹模型，有限 单元和有限差分时间域模型、高效的数值算法、高级数据分析和 可视化，以及模拟器的验证。 项目结果证明了功能并行大线程EMT模拟器的可行性 在使用非常详细的系统的情况下，可以在大规模系统仿真中实现计算加速 组件模型。该项目的主要受益者将是马尼托巴省高压直流研究中心(MHRC) 可以将开发的创新集成到其旗舰EMT仿真工具PSCAD/EMTDC中。这个 在该项目中接受培训的经验丰富的高素质工程师团队将为正在进行的 EMT模拟技术的进步，并可能被证明是MHRC未来的有价值的员工。

Electrical power systems are the largest man-made geographically distributed nonlinear structures that are composed of a variety of equipment such as generators, transformers, transmission lines, customer loads, spanning thousands of kilometers. Electromagnetic transient simulation (EMT) tools are used analyze and solve major operational and control problems in power systems. EMT simulation of large-scale power systems consumes so much computational power that parallel programming techniques are urgently needed in this area. Currently available EMT simulation tools were designed to run on single-core CPUs, and therefore, cannot take advantage of the multi-core CPUs and many-core graphics processors (GPUs) to manage computational load. Massive-thread computing is one of the key developments that can increase the EMT computational capabilities substantially when the processing unit has enough hardware cores. The aim of this project is to develop a host of innovations towards implementing a parallel massive-thread EMT simulator for large-scale power systems on the GPU and multi-core CPU hardware architectures. The main features of this work is to design and build massive-thread models for various power system and power electronic components, finite element and finite difference time domain models, efficient numerical algorithms, advanced data analysis and visualization, and validation of the simulator. The project results are expected to prove the feasibility of a functional parallel massive-thread EMT simulator that can achieve computational speed-up in large-scale system simulation while using very detailed system component models. The main beneficiary of this project will be Manitoba HVDC Research Centre (MHRC) which can integrate the developed innovations into their flagship EMT simulation tool PSCAD/EMTDC. The experienced team of highly qualified engineers trained in this project will contribute to the ongoing advancement of the EMT simulation technology and may prove to be valuable future employees for MHRC.