基于事件驱动与数据协同的电力电子混杂系统硬件在环仿真研究

硬件在环仿真已成为电力电子系统设计与验证的基础工具，但难以跨越多个时间尺度精确表征大规模电力电子混杂系统开关瞬态和系统暂态动力学行为。为了解决此问题，本项目拟开展基于事件驱动与数据协同的电力电子混杂系统硬件在环仿真关键技术研究。首先开展基于离散状态事件驱动机制的混合步长硬件在环仿真方法研究；其次开展同步周期数据协同实时硬件在环仿真关键技术研究，包括事件预测实时采样、自适应混合步长积分算法、解耦并行多速率仿真和自恢复容错方法，以提高仿真速度和精度；最后开展异步周期数据协同准实时硬件在环仿真关键技术研究，包括异步周期数据协同的时序控制方法和软硬件虚/实时间比值自适应调节技术，以扩大仿真规模和表征亚微秒时间尺度的特征。本项目预期研究成果将使仿真速度不再受限于仿真步长，能够自适应调节仿真精度，为硬件在环仿真跨越多个时间尺度精确表征大规模电力电子混杂系统动力学行为提供理论基础和关键方法。

Hardware-in-the-loop simulation has become a basic tool for the design and verification of power electronic systems, but it is difficult to characterize accurate dynamic behavior of switching transients and system transients of large-scale power electronic hybrid system across multiple timescales. In order to solve this problem, this project intends to carry out research on the key technology of hardware-in-the-loop simulation of power electronic hybrid systems based on the event driven and data collaboration. Firstly, it is researched that the hybrid time-step hardware-in-the-loop simulation method based on discrete state event driven framework. Secondly, the key technologies of synchronous period data collaboration real-time hardware-in-the-loop simulation are researched, including event detection real-time sampling, self-adapting hybrid time-step integral algorithm, decoupled parallel multi-rate simulation and self-recovery fault-tolerant interactive method, to improve the speed and accuracy of hardware-in-the-loop simulation. Finally, the key technologies of asynchronous period data collaboration quasi-real-time hardware-in-the-loop simulation are researched, including timing sequence control method of asynchronous period data collaboration, and adaptive adjustment technique of virtual/real time ratio for software and hardware, to increase the simulation scales and characterize sub-microsecond timescale characteristics. The expected research results of this project will make the simulation speed no longer limited by the simulation time-step, and can adjust simulation accuracy adaptively, and provide theoretical foundation and key method for hardware-in-the-loop simulation that accurately characterizes multi-time-scales dynamic behaviors of large-scale power electronic hybrid systems.