含新能源发电接入的复杂电网宽频带振荡在线辨识与阻尼控制

波动性新能源的高比例接入使得电力系统的惯量减弱、动态行为趋于复杂，低频振荡、次同步振荡等多模态宽频带振荡问题凸显，这已对现代电网安全稳定运行构成威胁。为此，本项目建立融合实时/历史/仿真数据的新能源电力系统动态特征库，提出数据驱动的宽频带振荡在线辨识方法，建立动态数据分类器，实现测量数据与振荡模态之间的自动快速匹配；提出控制设备/策略对振荡阻尼贡献的统一量化指标，站在全局稳定性角度优选反馈控制信号；提出宽频带振荡阻尼协调控制方法，形成振荡阻尼控制器备选集，建立振荡模态与阻尼控制器之间的映射关系；研发一套宽频带振荡在线辨识与阻尼控制系统，建立硬件在环的新能源电力系统实时仿真模型，开展半实物仿真与实验测试工作。通过本项目将形成一套数据驱动的宽频带振荡在线辨识与协调优化控制的理论与方法，这对于促进新能源规模化电网消纳、保障电网的安全稳定运行具有重要的科学意义。

The high penetration of renewable energy sources (RES), which has the characteristics of fluctuation, results in the reduction of system inertia, the complicate dynamic behavior and wide-frequency band oscillation modes, e.g., low frequency oscillations, sub-synchronous oscillations and so on. These changes bring a severe threat to the stable and secure operation of the modern power systems. To deal with this situation, a dynamic feature library of the RES-based power system, which integrates the real-time, historic, and simulated data, is established in this project. Based on this feature library, a data-driven online identification method of the wide-frequency band oscillation modes and a dynamic data classifier are proposed, which can be used to match the oscillation modes with the real-time measured data automatically. Then, a unified quantitative index is proposed to evaluate the contributions of controllable devices and control strategies to the oscillation damping. The optimal feedback control input signals for the damping controllers can be chosen from the point of overall stability of the power system, according to this index. A coordinated damping control method for the wide-frequency oscillation modes and a library of alternative damping controllers are proposed. The mapping relation between the oscillation modes and the damping controller can be set up. Finally, an online identification and damping control system of the wide-frequency band oscillations will be designed. The real-time hardware in-the-loop simulation model of the RES-based power system will be developed for the hardware-in-the-loop simulation and experimental test. This project will form a theory and methodology of a data-driven identification and a coordinated optimal control strategy for the wide-frequency band oscillations, which has an important scientific significance to promote the large-scale integration and consumption of the renewable energy generations, and to guarantee the secure and stable operation of the power grid.