Dynamic State and Parameter Estimation based on Robust Unscented Kalman Filters for Power System Monitoring and Control

提高电力系统的可靠性、安全性和弹性取决于快速、准确和鲁棒的动态状态估计器的可用性。这些估计器应该对测量值和模型参数值的粗误差具有鲁棒性，同时即使在存在大的动力系统模型不确定性和非高斯粗尾过程和观测噪声的情况下也能提供良好的状态估计。事实证明，目前文献中给出的基于卡尔曼滤波的动态估计器存在几个重要的缺点，使它们无法被电力公司用于实际应用。具体来说，它们无法处理(1)动态模型的不确定性和参数误差；（ii）系统非线性动态模型的非高斯过程和观测噪声；（iii）任何类型的异常值，这些异常值是由脉冲测量和系统过程噪声或不正确的系统参数值引起的，仅举几例；（iv）各种类型的网络攻击。为了应对这些挑战，本项目将采用鲁棒统计理论和鲁棒控制理论，为鲁棒动态状态和参数估计开发一个通用的理论框架。这种新的通用框架将为电力系统监测、控制、保护和安全分析提供可靠的实时状态和参数估计。此外，它将有助于下一代具有同步量测量的在线状态估计器和针对网络攻击的鲁棒检测器的重新设计。该项目还包含对STEM（科学、技术、工程和数学）领域感兴趣的K-12学生、本科生和研究生的综合教育议程。该项目将开创一个综合鲁棒统计理论和鲁棒控制理论的通用理论框架，用于网络物理系统的鲁棒动态状态和参数估计。具体而言，将广义极大似然型（GM）估计器、无迹卡尔曼滤波器和h -∞滤波器集成到一个统一的框架中，以产生各种集中和分散的鲁棒动态估计器。这些新的估计器将能够处理大的系统不确定性以及抑制三种类型的异常值，同时在大范围的非高斯过程和观测噪声下实现良好的统计效率。三种类型的异常值，包括观测值、创新值和结构异常值，是由不可靠的动态模型或具有数据质量问题的实时同步量测量引起的，这在电力系统中很常见。此外，稳健统计理论将扩展到结构化非线性回归模型。即将线性结构回归中的击穿点理论推广到以稀疏雅可比矩阵为特征的非线性动力学模型中，这正是电力系统的情况。为此，将研究所有提出方法的全局和局部击穿点。最后，开发的方法将在两个实际的电力系统上实施和测试，包括巴西南部电力系统和弗吉尼亚电力公司500千伏输电系统，通过一组冗余实时同步测量来观察。

The enhancement of the reliability, security, and resiliency of electric power systems depends on the availability of fast, accurate, and robust dynamic state estimators. These estimators should be robust to gross errors on the measurements and the model parameter values while providing good state estimates even in the presence of large dynamical system model uncertainties and non-Gaussian thick-tailed process and observation noises. It turns out that the current Kalman filter-based dynamic state estimators given in the literature suffer from several important shortcomings, precluding them from being adopted by power utilities for practical applications. To be specific, they cannot handle (i) dynamic model uncertainty and parameter errors; (ii) non-Gaussian process and observation noise of the system nonlinear dynamic models; (iii) any type of outliers that are induced by impulsive measurement and system process noises, or incorrect system parameter values, to cite a few; and (iv) all types of cyber attacks. To address these challenges, this project will resort to both robust statistical theory and robust control theory to develop a general theoretical framework for robust dynamic state and parameter estimation. This new general framework will provide reliable real-time state and parameter estimates for power system monitoring, control, protection, and security analysis. In addition, it will contribute to the next generation of online state estimators with synchrophasor measurements and the redesign of robust detectors against cyber attacks. The project also contains an integrated educational agenda for K-12 students, undergraduates and graduate students who are interested in the STEM (Science Technology Engineering and Mathematics) area.This project will pioneer a general theoretical framework that integrates both robust statistical theory and robust control theory for robust dynamic state and parameter estimation of a cyber-physical system. Specifically, the generalized maximum-likelihood-type (GM)-estimator, the unscented Kalman filter, and the H-infinity filter will be integrated into a unified framework to yield various centralized and decentralized robust dynamic state estimators. These new estimators will be able to handle large system uncertainties as well as suppress three types of outliers while achieving good statistical efficiency under a broad range of non-Gaussian process and observation noise. The three types of outliers, including observation, innovation, and structural outliers are caused by either an unreliable dynamical model or real-time synchrophasor measurements with data quality issues, which are commonly seen in the power system. Furthermore, the theories of robust statistics will be extended to structured nonlinear regression models. That is, the theory of breakdown point in linear structured regression will be extended to nonlinear dynamical models characterized by sparse Jacobian matrices, which is precisely the case for power systems. To this end, the global and local breakdown points of all the proposed methods will be investigated. Finally, the developed methods will be implemented and tested on two practical power systems, including the Southern Brazil power system and the Dominion Virginia Power 500-KV transmission system, which is observed through a set of redundant real-time synchrophasor measurements.