Search

This paper proposes a robust parameter estimator to deal with the problem of inaccurate power losses calculated from the estimated state variables. These power loss inaccuracies result from some incorrect branch impedance due to non-stationary ambient conditions or geometrical constraints unreported in the database. Specifically, we use the projection statistics and the coupling relationship between parameter errors and the measurements to detect incorrect branch parameters associated with leverage points, which are proved to show the largest influence on the state estimates. Then these suspicious parameters are augmented with the original system state variables for the generalized estimation. Finally, the generalized robust estimator is executed as a nonlinear constrained optimization problem. The necessary condition for the primal-dual interior point method-based solver is derived. Our proposed method is able to enforce each suspicious parameter within its validity domain expressed as inequality constraints. Thanks to the robustness of the proposed Schweppe-type Huber (SHGM)-estimator and the enhanced measurement redundancy using multiple night off-peak snapshots of a nearly constant power system state, incorrect parameter values and erroneous zero injections are reliably detected and effectively corrected and validated. The performance of the weighted least squares estimator, the generalized least absolute value estimator, and the SHGM-estimator are evaluated and compared with regards to the estimated values of impedances and losses. Field test results on the realistic 2000-bus French power system show that the SHGM-estimator outperforms the other two estimators. [ABSTRACT FROM AUTHOR]

This paper summarizes the technical activities of the Task Force on Power System Dynamic State and Parameter Estimation. This Task Force was established by the IEEE Working Group on State Estimation Algorithms to investigate the added benefits of dynamic state and parameter estimation for the enhancement of the reliability, security, and resilience of electric power systems. The motivations and engineering values of dynamic state estimation (DSE) are discussed in detail. Then, a set of potential applications that will rely on DSE is presented and discussed. Furthermore, a unified framework is proposed to clarify the important concepts related to DSE, forecasting-aided state estimation, tracking state estimation, and static state estimation. An overview of the current progress in DSE and dynamic parameter estimation is provided. The paper also provides future research needs and directions for the power engineering community. [ABSTRACT FROM AUTHOR]

Due to the communication channel noises, GPS synchronization process, changing environment temperature and different operating conditions of the system, the statistics of the system process and measurement noises may be unknown and they may not follow Gaussian distributions. As a result, the traditional Kalman filter-based dynamic state estimators may provide strongly biased state estimates. To address these issues, this paper develops a robust generalized maximum-likelihood unscented Kalman filter (GM-UKF). The statistical linearization approach is presented to derive a compact batch-mode regression form by processing the predicted state vector and the received measurements simultaneously. This regression form enhances the data redundancy and allows us to detect bad phasor measurement unit measurements and incorrect state predictions, and filter out unknown Gaussian and non-Gaussian noises through the generalized maximum likelihood-estimator. The latter minimizes a convex Huber function with weights calculated via the projection statistics (PS). Particularly, the PS is applied to a proposed 2-dimensional matrix that consists of temporally correlated innovation vectors and predicted states. Finally, the total influence function is used to derive the error covariance matrix of the GM-UKF state estimates, yielding the robust state prediction at the next time instant. Extensive simulations carried out on the IEEE 39-bus test system demonstrate the effectiveness and robustness of the proposed method. [ABSTRACT FROM AUTHOR]

Real-time state information provided by the state estimator plays a major role in power system monitoring and control. As a result, the convergence of the estimator under various system operating conditions becomes one of the key requirements. In this paper, a state estimation framework that achieves both statistical and numerical robustness is proposed. This estimation framework also generalizes several well-known estimators, including the weighted least squares estimator, the least absolute value estimator, the Huber Maximum-likelihood-estimator, and the Schweppe-type Huber generalized Maximum-likelihood (SHGM)-estimator. The statistical robustness of each estimator has been studied analytically through the total influence function. In addition, the dependence between the statistical robustness of an estimator and the numerical robustness of the iterative algorithm is investigated. To enhance the numerical robustness of the iterative algorithm that solves the SHGM-estimator, a fourth-order Levenberg–Marquardt approach-based SHGM (LM-SHGM) estimator is proposed. It integrates the statistical robustness of that estimator to address various types of bad data and the numerical robustness of the LM approach to handle highly stressed operating conditions. Numerical results carried out on the IEEE test systems demonstrate the good convergence and robustness of the proposed approach under various operating conditions. [ABSTRACT FROM AUTHOR]

This paper develops a robust power system state estimation framework that accounts for correlations and imperfect time synchronization of the measurements. In this framework, correlations of the measurements obtained from the supervisory control and data acquisition (SCADA) system and the phasor measurement units (PMUs) are separately calculated through the unscented transformation and a vector auto-regression (VAR) model. Specifically, the PMU measurements during the waiting period of two successive SCADA measurement scans are buffered via a VAR model whose parameters are robustly estimated using the projection statistics. The latter take into account their temporal and spatial correlations and provide the needed measurement redundancy to suppress bad data and mitigate imperfect time synchronization. In the case where the SCADA and the PMU measurements do not arrive simultaneously at the control center, yielding imperfect measurement time synchronization, either the forecasted PMU measurements or the prior SCADA measurements from the latest state estimation run are leveraged to restore system observability. Finally, a robust generalized maximum-likelihood (GM)-estimator is extended to integrate the measurement error correlations and to handle the outliers, also known as bad data. Simulation results that stem from a comprehensive comparison with other alternatives under various conditions demonstrate the benefits of the proposed framework. [ABSTRACT FROM AUTHOR]