Font Size: a A A

Research On Some Issues Of Power System State Estimation Based On WAMS

Posted on:2011-06-08Degree:DoctorType:Dissertation
Country:ChinaCandidate:H LiFull Text:PDF
GTID:1102360305953225Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
At present, smart grid construction is getting attention all around the world. In China, strong smart grid based on ultra-high voltage (UHV) power grid is under construction. It needs even much higher automation level of modern dispatching centers. During this process, power system state estimation is an important constituent of modern electric dispatching centers, and it is the basis of realizing the power system real-time analysis and control.The remote measurements for traditional state estimator are gathered by supervisory control and data acquisition (SCADA) system. In 1990s, with the widely application in power systems of wide area measurement system (WAMS), which is based on phasor measurement unit (PMU), it brings a new chance for the development of power system state estimation. At present, PMU measurements have become one of the most important data sources of power system. Therefore, it is an inevitable trend to integrate WAMS and energy management system (EMS) for power system analysis and control.According to WAMS and nowdays research of state estimation, the main contents of this thesis are as follows:1. The method of equivalent current measurement transformation is proposed to transform nonlinear static state estimator into a linear estimator; PMU measurements of state variables is transformed into equivalent correction measurements of the branch current correspondingly. They are incorporated into the linear estimator to do linear estimation for fast convergence and high filtering precision.2. A new method for selection of suitable PMU placement is presented. Based on traditional SCADA measurement system, a hybrid non-linear measurement model is built by installing PMU partially. And the PMU placements are selected by using the theory of outlets number and bus injections, which perfectly improves the system observability, state estimation precision and numerical stability of the state equations.3. An adaptive dynamic state estimation algorithm based on WAMS for power systems is introduced. According to the new algorithm, estimation model parameters are on-line identified and modified to improve the filtering performance under different scenarios such as normal operation, bad measurements, sudden load change/drastic generation variation and topology errors.4. Based on traditional extended Kalman filter (EKF) algorithm, two methods of modifying the model error covariance matrix are presented; the measurement function nonlinearities are incorporated in the two dynamic state estimation models. That is, when the filter is conducted, the unknown or imprecise noise statistic characteristic is identified and modified on-line simultaneously. Additionally, since the nonlinearities of the measurement function are integrated into the state estimation models, linearized error yielding from the linearization of nonlinear power system is fully compensated. Besides, the voltage magnitudes and phase angles measured by PMU are introduced in the observed measurements. All of these greatly improve the forecasting and filtering performances of the dynamic state estimation under different situations that include normal operation, sudden load change, bad measurements and topology error conditions.5. As WAMS/SCADA system inevitably contains various anomalies such as gross measurement errors and sudden changes of system status, a new method for detection, identification and elimination of anomalies during power system adaptive dynamic state estimation is presented. In the new method, according to the capability of forecasting the system state of the dynamic state estimator, normalized innovation and weighted innovations are used to diagnose, recognize and process the anomalies such as multi bad data, sudden load changes, network topology errors and the simultaneous occurrences of the three anomalies on the system states.Based on all of the above proposed methods, software is developed and tested with some examples. The simulation results show the feasibility and validity of all the presented methods.
Keywords/Search Tags:power system, wide area measurement system, static state estimation, dynamic state estimation, adaptability
PDF Full Text Request
Related items