Robert M. Gardner

Wide-area Frequency Based Event Location Estimation

This paper discusses the latest developments in wide-area frequency-based event location using FNET. FNET, meaning ldquoInternet based frequency monitoring networkrdquo, is a low cost and quickly deployable wide-area frequency measurement system with high dynamic accuracy. This project demonstrated the feasibility of using information from FNET to estimate the location of events in the electric grid. An on-line real time event detection tool using least squares methods was developed and tested on data from FNET. This paper outlines the methods used along with an examination of implementation feasibility.

Power system event location analysis using wide-area measurements

Power system frequency can tell volumes about the health and state of a power system. Moreover, the manner in which angle and frequency perturbations travel throughout the power grid has been closely studied in the past. Such perturbations are known to propagate throughout the grid as a function of time and space. Using measurements from a wide-area frequency monitoring network (FNET), one can observe the electromechanical wave signature of power system events in frequency data. In this paper, a brief synopsis of FNET is presented along with an analysis of several generator tripping events. The analysis is comprised of several event location methods and resulting observations. The eventual goal of this research is to formulate a robust event location algorithm that can be used online and in real-time to locate with sufficient accuracy the hypocenter of power system events irrespective of utility boundaries

Analysis of system oscillations using wide-area measurements

Analysis of power system oscillations is fundamental to understanding power system stability and performance. These oscillations clearly manifest themselves in power system frequency. Virginia Techs Internet-based frequency monitoring network (FNET) collects and organizes frequency data from throughout the grid in real time. After this frequency data is filtered with robust statistical methods, oscillations are easily extracted from the data. This paper details the value of these wide-area measurements in oscillation analysis and the development of a graphical tool efficiently analyze these oscillations

Non-Parametric Power System Event Location Using Wide-Area Measurements

Previous work has shown that the study of the electromechanical wave signature of a power system perturbation, such as a generator trip, can lead to an understanding of how to approximate the origin of the perturbation (R.M. Gardner, et al., 2006). In this paper, a brief synopsis of FNET is presented along with an analysis of generator tripping events. The analysis is comprised of several non-parametric event location methods and resulting observations. The eventual goal of this research is to formulate a robust event location algorithm that can be used online and in real-time to locate with sufficient accuracy the hypocenter of power system events irrespective of utility boundaries

Generator Trip Identification Using Wide-Area Measurements and Historical Data Analysis

Changes in power system operating conditions cause dynamic changes in angle and frequency. These changes propagate through the system as disturbances in frequency as previously studied (J.S. Thorp, et al., 1998), (L. Huang, et al., 2001). The disturbances in system frequency travel with finite speed across the system area (M. Parashar, et al., 2004). Thus a sudden loss of generation in one area would be seen as a frequency disturbance in other areas of the power system. Through use of wide-area frequency measurements these traveling frequency disturbances can be observed and recorded. This paper focuses on the analysis of this recorded data and how the specific properties of a generator trip can be determined. Future generator trips can then be compared against the recorded frequency signatures to determine a probable match. In this way the trip location can be determined. This analysis and comparison process is capable of being completely automated to be used in real time to determine the health of the power system as a whole

Off-line event filter for the wide area frequency measurements

System-wide frequency is one of most important vital signs of power system. To better understand the dynamics of large power systems, wide-area, synchronized monitoring of system frequency is needed. The concept of an Internet based, real-time, GPS synchronized, wide-area frequency monitoring network (FNET) was proposed in 2000. Today, FNET consists of more than 30 frequency disturbance recorders (FDRs) throughout the US with an information management system (IMS) based at Virginia Tech. In the near future, more FDRs will be deployed. With the help of FNET, a wide-area power system disturbance event filter has been developed. The filter helps pick out the disturbances from massive amounts of daily monitoring data and collect valuable information for the disturbance analysis

Estimating Speed of Frequency Disturbance Propagation Through Transmission and Distribution Systems

Changes in power system operating conditions cause dynamic changes in angle and frequency. These changes propagate through the system as disturbances in frequency as previously studied (J.S. Thorp, et al., 1998), (L. Huang, et al., 2001). The disturbances in system frequency travel with finite speed across the system area (M. Parashar, et al., 2004). Thus a sudden loss of generation or load in one area would be seen as a frequency disturbance in other areas of the power system. Through use of wide-area frequency measurements these traveling frequency disturbances can be observed and recorded to be used for real time system analysis. Currently the majority of wide area measurements are taken by PMUs and similar devices at the transmission level. This paper focuses on the effects of performing these measurements at the distribution level and how this additional layer of the system can affect the measurements. It is important to note that the relevant effects of machine inertia are not considered in this study. Rather this paper focuses on system/line impedance in the determination of electromechanical wave propagation speed

Generation-Load Mismatch Detection and Analysis

Using data from high-resolution wide-area GPS-synchronized measurements to detect and analyze disturbances within a synchronous electrical interconnection is not trivial. Deterministic methods often fall prey to the non-stationary noisy nature of this type of data. This paper dwells on the location of noteworthy events within “blue-sky” ambient data taken from particularly noisy environments. The statistical processing used to detect and analyze power system events is congealed into a concise and tunable algorithm that is applicable to a multitude of data types. The data analyzed herein is drawn from FNET, a wide-area measurement system deployed worldwide at standard end-user distribution voltages.

Wide area synchronized measurements and inter-area oscillation study

The Frequency Monitoring Network (FNET) provides a low-cost, easily deployable method for wide-area monitoring. FNET has the capability to realize long term, high density and continuous monitoring of large interconnected power systems. In effect, there are potentials for ground-breaking research on wide-area monitoring and control. In this paper, low frequency oscillations in the Eastern Interconnection (EI) are analyzed based on generation trips detected by FNET in 2007. The oscillation data were selected in a way such that their locations are distributed throughout the EI system. In addition, some preliminary statistical analysis of the dynamic characteristics is presented. Through illustrating the control method of system stability, it is shown that the analysis of oscillation mode shape distribution is very important. As an example, some preliminary simulation results are provided for the control of system low-frequency oscillation due to transmission line faults or other changes in system topology.

Power system frequency oscillation characteristics

This paper reports the initial results on low frequency oscillation monitoring and analysis combining practical measurements from an Internet-based frequency monitoring network (FNET) and simulations. Typical oscillation characteristics excited by various disturbances are shown using simulations and real measurements. The corresponding time-space dynamic propagation patterns are also provided. The frequency domain characteristics are also provided revealing oscillation modes and damping factors.