Mahmood Saadeh

A unified silicon/silicon carbide IGBT model

A new physics-based IGBT compact model has been developed for circuit simulation of silicon (Si) or silicon carbide (SiC) devices. The model accurately predicts the steady-state output, transfer and switching characteristics of the IGBT under a variety of different conditions. This is the first IGBT model to predict the behavior of p-channel SiC IGBTs. Previous work on IGBT models has focused on Si n-channel IGBTs [1]. The unified model is not limited to SiC p-channel IGBTs; the user has the option to select between Si or SiC, and n-channel or p-channel, making it the first IGBT model that captures the physics of all of these device and material types. The model also accounts for temperature effects, often referred to as temperature scaling. The model have been experimentally validated up to 125 °C for silicon and 300 °C for SiC. Validation of n-channel and p-channel devices for both Si and SiC was accomplished by fitting the steady-state characteristics and inductive load switching transient waveforms. 15-kV p-channel IGBTs supplied by Cree were among those used for validation. The fitting was achieved using Certify, a software tool developed at the University of Arkansas. A parameter extraction recipe for the model was developed for simple parameter extraction using data that are readily available from datasheets. That fitting tool is available to the public through the National Center for Reliable Electric Power Transmission website (ncrept.eleg.uark.edu). The model and parameter extraction recipe will also be made available to the public through NCREPT.

A risk-based planning method for N-1-1 contingency analysis of transmission systems

A risk-based method is introduced for assessing the transient stability characteristics of electric power systems under N-1-1 contingency analysis. It is shown that the risk and corresponding severity depend on factors including the pre-fault load-flow, the clearing time of circuit breakers and the timing of the second fault. To illustrate the assessment of transient stability of a power system, the method is applied to 68-bus, 16-machine IEEE benchmark system. Simulation results using MATLAB-Simulink with repeated random trials are included to confirm the effectiveness of the proposed method for a statistically large set of sample cases.

A methodology to coordinate solid-state fault current limiters with conventional protective devices

High short-circuit currents can cause equipment failures that conventional protective devices may not avoid. Alternatively, solid-state fault current limiters (SSFCL) are designed to limit high levels of short circuit currents, in particular, within quarter cycle. However, the SSFCL may cause sensor and protection equipment malfunction; this may lead to mis-coordination and false tripping between existing protective devices, and thus reduce system reliability. This paper addresses a methodology to coordinate conventional protective devices and a thyristor-based SSFCL in a distribution system, and analyzes potential coordination issues and effectiveness of the proposed method. It also describes an approach to produce SSFCL time-current characteristic curves (TCC) and their use in protection coordination studies. Lastly, the analysis includes SSFCL to recloser and SSFCL to fuse coordination cases and includes simulation results of several fault scenarios. Lastly, the paper describes.

Anti-series normally-On SiC JFETs operating as bidirectional switches

Ac-ac matrix converters and cycloconverters require bi-directional switches, which are typically formed by two antiparallel thyristors or a two-switch (IGBT/MOSFETs) two-diode configuration. As silicon carbide (SiC) and gallium nitride (GaN) devices become more available, it is possible to have higher voltage FETs with low conduction and switching losses and reverse conduction capability, which allows the elimination of the diodes in a bidirectional switch. This paper will investigate a bidirectional switch formation that is formed by using two normally-on SiC JFETs in anti-series with no anti-parallel diodes.

A solid state fault current limiter control algorithm

The fault current limiter (FCL) is considered as the ideal solution to limit the fault current in integrated electrical power systems with distributed energy resources because of its speed of response. However, there are some practical application issues for FCL in commercial industrial deployments. In this paper, the coordination issue of FCL with the other protective devices is discussed making use of simulation and experimental results. The fundamental point of operating a FCL is that it introduces an additional equivalent impedance into the power system and thus influencing the operation of the other protective devices, sometimes, leading to malfunction. A formula is proposed in this paper to estimate the impedance range of the equivalent FCL in order to maintain the coordination with the other protective devices. An example is used to demonstrate that the fault current can be limited by controlling the switching position of a solid state fault current limiter (SSFCL) while maintaining system coordination.

Estimation of the bus admittance matrix for transmission systems from synchrophasor data

This research proposes a technique for determining the bus admittance matrix Ybus for large-scale power systems from recorded synchrophasor measurements. The approach is based on recognizing that measurements of bus injection currents Ibus can be viewed as signals produced by a random process. In this manner, the corresponding bus voltages Vbus are also stochastic signals that are related through a cross-covariance matrix to the injection current vector Ibus. Using estimation techniques developed for statistical signal processing, the cross-covariance matrix is shown to be Ybus. The increasing use of synchrophasors has enabled large-scale data collection of time synchronized bus injection currents and voltages. The new Zbus estimation method is applied to the IEEE 68 bus benchmark system to demonstrate the validity of approach. The accuracy and convergence rate of the method is evaluated under conditions corresponding to wide-area synchrophasors data collection. The results indicate that the method is broadly applicable to determining Ybus and Zbus for electric power transmission and distribution systems equipped with synchrophasor data collection technology.IEEE 68 bus benchmark systemwide-area synchrophasors data collection.electric power transmissiondistribution systemssynchrophasor data collection technology.

Analyzing variable time between N-1-1 contingencies in assessing NERC TPL-001-4 multiple events reliability

There is increasing emphasis on accounting for multiple sequential faults when analyzing for system reliability. This research considers N-1-1 contingency analysis as described in NERC TPL-001-4. Existing guidelines allow for the power system to respond to a first event prior to a second fault occurring. This research develops a method to evaluate the more complex situations where there is a variable amount of time between the first and second contingency. It is generally possible for the cumulative effect of transient interactions to be influenced by the timing of a second disturbance. A probabilistic approach is proposed to identify the most severe transient stability outcomes. The results of a large scale simulation study on a 68-bus IEEE stability benchmark model are presented to confirm the benefits of the proposed probabilistic method.

A proposed benchmark model for wind energy transmission systems

The results of developing a transient stability benchmark model are presented. This benchmark model represents the effects of areas with large wind generation. The proposed benchmark system is based on the transmission network in the south-central region of the US. The model is first derived based on transmission interconnections, conventional and wind generation and the associated geographical information. Generating station and transmission geographical results were determined from the U.S. Energy Information Administration maps features (EIA). Steady-state power flow analysis is performed to check the reasonableness of the results in terms of the impact on the surrounding 345 kV transmission grid. The power flow analysis is also used to establish the initial conditions for subsequent transient stability analysis. Fault conditions were evaluated under large wind generation conditions using small-signal (eigenvalue) analysis. The results suggest that the model captures the principle effects of wind generation related transients and power flows. This method of model development results in a benchmark system that allows system planners to evaluate the impacts on power system reliability associated with wind energy transmission networks. The results suggest that the model is realistic and somewhat accurate. This is one of the first benchmark systems for wind energy transmission networks. The use of a realistic wind energy benchmark system is of utmost importance as wind energy systems have certain characteristics that need to be accounted for when developing new technologies for such systems.

Improved stability design of interconnected distributed generation resources

This work provides a design method for achieving a specified level of stability for inverter-based interconnected distributed generation. The stability of parallel connected distributed energy resources determined from a linearized state-space model of the inverter dynamics that includes the admittance matrix of the interconnecting distribution lines. Each inverter uses a localized droop control scheme with the associated voltage and frequency measurements obtained through the application of an enhanced phase locked loop. Previous work on this topic has focused on single inverters connected to an infinite bus without modeling of delays from a phase locked loop implementation. This proposed method overcomes both of these limitations of previous research. A detailed large-signal simulation of a three-bus interconnected power system is analyzed under two different network admittance values. Results confirm the effectiveness of the proposed stability design method.