Edson L. Geraldi

Benchmark Models for the Analysis and Control of Small-Signal Oscillatory Dynamics in Power Systems

This paper summarizes a set of six benchmark systems for the analysis and control of electromechanical oscillations in power systems, recommended by the IEEE Task Force on Benchmark Systems for Stability Controls of the Power System Dynamic Performance Committee. The benchmark systems were chosen for their tutorial value and particular characteristics leading to control the system design problems relevant to the research community. For each benchmark, the modeling guidelines are provided, along with eigenvalues and time-domain results produced with at least two simulation softwares, and one possible control approach is provided for each system as well. Researchers and practicing engineers are encouraged to use these benchmark systems when assessing new oscillation damping control strategies.

A New and Fast Method for Preventive Control Selection in Voltage Stability Analysis

A new method for selection of the most effective controls to prevent voltage instability in electrical power systems is developed in this paper. The proposed method is fast and suitable for real time operation. It is based on a sensitivity analysis of both a maximum loadability estimate (which is obtained via Look-Ahead method) and the load flow solution with respect to the selected controls. These sensitivities are calculated without computing the maximum loadability points, which significantly speeds-up the analysis. The method was successfully tested in a reduced model of the south-southeast Brazilian system composed of 107 buses.

Estimation of synchronous generator model parameters operating under unbalanced three-phase conditions

In this paper, a trajectory sensitivity method (TSM) is used to estimate parameters of a synchronous generator model suitable for unbalanced conditions which are typical of distribution systems. The current formulation and mechanical equations which represent the synchronous machine have parameters whose value modification cause significant changes in systems output. Trajectory sensitivity method is used to estimate the parameters which most affect the behavior of the system and whose trajectory sensitivity functions have comparable values. The obtained results show the method is able to estimate the parameters with great accuracy, once the initial values of the parameters are inside the convergence region, even when the system is liable to a high level of imbalance. It can also be noticed through the results that, even considering great modifications in the parameters, the convergence region is large enough to guarantee the correct estimation of the parameters.

An approach for statistical quantitative analysis of voltage fluctuations using extended participation factors

In this paper, a new concept of participation factors is used to investigate the presence of the voltage fluctuations induced by the electromechanical oscillations in systems with distributed generation. Unlike the originally proposed participation factors (which will be called traditional participation factors), the extended participation factors is calculated taking into account a set of initial conditions, which makes them able to detect and quantify the voltage fluctuation from a statistical viewpoint. In order to obtain a set of initial conditions that efficiently reflects the random nature of the plausible disturbances in the system, a Monte Carlo method and a normal distribution of the disturbance parameters are used to build this set. The obtained results show the efficiency of extended participation factors in detecting and statistically quantifying voltage fluctuations. In addition, the results indicate that it is possible to infer the probability of occurrence of voltage fluctuation in the studied system considering a statistical approach.

An approach to estimate parameters of a three-phase synchronous generator model

In this paper, in order to estimate the operational parameters of synchronous generator model suitable for unbalanced conditions, an approach which uses Trajectory Sensitivity Functions (TSF) and is based on Unscented Kalman Filter (UKF) is developed. In this framework, iniatilly, TSF are used to identify the parameters which most affect the behavior of the system. After the ordination of the parameters according to the TSF values, an iterative process which uses an estimation method based on the UKF is employed. Analyzing the results obtained by the approach when it estimates the parameters of a model of synchronous generator connected to a distribution system, we can verify that the proposed approach works efficiently.

Application of modal estimation techniques for small signal stability assessment in distribution systems considering measurement noise

The increase of distributed generation in the global scenario has motivated the study of the electromechanical oscillation phenomenon in distribution systems with load imbalance, an intrinsic characteristic of this type of system. In this paper, after showing that typical linearization procedures cannot be applied to the assessment of the small-signal stability of unbalanced distributed generation systems, this assessment is done with the application of a modal estimation technique: the Prony method. This approach enables the calculation of oscillation frequencies and damping ratios under unbalanced conditions, as earlier works from the authors have shown. However, under realistic conditions, measurement noise patterns with different spectra may degrade the results of the proposed application. The main contribution of this paper is the evaluation of the impact of noise over the small-signal stability assessment technique proposed by the authors for unbalanced distribution generation systems, aiming at a better suited technique for practical applications. The results show that, while frequency remains mostly unaffected by the presence of noise, the identification of the damping ratio can be significantly influenced by this factor.

An approach to evaluate the occurrence of flicker induced by electromechanical oscillations

In this paper, an approach to evaluate the occurrence of flicker induced by electromechanical oscillations is proposed. This approach employs the extended participation factors (EPF) to statistically quantify the presence of electromechanical modes in the dynamics of the terminal voltage of a distributed synchronous generator. Once the results obtained by the EPF indicate a high probability for the manifestation of the electromechanical mode in the terminal voltage of the distributed generator, an estimate for the probability of occurrence of flicker can be obtained.

An approach to refine linear models used in small signal stability assessment

In this paper, an approach is proposed to improve linear models adopted in small-signal stability studies using information extracted from measured outputs of real system, as small disturbance data recorded by phasor measurement units (PMU). With this aim, a trajectory sensitivity technique is applied on the mismatch between the output of the simulated model and the response of the real system. Using this error signal and the sensitivity curves, this method can refine linear model in order to obtain a good match between the model and the measurements. In addition, this technique allows to identify which coefficients of the linear model can be refined from available data. Results obtained for a test case show that the proposed approach, which uses signals that can be directly measured by PMUs, is effective to calibrate the coefficients of interest in the system linear model.

A new and fast method for preventive control selection in voltage stability analysis

Summary form only given. A new method for selection of the most effective controls to prevent voltage instability in electrical power systems is developed in this paper. The proposed method is fast and suitable for real time operation. It is based on a sensitivity analysis of both a maximum loadability estimate (which is obtained via Look-Ahead method) and the load flow solution with respect to the selected controls. These sensitivities are calculated without computing the maximum loadability points, which significantly speeds-up the analysis. The method was successfully tested in a reduced model of the south-southeast Brazilian system composed of 107 buses.