Standstill Frequency Response Research Articles

Modelling and identification of synchronous machines, a new approach with an extended frequency range

Correct modeling of turbogenerators is difficult, as currents are flowing during transients not only in the field and damper winding but also in the solid rotor iron. To model this effect, especially at high frequencies, new model structures called ladder networks are introduced. Their parameters are identified by combining the standstill frequency response measurements with the results of the standardized three-phase no-load short-circuit test. By means of this combination the errors in the standstill frequency responses, caused by the iron nonlinearity and the contact uncertainty of the slot wedges at standstill, are expected to be reduced. Both the new ladder networks and the standard models are used to simulate disturbances for comparison. In some cases (three-phase terminal fault out-of-phase synchronization) both models yield equivalent results, but in other cases (operation after a close-up fault) the results of the two models can differ considerably. >

Determination of the model parameters of machines from the reactance operators x/sub d/(p), x/sub q/(p) (evaluation of standstill frequency response test)

A method is presented for determining the characteristic quantities and model parameters from the given frequency responses x/sub d/(js), x/sub q/(js). The curves of values measured on a large turbogenerator at standstill are used as an example. The characteristic values and model parameters are determined for various models (with four, three, and two rotor circuits in each axis) and are compared with the results of short-circuit tests and with advanced calculations.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Modeling of Induction Motors from Standstill Frequency Response Tests and a Parameter Estimation Algorithm

ABSTRACT Measurements from standstill frequency response test (SSFR) together with the least error squares (LS) parameters estimation algorithm are used for modelling of the induction motors. The equivalent circuit parameters, which are used for studying the performances of the induction motor, can be derived from this model. A comparison between estimated parameters using the proposed algorithm and the manufacture's data available is performed, this comparison showed a good agreement of the proposed algorithm.

A frequency-domain maximum likelihood estimation of synchronous machine high-order models using SSFR test data

The authors propose a numerical scheme for processing noisy signals originating from standstill frequency response (SSFR) tests on synchronous machines. Instead of using a univariate nonlinear least-squares procedure to fit only the weighted sum of magnitude responses, they minimize a multivariate prediction error criterion based on the determinant of the residuals covariance matrix. The algorithm pertains to a large class of prediction error methods and results in a multiresponse nonlinear regression procedure related to the maximum likelihood viewpoint when the residuals distribution is Gaussian. To demonstrate the efficiency of the proposed scheme, the implementation was tested using noisy simulated data, based on the Rockport model 3.3. It is shown, using actual data from the Nanticoke turbogenerator, dating back to the EPRI-project RP-9997-2 (1980), that the frequency-domain maximum likelihood approach can be effective for direct estimation of generalized circuits with up to five equivalent windings per axis, providing satisfactory predictions of both magnitude and phase as far as the 16th harmonic. >

Synchronous Machine Parameter Identification

ABSTRACT This paper presents a survey of synchronous machine modeling and parameter identification. In addition, the paper presents results of a study conducted to estimate the effects of measurement noise on the estimation of machine parameters from Standstill Frequency Response (SSFR) test data. Results obtained indicate that because the test data are inherently noise-corrupted, multiple solution sets can be obtained. Furthermore, the effect of noise on time-domain parameter estimation of synchronous machine models axe studied. It is shown that a unique set of parameters can be obtained and the noise effects can be dealt with effectively when the maximum likelihood estimatiou (ML) technique is used to estimate machine parameters.

Impact of synchronous machine constants and models on the analysis of torsional dynamics

The authors investigate the impact of synchronous machine parameters and second-order and third-order models on the phenomenon of torsional dynamics. The results were compared with those obtained from the conventional (2d, 1q) model. For each model, the effect of rotor-field differential leakage flux on the damping of the torsional oscillations is studied. Numerical values of the elements corresponding to each model were obtained based on a parameter fitting process to a set of stand-still frequency response (SSFR) test results. An eigen analysis method was used for the studies. The eigen analysis results were verified by digital time-domain simulation, using the electromagnetic transients program (EMTP). >

Least absolute value estimation of the generalized operational impendances of solid-rotor synchronous machines from standstill frequency response test data

The generalized operational impedances of solid-rotor synchronous machines are estimated from standstill frequency response (SSFR) test data together with a linear least absolute value parameter estimation algorithm. The proposed algorithm has been applied using measurements made elsewhere on a large turbogenerator. The results obtained using the proposed algorithm are in good agreement with those obtained previously for the same turbogenerator using another technique.

Maximum likelihood estimation of generator stability constants using SSER test data

The performance of the maximum likelihood (ML) method when used to determine simulation data for generators from standstill frequency response (SSFR) tests is evaluated. The robustness of the ML method is demonstrated by analyses made with SSFR data from tests on the Rockport 722 MVA generator. It is shown that a unique set of parameters can be obtained, and the noise effects can be dealt with effectively when the ML technique is used to estimate machine parameters. >

Synchronous-machine standstill frequency-response test data analysis

Frequency response methods used in the development of linear transfer functions representing nonlinear processes are extended to standstill frequency response tests for synchronous machines, thus providing an independent check on the equivalent circuit method of IEEE Standard 115A. By iterative subtraction of three separate transfer function elements from the Bode plot of the experimental data an overall transfer function can be synthesized, leaving a small residue of Bode-plotted data that represent irregularities due to nonlinearities, noise, or erroneous data. Suggested limits for the range of test frequencies and the magnitudes of residual transfer function are given.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Test and simulation of network dynamic response using SSFR and RTDR derived synchronous machine models

The comparative testing of four 60 Hz two-pole generators indicates practical advantages for the SSFR (standstill frequency response) method over the RTDR (rotating time-domain response) method when synchronous machine models are to be synthesized from test data for network dynamic performance analysis. Manufacturer-provided data are compared to SSFR and RTDR derived parameters. All adequately duplicate synchronization, load rejection, and line switching tests. Deterministic sensitivity of CFCT (critical fault clearing time) to selected machine parameters is reported.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Time-domain identification of synchronous machine parameters from simple standstill tests

A set of standstill measurements that allow time-domain identification of linear model parameters for direct and quadrature axes of a synchronous machine is proposed. The advantages of this method over conventional standstill frequency response testing include the simplicity of test equipment, the higher current levels achieved, the speed with which the measurements can be obtained, and the fact that the models are obtained directly in parametric form (in per-unit or measured units). The proposed method is illustrated by results measured on a 3 kVA, 220 V microalternator. >

A 'three transfer functions' approach for the standstill frequency response test of synchronous machines

A three transfer functions approach for the standstill frequency response (SSFR) test of synchronous machines is proposed. Network theory is employed for the explanation of the three-function approach. The three-function and two-function approaches, as well as the one-function approach, are compared. The verification of the approach is obtained through simulations and an application to an SSFR test on a 3 kVA laboratory microalternator. The accuracy of the determined d-axis model parameters, particularly of the rotor circuits, is improved by the use of the suggested third transfer function (L/sub afo/(s)) together with the two transfer functions (L/sub d/(s) and G(s)). This requires that the measurements of the three transfer functions be taken in the SSFR test and be involved in the d-axis model fitting instead of the common practice involving only L/sub d/(s) and G(s). >

The Effects of Noise on Frequency-Domain Parameter Estimation of Synchronous Machine Models

The authors present results of a study conducted to evaluate the effects of measurement noise on the estimation of machine parameters from standstill frequency response test data. Results obtained indicate that, because the test data are inherently noise-corrupted, multiple solution sets can be obtained. Moreover, some of the estimated machine parameters could be unrealistic. It is shown that the results are very sensitive to the value of armature resistance used in the data analysis. Even a 0.5% error in the value of armature resistance could result in unrealistic estimation of the machine parameters. >

Derivation of Induction Motor Models from Standstill Frequency Response Tests

Standstill Frequency Response (SSFR) tests were conducted on four induction motors with ratings of 300, 400, 1000, and 1750 horsepower. Using the SSFR test data, an equivalent circuit was derived for each motor. The speed-torque and speed-current curves calculated from the equivalent circuit were compared with those reported by the manufacturer.

Derivation of induced motor models from standstill frequency response tests

Standstill frequency response (SSFR) tests were conducted on four induction motors with ratings of 300, 400, 1000, and 1750 horsepower. Using the SSFR test data, an equivalent circuit was derived for each motor. The speed-torque and speed-current curves calculated from the equivalent circuit were compared with those reported by the manufacturer. The comparison showed good agreement of the SSFR-based model with manufacturer's performance curves in the case of the 300 hp motor. Including a simple approximation for saturation effects provided reasonable agreement for the 1000 hp and 1750 hp motors. Results for the 400 hp motor were at considerable variance with manufacturer's curves. >

The effects of noise on frequency-domain parameter estimation of synchronous machine models

The authors present results of a study conducted to evaluate the effects of measurement noise on the estimation of machine parameters from standstill frequency response test data. Results obtained indicate that, because the test data are inherently noise-corrupted, multiple solution sets can be obtained. Moreover, some of the estimated machine parameters could be unrealistic. It is shown that the results are very sensitive to the value of armature resistance used in the data analysis. Even a 0.5% error in the value of armature resistance could result in unrealistic estimation of the machine parameters.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Determination of synchronous machine stator and field leakage inductances from standstill frequency response tests

It is shown that the identification of d-axis stator and field winding leakage inductance appears to be possible using standstill frequency-response rests with field open (applying voltage to the stator), and with the stator open (applying voltage to the field). Measurements of applied and induced voltages and currents to the excited winding form a sufficient set of data to identify the leakage inductances of both the stator and field windings. This method is illustrated and applied to limited published standstill frequency response test data. The results indicate a variation in leakage inductance, depending on the frequency of the test. The reasons for such variations are explored and hypothesized.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A Study of the Frequency Response of Turbogenerators with Special Reference to Nanticoke G. S.

The results from the standstill frequency response tests on three selected turbogenerators are critically examined. Particular attention is paid to the Nanticoke machine. A reason for the large difference between this machine's performance as predicted using data derived from standstill tests as compared to that predicted using data from on-line tests is proposed. Use is made of the finite element method linked to simple external circuits.

Synchronous Machine Modelling by Standstill Frequency Response Tests

The synchronous machine is modelled in terms of operational impedances derived from standstill frequency response measurements. A curve-fitting procedure in the complex plane yields all the parameters including the armature resistance without having to pre-process the data or make other measurements. The validity of the model is demonstrated for a micro-alternator by comparing the sudden short-circuit current predicted by the model with that obtained experimentally. Frequency response data obtained at different current levels are used to show that a useful model can be derived from tests at low per-unit current.

Synchronous Machine Modeling by Standstill Frequency Response Tests

Synchronous Machine Modeling by Standstill Frequency Response Tests