OMICRON-SFRA-measurement-results-Predl-Paper-omicronized.pdf

Interpretation of Sweep Frequency Response Analysis (SFRA) Measurement Results By Florian Predl OMICRON Australia

Interpretation of Sweep Frequency Response Analysis (SFRA) Measurement Results Florian Predl OMICRON Australia This paper was presented in Techcon Asia-Pacific 2016.

Abstract Power transformers are critical components in an electrical power network. Testing, diagnostics and reliable condition assessment of power transformers becomes increasingly relevant due to the aging of transformer fleets around the globe. Transfer function measurements have been used as a diagnostic tool to detect mechanical failures in power transformers. Geometrical changes in the transformer windings and core due to mechanical stress can be reflected as a change in the RLC parameters of the equivalent circuit of the power transformer. Such changes can be detected through the change in the transfer function. This paper discusses the measurement principle of sweep frequency response analysis (SFRA) on power transformers and result interpretation, which is complimented with case studies.

Introduction Transfer function measurements have extensively been used in the last years as a key diagnostic tool on power transformers. The transfer function of a power transformer winding is sensitive to mechanical and electrical influences. This diagnostic measurement is based on the fact that a geometrical change in the transformer windings and core, as a result of a mechanical impact is causing a change in the complex RLC network. Therefore, by measuring the transfer function such changes to the network can be detected. The measurement is conducted off-line, i.e. the power transformer under test has to be deenergized and has to be taken out of service, though on-line measurements have also been explored and studied [1], [2], [3], [4]. To date the suitability of the on-line applicability has not yet been proven. The first steps towards frequency response analysis (FRA) on power transformers were made in Poland in 1966 [5]. The measurement method being utilized was the low voltage impulse method (LVI). The method had been further refined in Britain and the United States. The main motivation behind the LVI measurement method was to assist in determining whether power transformers under short-circuit tests have passed or failed. The LVI method is also known as the impulse frequency response analysis (IFRA). Over the years the FRA method has been proven to be a powerful tool for detecting and diagnosis of the active part of power transformers [6]. The sensitivity of the FRA method allows to detect geometrical deformations in the windings of power transformers before the occurrence of a major or even a catastrophic failure. When talking about FRA it is important to distinguish between IFRA and SFRA (sweep frequency response analysis). www.omicronenergy.com | [email protected]

As stated in [7] the sweep frequency method was invented by Dick and Erven of Ontario Hydro Research Laboratories between 1975 and 1977. IFRA versus SFRA Considering that a power transformer corresponds to a linear time-invariant system, its response can be studied by the means of its transfer function. The transfer function can be measured in either time domain or frequency domain. This is illustrated in Figure 1 below.

Figure 1 Response of a linear time-invariant system

The IFRA method has been further developed from the historical LVI. The IFRA method is performed in time domain. Therefore the power transformer is excited with a broadband impulse signal which is applied to one end of the winding under test. The response signal is measured on the other end of the winding. Both signals are filtered with anti-aliasing filters before they are transformed into frequency domain by the means of FFT. The complex transfer function results from the quotient between the Fourier transformed output and input signals, see Figure 2.

Figure 2 Principle operation of IFRA

www.omicronenergy.com | [email protected]

SFRA proceeds by applying a sinusoidal signal of constant amplitude and variable frequency to one end of the winding under test (U1(f)). The response is measured on the other end of the winding (U2(f)). The response will vary in amplitude and phase. The transfer function (H(f)) is a comparison of the applied signal and the response, see principle operation of SFRA in Figure 3. As the SFRA method measures in frequency domain there is no further signal processing by the mains of FFT required.

Figure 3 Principle operation of SFRA

As it can be appreciated, the measurement setup consists mainly in a network analyzer and measurement cables. Given the fact that the cables have also a capacitance, these shall be grounded. Ideally, braids are used for grounding the shield of the measurement cables. A braid has a lower inductivity compared to a single wire with the same cross-section. Furthermore, using single braided wires will reduce the impact of the skin effect on the test results, especially at very high frequencies. Usually, a measuring resistance is needed for producing the voltage drop U2(f). The wave impedance of the measuring cables shall be the same as the resistance of the measurement input. Figure 4 shows the test setup of the transfer function in frequency domain.


Figure 4 Transfer function measurement in frequency domain on a power transformer

The transfer function of a power transformer is measured according to (1): H(f) = U2 (f) / U1 (f)

(1)

There are a number of possible methods of presenting the results of measurements made using the sweep frequency response method. The most widespread is to plot a graph of the amplitude, as measured by the network analyzer, over frequency. Both linear and logarithmic scales are used [8]. The amplitude is defined by: k(f) = 20 * log10(U2 (f) / U1 (f))

(2)

The phase, as measured by the network analyzer, is defined by:

 (f) = tan-1(  U1(f) /  U2(f))

(3)

This work will focus on the SFRA method only, since the SFRA method is superior to the IFRA method. The IFRA method lacks of reproducibility for in-site tests. Another major disadvantage is its low measurement resolution at very low frequencies. Type of measurements The SFRA measurement is typically performed on all accessible windings. The transfer function of each test can be compared to reference data. www.omicronenergy.com | [email protected]

The core and winding structure of a power transformer can be seen as a complex electrical network of resistances, self-inductances, ground capacitances, coupling inductances and series capacitances as schematically shown in Figure 5. The frequency response of such a network is unique and can therefore be considered as a fingerprint of the power transformer under test.

Figure 5 Simplified network behavior and equivalent circuit diagram of a transformer’s active part

For understanding the shape of the SFRA responses it is essential to have a clear idea about the behavior of the transfer function. The transfer function is dependent on the measurement resistance (R_m) which is usually 50Ω and the impedance of the power transformer under test (Z_transformer): H(f) = U2 (f) / U1 (f) = R_m / (R_m + Z_transformer)

(5)

From an electrical point of view a power transformer is a combination of resistances, inductances and capacitances. A first approach is necessary for distinguishing between the effects of the core and the windings in the transfer function. The frequency response of the iron core is a result of the magnetizing inductance (L_m), the core power losses (R_m) and the parasitic capacitance coupled with the iron core (C_g1 and C_g2). The response prevails at very low frequencies. The frequency response of the windings is a result of the cooper losses (R_1 and R_2), leakage inductances (L_1 and L_2) and other parasitic capacitances (C_s1, C_s2 and C_12). The response of the windings prevails as the test frequency is being increased. This is due to the fact that the inductance and parasitic capacitance of the iron core is much greater than the leakage inductivity and parasitic capacitances of the windings. In Figure 6 a typical FRA response is described with reference to the parameters of the equivalent circuit diagram of a power transformer. At low frequencies, typically between 20Hz and 1kHz (frequency ranges are depending on the transformer to be tested) the magnetizing inductance dominates the response. The first parallel resonance frequency is www.omicronenergy.com | [email protected]

due to the resonance between the magnetizing inductance of the iron core and the parallel, respectively parasitic capacitance of the power transformer. It can be seen that for both phase A and phase C two parallel resonance points take place, whereas for phase B only one parallel resonance takes place. This is due to the two magnetic paths in the iron core when the test signal is injected at either phase A or phase C. At medium frequencies, typically between 1kHz and 10kHz, the parallel capacitance and the mutual inductances are dominating the response. The mutual inductances are due to the mutual coupling effect between the high voltage (HV) and low voltage (LV) winding of a power transformer. Hence, this frequency range is often referred to as the mutual coupling frequency range. At high frequencies, typically between 10kHz and 1MHz, the response is dominated by the winding capacitances and inductances. Any mechanical change within the winding structure would affect this frequency range. At frequencies of 1MHz and beyond, the effect of the actual measurement setup will have a great impact the response. Therefore, it is not standard practice to analyze traces at frequencies beyond 1MHz. The upper frequency limit for analyzing purposes depends very much on the physical size of the power transformer and is lower the bigger the transformer. Typically, for a large power transformer (>500MVA, >400kV) the upper frequency limit is approx. 0.5MHz. For a small distribution transformers (<5MVA) the upper frequency limit is approx. 4MHz).

Figure 6 Simplified network behavior and equivalent circuit diagram of a transformer’s active part

There are different types of FRA measurements. In the existing recommended practices [9] and [10] different types of measurements have been standardized. According to the CIGRE terminology these measurements can be subdivided into four groups:    

End to end open circuit test End to end short circuit test Capacitive inter winding test Inductive inter winding test

The most common type is the end to end open circuit test. This type of test provides information about both the winding and the core. The end to end short circuit test is normally performed for on-site measurements in which only problems in the winding structure have to be identified. Inter winding tests have recently been introduced.


Capacitive inter winding measurement seem to be a potential type of FRA measurements due to its higher sensitivity in the detection of radial deformations. Figure 7 shows the typical response of an end to end open circuit test on a YN connected power transformer. The source (yellow) and the reference input (red) are connected to one phase. The response (blue) is measured on the neutral as per the IEC 60076-18 standard [11].

Figure 7 Typical response of an end to end open circuit test (left) and connection diagram (right)

Figure 8 shows the typical response of an end to end short circuit test on a YN connected power transformer. A comparison between the response of a end to end open circuit test and an end to end short circuit test is shown on the left hand side. At low frequencies the differences between the two measurement methods is caused by the short circuit of the magnetic core. At higher frequencies the response of both methods line up as the actual winding structure is dominating the response at high frequencies. The end to end short circuit test is very sensitive to any change in the leakage channels. Therefore, it is the preferred method for detecting any axial or radial movements of the windings.


Figure 8 Typical response of an end to end short circuit test (left) and connection diagram (right)

Figure 9 shows the typical response of the capacitive inter winding test. The capacitive inter winding test has proven to be sensitive for detecting any radial deformations within the power transformer active part. The test signal is applied to one phase on the high voltage winding. The response is measured on the low voltage winding which corresponds to the high voltage winding according to the power transformer vector group.

Figure 9 Typical response of a capacitive inter winding test (left) and connection diagram (right)

Figure 10 shows the typical response of the inductive inter winding test. This is the least common test method. The measurement is carried out on two adjacent coils in order to measure the transfer admittance of the power transformer. The test signal is applied to one end of the high voltage winding. The other end of the high voltage winding has to be connected to ground in order to allow for a magnetic flux to build up. The response is


measured on the corresponding low voltage winding. The other end of the corresponding low voltage winding has to be grounded as well to be able to pick up the induced voltage.

Figure 10 Typical response of an inductive inter winding test (left) and connection diagram (right)

Interpretation of test results SFRA is a comparative measurement method. This implies that any sort of reference data has to be available in order to analyze the test results. This means results of an actual test, which is usually a certain set of curves, are compared to reference baseline data. Three methods are commonly used to assess the measured traces: 

Time-based comparison Current SFRA results are compared to previous results on the same power transformer under test



Type-based comparison Current SFRA results are compared to another power transformer of the same design (sister unit)



Phase-based comparison Current SFRA results of one phase are compared to the results of the other phases of the same power transformer under test

Considering this fact, the very first question which has to be answered by the test engineer is the following: 

Is reference data available?

A key point for proper diagnostics is the correct documentation of the SFRA measurement in regards to the way the test has been carried out. SFRA is very sensitive. The actual test www.omicronenergy.com | [email protected]

setup including the cable alignments, connection techniques, clamps, etc. have an impact on the test results. Best practice is to take pictures of the test setup and save them together with the measurement data. This significantly helps to achieve the highest possible degree of repeatability while reproducing the same arrangement for the measurement. The available data helps the engineer to get an idea of the expected results which is of vital importance for avoiding measurement mistakes. Even differences considered to be minor can have a huge impact on the test results, e.g. tap changer position. According to the reason of the actual test additional measurements might be recommendable because different types of SFRA measurements are best-suited for different types of investigations. A time-based comparison is the easiest way to compare and assess the condition of the transformer under test. Any change to the fingerprint results of the transformer have to be further investigated. If no historical data of the concerning transformer is available, the use of SFRA data from a sister transformer, if available, can often be used. The approach is of analyzing the data is generally the same as described for the time-based comparison. However, minor deviations between curves of sister transformers can normally not be excluded [12]. Without any sort of reference data, neither a fingerprint nor from a sister transformer, it is common practice to compare the phases of a transformer against each other. For threephase transformers it must be noted that the middle phase usually differs from the two outer phases in the magnetic core region (low frequency range up to a few kHz) as well as in the higher frequency regions which are related to the winding structure. It must be noted that sometimes also healthy windings may not correlate well, according to their winding design. In such cases further investigation, for instance a type-test comparison, is required. It can be estimated that app. 90% of all SFRA measurements can be evaluated without fingerprint data of the power transformer under test being available [12]. Figure 11 shows the SFRA traces of a 10MVA, 33kV / 2x1.31kV transformer which has been investigated. There were no reference data available at the time. Therefore a phase-based comparison was the only option. On the left hand side the SFRA traces of the high voltage phases are displayed and on the right hand side the SFRA traces of the low voltage phases are displayed.

Figure 11 www.omicronenergy.com | [email protected]

Phase-based comparison of a 10MVA, 33kV / 2x1.31kV transformer being investigated

It can be seen, that the three phases of the HV winding are matching up quite nicely. In particular the serial and parallel resonance frequencies are very similar which is an indicaton that the phases are symmetrical and the winding is healthy. The SFRA traces for the LV winding show some deviations, especially for phase w-n, in the higher frequency range. This could actually indicate a problem within the winding structure. However, an actual problem in the winding structure of the LV winding should also be reflected in the response of the HV winding in the mutual coupling frequency range. It was decided to measure a sister transformer to be able to give a more reliable judgement on the test results. In Figure 12 a comparison of the suspicious phase w-n on the low voltage winding is shown. The reference trace is from the same phase on a sister transformer (type-based comparison). The two responses match up very nicely. The transformer was assessed to be healthy based on the SFRA test results.

Figure 12 Type-based comparison of a 10MVA, 33kV / 2x1.31kV transformer being investigated

Above mentioned example highlights the importance of having proper reference data available and accessible in order to be able to reliably assess the condition of the power transformer under test. A phase based comparison can be quite challenging as obvious differences in between two traces can either be indicating a serious issue within the winding structure or it can be simply caused by a slight different design between the phases. www.omicronenergy.com | [email protected]

Standards and guidelines The first valid standard in the world for FRA testing was established in China. The standard, named DL 911/2004, only refers to SFRA measurements. The standard covers subjects including the test principle, requirements for testing instruments, testing methods and the analysis of the results. An example is the measurement for star connected windings, for which the standard requires the injection of the signal into the neutral terminal and the measurement of the response at the phase terminal. In two appendices, various test examples are given The standard evaluates a frequency range between 1kHz and 1MHz and is unique in that it gives a rule about how to judge test results based on a calculation of covariance [13]. In 2002 the IEEE established a Task Force concerned with FRA and followed it up with a Working Group founded in 2004. The scope of this Working Group is the creation of a guide for the application and interpretation of FRA for oil immersed transformers. As with the Chinese standard, the IEEE recommends a three lead test system for source, reference and measurement. This is in accordance with common scientific knowledge and is supported by all SFRA test instrument manufacturers [10]. The standard, named “IEEE Guide for the Application of Frequency Response Analysis for Oil-Immersed Transformers – IEEE Std. C57.149” was released in 2012. The CIGRE Study Committee A2 – Transformers – decided in 2003 to establish a Working Group on application of FRA on power transformers. This working group A2/26 with the title “Mechanical Condition Assessment of Transformer Winding using Frequency Response Analysis (FRA)” started its work in 2004 and ended with the publication of the CIGRE report No 342 [9] in April 2008. During this period, besides regular meetings the Working Group organized two FRA test workshops. During these workshops a large number of practical investigations were performed. The main results and conclusions from the comparative tests were as follows: 

All test equipment produced essentially the same measured responses of the test objects over a mid-range of frequencies from about 10kHz to 500kHz. After the standardization of good cabling practices was agreed by the Working Group members, it was possible to get perfectly reproducible results at the same test object up to approx. 1.5MHz, even with different test instrument vendors. This is sufficient for a reliable condition assessment.



The impulse method (IFRA) was unable to reproduce the low frequency response because of digitizers set to acquire only the higher frequencies.



Some swept frequency methods also did not have sufficient dynamic range to reproduce the typical 90dB minimum obtained with a 50Ω measuring impedance [9]

Finally, the Working Group worked out the best practice to take full advantage of the proven sensitivity of FRA for condition assessment of power transformers. The resulting document is a valuable source of information and a helpful guide for the practical application of FRA.


In 2009 the IEC started to establish a Working Group to develop a standard for FRA testing [11]. The standard, named IEC 60076-18 – Power transformers - Part 18: Measurement of frequency response, was released in July 2012. The standard addressed important topics such as the measurement method, the frequency range which has to be covered and the density of measurement points and technical specifications for the measuring equipment. In Annex A the measurement lead connections are discussed. Contradict to the Chinese standard, the IEC standard requires the injection of the test signal into the phase and the measurement of the response at the neutral terminal. Factors affecting the reproducibility The SFRA method is a very robust test method which also provides unique information on the integrity of the active part of the transformer. However, it is important to know that different effects can have an impact on the reproducibility of the test results [14], [15] and [16]. Below list summarizes the most common effects affecting the reproducibility:                  

Connection of non-tested windings Connection of the open delta winding Temperature and moisture Tap position Bushings (connections, contact resistances, etc.) Insulation oil condition Core grounding (whether core is grounded or not) Tank (whether tank is grounded or not) Measurement mistakes Point of injection of the test signal (phase vs. neutral) FRA instrument Connection technique Arrangement of cables Type of measurement (IFRA vs. SFRA) Measurement impedance (50Ω vs. high impedance input) Measurement cables (wave impedance, etc.) Electromagnetic interferences Residual magnetism of the iron core

Due to the effects that these factors may have in the results, it is crucial to document the measurement setup and conditions as good and complete as possible. The example in Figure 13 below shows the impact of residual magnetism of the iron core on the SFRA test results. One can clearly see that a magnetized core will affect the low frequency range where the magnetizing inductance of the iron core is dominating the frequency response. The magnetizing inductance of the iron core is heavily influenced by residual magnetism. Residual magnetism is causing the main inductivity to decrease. According to Thomson’s formula this will result in a higher resonance frequency:



f_resonance = 1 / 2 *  * L * C




(6)

This shift in the resonance frequency can clearly be seen in case of a magnetized core in below comparison.

Figure 13 Effect of residual magnetism on the SFRA test results

The effect of the tertiary winding treatment can be seen in Figure 14. The power transformer which was investigated had an open delta tertiary winding. According to international standards and the CIGRE guideline the open delta winding has to be closed but not connected to ground for all SFRA measurements. The medium frequency range which is dominated by the mutual coupling is affected most in this example. The mutual coupling between the winding to be tested and the tertiary windings changes depending on the tertiary winding treatment. Hence, differences in the SFRA response can be expected in the medium frequency range. A radial or axial movement of either the HV or LV winding would have a similar impact on the winding to be tested. Therefore, it is very important to pay particular attention to the tertiary winding configuration to make sure that a change of the SFRA response in the mutual coupling frequency range is not misinterpreted.


Figure 14 Effect of tertiary winding treatment

If the power transformer under test has a tap changer for regulating either voltage or phase shift the tap changer position, the winding was tested at, has to be documented. In Figure 15 the impact of the tap changer position can be seen. The tap changer position can affect the entire frequency range relevant for the assessment. When adding or removing turns it will not just affect the copper losses, leakage inductances, parasitic capacitances of the individual turns, but also the magnetizing inductance and parasitic capacitance coupled with the iron core. This explains why the response changes over the entire frequency range rather than just a portion of the range.


Figure 15 Effect of tap changer position

Case studies Case study #1: Power Transformer 150kV/20kV 30MVA A power transformer (150kV/20kV 30MVA) was investigated after it tripped on the tank pressure relay. Additionally, the Buchholz relay was giving an alarm. The first visual on-site inspection on the bushings, cooling system, OLTC and mechanical structure of the main tank did not reveal any obvious faults. The DGA result of the oil sample was revealing a high concentration of carbon-monoxide (CO) of 935ppm. A decomposition of the transformer paper insulation was suspected. The test results for tan delta and capacitance was not indicating any severe problem. Interestingly, the winding resistance measurement on phase B and phase C of the HV winding were aligning quite well. However, it was not possible to obtain any readings for phase A. It appears that the primary winding has an open circuit. It was decided to follow-up with an SFRA measurement. There were neither benchmark data from the same transformer nor a sister transformer available, thus the only option was a phase-based comparison. In Figure 16 the SFRA response of all three HV phases are shown. It can be clearly seen that phase A (H0 H1) shows capacitive behavior at low frequencies rather than inductive behavior. This is a clear indication of an open circuit in the winding.


Figure 16 SFRA response of the HV winding

In Figure 17 the SFRA response of the LV windings can be seen. The response of the LV winding is confirming the findings of response of the HV winding. Phase A of the LV windings shows huge differences in the mutual coupling frequency range, indicating a problem at the HV winding on phase A.


Figure 17 SFRA response of the LV winding

It was decided to open the transformer tank and inspect the active part of the power transformer. It was found that the lead connecting the main winding to the tap winding was interrupted (Figure 18), which had caused the capacitive rather than the inductive effect at low frequencies [17].


Figure 18 Fault on the lead connecting the main winding to the tap winding

Case study #2: Power Transformer 115kV/34.5kV 30MVA A power transformer (115kV/34.5kV 30MVA) tripped out on differential protection after one 115kV overhead line was falling on top of the 34.5kV line. The protection tripped first on phase B, shortly followed by the other two phases. The DGA was indicating a high concentration of Acetylene (C2H2) of 21ppm, indicating arcing in oil or overheating of the cellulose material. It was agreed to perform an SFRA measurement to check the integrity of the active part. A phase-based comparison was to be conducted as there was no fingerprint or sister unit available at the time. Figure 19 and 20 show the response of the HV winding and LV winding respectively.


Figure 19 SFRA response of the HV winding

The response of the HV winding is not indicating any severe deviations. All three phases line up nicely in the mutual coupling frequency range and the winding frequency range.

Figure 20 SFRA response of the LV winding

It can be clearly seen in Figure 20 that phase B shows some significant differences in the resonance frequencies compared to phase A and phase C. All main resonance frequencies in the winding frequency range are shifted towards lower frequencies. In the CIGRE report this behaviors has been classified as buckling effect [9]. Buckling effect can happen due to Lorenz’ Force when there are high short-circuit currents involved. There are two types of winding buckling: forced buckling and free buckling. Free buckling can happen when there is no mechanical support for the winding, where the mechanically weakest point of the winding can start to buckle. Forced buckling can happen when there is mechanical support for the winding. The conductors can bend between the supports all along the circumference. It was decided to pull-out the active part of the power transformer for further inspection. A visual inspection of the active part confirmed buckling on phase B (Figure 21). The burn marks on the cellulose material where caused by the welding machine used to open the lid of the transformer tank.


Figure 21 Buckling on phase B

Case study #3: Power Transformer 69kV/23kV 33MVA After a fault event a power transformer (69kV/23kV 33MVA) on a mobile substation was investigated. A time-based comparison was possible. The power transformer passed all standard electrical tests such as winding resistance, turns ratio, capacitance and tan delta measurements. In Figure 22 a comparison with the fingerprint (red plot) of the HV phase A-B is shown. The differences at lower frequencies are to due residual magnetism in the iron core. No obvious deviations can be seen. In Figure 23 a comparison of the suspected LV phase B is shown. There are obvious deviations in the high frequency part which is dominated by the parasitic capacitance and leakage inductance of the winding (red plot). The differences at very low frequencies are due to residual magnetism which causes a change in the magnetizing inductance. Based on the SFRA results it was decided to ship the transformer back to the factory for a closer inspection, despite the fact that all other standard electrical tests did not indicate any problem. The visual inspection of the active part revealed a mechanical problem of the secondary lead holder. The lead holder supporting the leads to the tap winding slipped down which caused a change in the response at high frequencies on the LV winding.


Figure 22 SFRA response of the HV winding phase B

Figure 23 SFRA response of the LV winding phase B www.omicronenergy.com | [email protected]

Figure 24 Secondary lead holder slipped down on phase B

Conclusion The SFRA measurement is a powerful method for detecting and diagnosing defects in the active part of a power transformer. Reliable information about the mechanical and electrical condition of the core, windings, internal leads and contacts can be gathered using the described diagnostic method. No other single test method for the condition assessment of power transformers can deliver such a diversity of information. On the other hand, the engineer relying on this method has to be aware of its limitations as well. The key for a successful application is the reproducibility. Therefore particular attention would be required on the test setup. A lot of work has been put into the international standards and the CIGRE guide for SFRA measurements. Acknowledgements This work is dedicated to my good friend, who also happens to be my co-worker, Wenyu Guo. It is his knowledge and passion for sweep frequency response analysis convinced me to write a paper about it.


References 1. T. Leibfried, K. Feser, Off-line- and On-line-Monitoring of Power Transformers using the Transfer Function Method, Conference Record of the 1996 IEEE International Symposium on Electrical Insulation, Montreal, Quebec, Canada, June 16-19,1996. 2. R. Wimmer, K. Feser, "Calculation of the transfer function of a power transformer with online measuring data", APTADM Wroclaw 2004. 3. T. Leibfried, K. Feser, Monitoring of Power Transformers using the Transfer Function Method, IEEE Transactions on Power Delivery, Vol. 14, No. 4, October 1999, pp.13331341. 4. L. Coffeen, J. McBride, D. Cantrelle, Initial Development of EHV Bus Transient Voltage Measurement: An Addition to On-line Transformer FRA, EPRI Substation Equipment Diagnostics Conference Orlando, FL March 2008. 5. R.C. Degeneff, "A General Method for Determining Resonances in Transformer Windings", IEEE Transactions on Power Apparatus and Systems, Vol. 96-2, 1977, pp. 423430. 6. Frequency Response Analysis on Winding Deformation of Power Transformers, The Electric Power Industry Standard of People’s Republic of China, Std. DL/T911-2004, ICS27.100, F24, Document No. 15182-2005, June 1st, 2005. 7. S.A. Ryder, Diagnosing Transformer Faults Using Frequency Response Analysis, IEEE Electrical Insulation Magazine March/April 2003. Vol. 19, No. 2, pp.16-22. 8. S. Ryder, S. Tenbohlen, A comparison of the swept frequency and impulse response methods for making frequency response analysis measurements, Doble Conference, 2003. 9. CIGRE WG A2.26, "Mechanical condition assessment of transformer windings: guidance, FRA standardization, further improvements", 2008. 10. IEEE C57.149-2012, IEEE Guide for the Application and Interpretation of Frequency Response Analysis for Oil-Immersed Transformers 11. IEC 60076-18 Edition 1.0 2012-07 Power transformers – Part 18: Measurement of frequency response 12. Alexander Kraetge, Michael Krüger, Juan L. Velasquez, Maximilian Heindl, Stefan Tenbohlen, Experiences with the practical application of Sweep Frequency Response Analysis (SFRA) on power transformers 13. A. Kraetge, M. Krüger, P. Fong, Frequency Response Analysis – Status of the worldwide standardization activities


14. N. Abeywickrama, Y.V. Serdyuk, S.M. Gubanski, Effect of Core Magnetization on Frequency Response Analysis (FRA) of Power Transformers, IEEE Transactions on Power Delivery, Vol. 23, No. 3, 2008. 15. S. Tenbohlen, R. Wimmer, K. Feser, A. Kraetge, M. Krüger, J. Christian, The influence of grounding and connection technique on the repeatability of FRA-results, Proceedings of the XVth International Symposium on High Voltage Engineering, University of Ljubljana, Ljubljana, Slovenia, August 27-31, 2007. 16. Frequency Response Analysis on Winding Deformation of Power Transformers, The Electric Power Industry Standard of People’s Republic of China, Std. DL/T911-2004, ICS27.100, F24, Document No. 15182-2005, June 1st, 2005. 17. H.I. Septyani, H. Maryono, A.P. Purnomoadi, U. Sutisna, Sweep Frequency Response Analysis for Assessing Transformer Condition after an Incident Biography Florian Predl (12 November 1986) is currently employed as an Field Application Engineer at OMICRON Australia. He commenced with OMICRON Austria in 2007 as an application engineer with special focus on advanced instrument transformer diagnostics within the Engineering Services team. In 2013 Florian joined the OMICRON team in Australia. [email protected]



OMICRON-SFRA-measurement-results-Predl-Paper-omicronized.pdf

Recommend Documents