Process for identifying abnormalities in power transformers

• US 20030189431A1
• Filed: 12/04/2002
• Published: 10/09/2003
• Est. Priority Date: 04/03/2002
• Status: Active Grant

• Current Assignee: Hitachi Energy Switzerland AG (Hitachi, Ltd.)
• Original Assignee: ABB Technology AG (ABB Limited)
• Inventors: Perkins, Mark D., Fazlagic, Asim
• Granted Patent: US 6,870,374 B2
• Time in Patent Office: Days
• US Class Current: 324/509
• CPC Class Codes: G01R 27/2688 Measuring quality factor or dielectric loss, e.g. loss angle, or power factor power factor related to power measurements G01R21/006; testing capacitors G01R31/016
• CPC Class Codes: G01R 31/1263 of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
• CPC Class Codes: G01R 31/52 Testing for short-circuits, leakage current or ground faults
• CPC Class Codes: G01R 31/62 Testing of transformers

Abstract

A preferred method for identifying a type of abnormality in an insulation system of a power transformer comprises measuring dielectric losses in a section of the insulation system, calculating theoretical dielectric losses for the section based on the material properties, geometry, and temperature of the section, and generating a graphical representation of a percent difference between the measured and calculated dielectric losses. The preferred method also comprises comparing a shape of the graphical representation with a shape of one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality.

Claims

1. A method for identifying a type of abnormality in an insulation system of a power transformer, comprising:

measuring dielectric losses in a section of the insulation system;

calculating theoretical dielectric losses for the section based on the material properties, geometry, and temperature of the section;

generating a graphical representation of a percent difference between the measured and calculated dielectric losses; and

comparing a shape of the graphical representation with a shape of one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality.

2. The method of claim 1, wherein measuring a dielectric loss in the section comprises performing a dielectric frequency response test on the section.

3. The method of claim 2, wherein performing a dielectric frequency response test on the section comprises performing a plurality of power-factor tests on the section, each of the power factor tests being conducted using an input-voltage frequency different from an input-voltage frequency used for the other ones of the power factor tests.

4. The method of claim 1, wherein generating a graphical representation of a percent difference between the measured and theoretical dielectric losses comprises generating a dielectric frequency response signature for the section.

5. The method of claim 4, wherein generating a dielectric frequency response signature for the section comprises calculating a difference between each of the theoretical dielectric losses and a corresponding one of the measured dielectric losses, and calculating a quotient of the difference and a corresponding one of the theoretical dielectric losses by dividing the difference by the corresponding one of the theoretical dielectric losses.

6. The method of claim 5, wherein generating a dielectric frequency response signature for the section further comprises multiplying the quotient by one hundred.

8. The method of claim 1, further comprising matching the shape of the graphical representation of a percent difference between the measured and calculated dielectric losses with a shape of one of the one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality.

9. The method of claim 1, further comprising identifying the section as having an abnormality by comparing the measured dielectric losses in the section to the theoretical dielectric losses for the section.

10. The method of claim 1, further comprising electrically isolating the section before measuring the dielectric losses in the section.

11. The method of claim 1, wherein comparing a shape of the graphical representation with a shape of one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality comprises determining whether the shape of the graphical representation substantially differs from the shape of the one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality.

12. The method of claim 1, wherein calculating theoretical dielectric losses for the section based on the material properties, geometry, and temperature of the section comprises calculating the theoretical dielectric losses without accounting for effects of any abnormalities in the section.

7. The method of claim I, further comprising reading the one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of insulation each known to have a specific type of abnormality from a data base.

13. A method, comprising:

measuring a dielectric frequency response of a section of insulation within a power transformer;

calculating a theoretical dielectric frequency response for the section;

generating a dielectric frequency response signature for the section; and

comparing the dielectric frequency response signature to the dielectric frequency response signature of other sections of the insulation known to have particular types of abnormalities.

14. The method of claim 13, further comprising identifying the presence of an abnormality in the section by comparing the measured dielectric frequency response to the theoretical dielectric frequency response.

15. The method of claim 13, further comprising identifying the presence of an abnormality in the section by performing a power factor test on the section.

16. A method for classifying a type of abnormality in a section of insulation in a power transformer, comprising:

energizing a conductor insulated by the section of insulation with an alternating-current voltage;

varying a frequency of the alternating-current voltage between a plurality of discrete frequencies within a predetermined range;

measuring the dielectric loss in the section of insulation at each of the discrete frequencies;

calculating a theoretical dielectric loss for the section of insulation at each of the discrete frequencies;

calculating a percent difference between the measured dielectric loss and the theoretical dielectric loss at each of the discrete frequencies;

generating a substantially curvilinear representation of the percent differences between the measured dielectric loss and the calculated dielectric loss at each of the discrete frequencies; and

comparing a shape of the substantially curvilinear representation to a shape of one ,or more predetermined curvilinear representations of a percent difference between a measured dielectric loss and a calculated dielectric loss for a section of insulation known to have a particular type of abnormality.

17. The method of claim 16, further comprising electrically isolating the conductor from other conductors in the power transformer before energizing the conductor.

18. The method of claim 16, further comprising identifying the section of insulation as having an abnormality by comparing at least one of the measured dielectric losses to a corresponding one of theoretical dielectric losses.

19. The method of claim 16, wherein varying a frequency of the alternating-current voltage between a plurality of discrete frequencies within a predetermined range comprises varying the frequency of the alternating-current voltage between approximately 0.01 hertz and approximately 1,000 hertz.

20. The method of claim 16, further comprising reading the one or more predetermined curvilinear representations of a percent difference between a measured dielectric loss and a calculated dielectric loss for a section of insulation known to have a particular type of abnormality from a data base.

21. The method of claim 16, further comprising matching the substantially curvilinear representation of the percent differences between the measured dielectric loss and the calculated dielectric loss at each of the discrete frequencies to one of the one or more predetermined curvilinear representations of a percent difference between a measured dielectric loss and a calculated dielectric loss for a section of insulation known to have a particular type of abnormality.

22. The method of claim 16, wherein calculating a theoretical dielectric loss at each of the discrete frequencies comprises calculating the theoretical dielectric loss at each of the discrete frequencies with accounting for any abnormalities in the section of insulation.

23. A method, comprising:

measuring a dielectric loss in a section of an insulation system of a power transformer in response to an alternating-current voltage varied over a predetermined range of frequencies;

calculating a theoretical dielectric loss for the section based on the material properties, geometry, and temperature of the section of insulation;

generating a representation of a percent difference between the measured and calculated dielectric losses; and

determining whether a shape of the representation substantially differs from a representation of a percent difference between measured and calculated dielectric losses in another section of the insulation previously identified as having a specific type of abnormality.

24. A method for identifying a type of abnormality in an insulation system of a power transformer, comprising:

measuring dielectric losses in a section of the insulation system;

calculating theoretical dielectric losses for the section;

identifying the section as having an abnormality by comparing the measured dielectric losses in the section to the theoretical dielectric losses for the section;

generating a graphical representation of a percent difference between the measured and calculated dielectric losses; and

comparing a shape of the graphical representation with a shape of one or more graphical representations of a percent difference between measured and theoretical dielectric losses in one or more other sections of the insulation each known to have a specific type of abnormality.

25. The method of claim 24, further comprising electrically isolating the section of the insulation system before measuring the dielectric losses in the section of the insulation system.

26. A method, comprising:

measuring a dielectric frequency response of a section of insulation within a power transformer;

calculating a theoretical dielectric frequency response for the section;

identifying the section as having an abnormality therein by comparing the measured dielectric response and the theoretical dielectric response;

generating a dielectric frequency response signature for the section; and

comparing the dielectric frequency response signature to a dielectric frequency response signature of other sections of the insulation known to have particular types of abnormalities.