Partial discharge monitoring method and system

• US 9,229,042 B2
• Filed: 07/22/2009
• Issued: 01/05/2016
• Est. Priority Date: 08/06/2008
• Status: Active Grant

• Current Assignee: Eskom Holdings Limited
• Original Assignee: Eskom Holdings SOC Limited
• Inventors: Higgins, Simon
• Primary Examiner(s): Phan, Huy Q
• Assistant Examiner(s): RHODES-VIVOUR, TEMILADE S
• Application Number:
  US13/057,712
• Publication Number:
  US 20110248721A1
• Time in Patent Office: 2,358 Days
• Field of Search: 324405-413, 324/420, 324509-513, 324520-537
• US Class Current: 1/1
• CPC Class Codes: G01R 1/00 Details of instruments or arrangements of the types included in groups G01R5/00 - G01R13/00 and G01R31/00 constructional details particular to electromechanical arrangements for measuring the electric consumption G01R11/02
• CPC Class Codes: G01R 31/1272 of cable, line or wire insulation, e.g. using partial discharge measurements locating faults in cables G01R31/083
• CPC Class Codes: H01B 1/00 Conductors or conductive bodies characterised by the conductive materials; Selection of materials as conductors
• CPC Class Codes: H01L 21/00 Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
• CPC Class Codes: H01L 2221/00 Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof covered by H01L21/00
• CPC Class Codes: H02B 1/00 Frameworks, boards, panels, desks, casings; Details of substations or switching arrangements

Abstract

The invention relates to a method and system of monitoring partial discharges occurring in an electrical system, and to a method of measuring or analyzing partial discharges occurring in an electrical system. The method comprising receiving a signal or impulse, or information associated therewith, from the electrical system; breaking the received signal or impulse into predefined frequency components; and displaying a peak of the received signal or impulse on a scatter plot with other peaks associated with similar predefined frequency components.

Claims

1. A method of monitoring partial discharges occurring in an electrical system, the method comprising:

receiving a pulse from the electrical system;

identifying if the pulse is noise or a duplicate pulse;

if the pulse is not noise or a duplicate pulse then converting the pulse from an analogue to a digital signal;

breaking the pulse into two or more predefined frequency components;

normalizing the two or more predefined frequency components to a maximum level;

comparing the amplitudes of the two or more normalized predefined frequency components associated with the received pulse with amplitudes of other stored pulses at the same predefined frequency components to identify similar pulses which indicate a known fault condition;

if the pulse is identified as a pulse which indicates the known fault condition, then storing data in a database associating the pulse with the two or more normalized predefined frequency components and the known fault condition;

grouping pulses with similar normalized predefined frequency components in a scatter plot stored in the database;

if the normalized predefined frequency components of the pulse are not similar to the normalized frequency components of a current grouping then creating a new grouping of pulses stored in the database; and

if the pulse is identified as a pulse which indicates the known fault condition, then notifying a user that the fault condition exists.

2. The method as claimed in claim 1, the method comprising generating a scratchpad to store a predefined number of normalized spectra for different pulses together with peak values associated with pulses.

3. The method as claimed in claim 1, the method comprising performing signal processing on the received pulse.

4. The method as claimed in claim 1, wherein comparing the frequency spectrum associated with the detected or received pulse with the existing fault spectra stored in the database is done by way of a fault matching algorithm.

10. The method as claimed in claim 1, comprising generating scatter plots from a sub-set of a fault spectra of the received pulse.

11. The method according to claim 4, wherein the fault matching algorithm is the sum of squared differences as given in the following equation:

Sum of squared differences,

12. The method of claim 1, wherein the step of identifying if the pulse is noise or a duplicate pulse comprises

determining a direction of travel of the received pulse,

determining if the received pulse is noise or not by at least determining if the pulse occurs in two or more phases of the electrical system, and

determining if the received pulse is cross coupled between the two phases.

5. A system for monitoring partial discharges occurring in an electrical system, the system comprising:

a database for storing a plurality of fault spectra;

a monitoring module for monitoring the electrical system to receive pulses occurring in the electrical system; and

an analog to digital converter to convert the pulse from an analog signal to a digital signal;

a validation module arranged to validate the received pulse by identifying if the pulse is noise or a duplicate pulse;

a peak detector to determine a peak value of the received pulse;

a frequency spectrum generating module operable to generate a frequency spectrum of the pulse received by the monitoring module by breaking the pulse into two or more predefined frequency components;

a normalizing module arranged to normalize the peak values of each of the two or more predefined frequency components of the received pulse to a maximum level;

a comparator operable, once a pulse is received, to;

compare the amplitudes of the two or more normalized predefined frequency components associated with the received pulse with amplitudes of other stored pulses at the same predefined frequency components to identify similar pulses which indicate a known fault condition;

if the pulse is identified as a pulse which indicates the known fault condition, then storing data in a database associating the pulse with the two or more normalized predefined frequency components and with the known fault condition;

group similar pulses in a scatter plot stored in the database; and

a display to display a point on the scatter plot representing the digitized peak of the received pulse together with points representing other identified similar pulse peaks.

6. The system as claimed in claim 5, the system being arranged to raise a flag if a frequency spectrum of the received pulse substantially matches any of the existing fault spectra.

7. The system as claimed in claim 6, wherein the system is arranged to store data or the frequency spectrum of the received pulse in the database.

8. The system as claimed in claim 5, wherein the system comprises a data generating module operable to generate data associated with the received pulse.

9. The system as claimed in claim 5, wherein the monitoring module is in communication with a plurality of sensors, the sensors being arranged to monitor each phase of the electrical or power system.