Traveling wave based single end fault location

• US 10,295,585 B2
• Filed: 11/08/2017
• Issued: 05/21/2019
• Est. Priority Date: 11/11/2016
• Status: Active Grant

• Current Assignee: Schweitzer Engineering Laboratories, Incorporated
• Original Assignee: Schweitzer Engineering Laboratories, Incorporated
• Inventors: Schweitzer, III, Edmund O., Guzman-Casillas, Armando, Kasztenny, Bogdan Z., Tong, Yajian, Mynam, Mangapathirao Venkata
• Primary Examiner(s): Phan, Huy Q
• Assistant Examiner(s): Frederiksen, David B
• Application Number:
  US15/806,959
• Publication Number:
  US 20180136269A1
• Time in Patent Office: 559 Days
• CPC Class Codes: G01R 31/085 in power transmission or distribution lines, e.g. overhead
• CPC Class Codes: G01R 31/11 using pulse reflection methods

Abstract

Traveling wave information from a single end of an electric power delivery system is used to determine a fault location on a power line of the electric power delivery system. Hypotheses of which of a plurality of received traveling waves represents a first reflection from the fault are evaluated. A determination of an arrival time of the first reflection from the fault is used to calculate a distance from the single end of the power line to the fault location.

Claims

1. A system to determine a relative location of a fault in a power line of an electric power delivery system comprising:

a traveling wave detector in electrical communication with a first location on the power line of the electric power delivery system to;

detect an arrival time (TPK) of each of a plurality of traveling wave amplitude peaks (VPKs) received at the first location on the power line, and

identify hypothesis TPKs as TPKs corresponding to VPKs detected after a first detected VPK that have the same polarity as the first detected VPK at the first location on the power line; and

a fault location determination subsystem comprising at least one processor and a memory to;

calculate a fault reflection value, F(H), for each hypothesis TPK,

calculate remote reflection value, R(H), associated with a remote location on the power line, for each hypothesis TPK using the hypothesis TPK, a length of the power line between the first location and the remote location, and traveling wave propagation time (TWLPT),

compare the F(H) and the R(H) of each hypothesis TPK with a set of time difference values (DT set) that includes time difference values between the pair-combinations of detected TPKs associated with VPKs received after the first detected VPK,

determine a first number of matches (NM) as the number of time difference values in the DT set that match the F(H) of each hypothesis TPK, and

determine a second number of matches (N1_M) as the number of time difference values in the DT set that match the R(H) of each hypothesis TPK; and

a fault distance calculation subsystem to calculate a distance from the first location to a fault location based, at least in part, on;

(i) the length of the power line between the first location and the remote location,

(ii) the traveling wave line propagation time (TWLPT), and

(iii) one of the hypothesis TPKs selected based on a function of NM and N1_M.

2. The system of claim 1, wherein the traveling wave detector is configured to detect the TPKs of each of the VPKs received within an observation window.

3. The system of claim 1, wherein the fault location determination subsystem is configured to calculate the F(H) of each hypothesis TPK as corresponding to a difference between the hypothesis TPK and the TPK of the first detected VPK.

4. The system of claim 1, wherein the traveling wave detector is configured to detect the TPKs of each of the VPKs received after the first detected VPK relative to the TPK of the first detected VPK, such that the F(H) of a hypothesis TPK is equal to the hypothesis TPK.

5. The system of claim 1, wherein the fault location determination subsystem is configured to calculate the R(H) of each hypothesis TPK as corresponding to twice the TWLPT less the F(H) value of the hypothesis, such that for each hypothesis TPK, R(H)=2*TWLPT−

F(H).

6. The system of claim 1, wherein the fault location determination subsystem is further configured to:

determine a set of expected traveling wave arrival times (ETWs) for each hypothesis TPK;

determine a third number of matches (NS), as the number of detected TPKs that match the ETWs for each hypothesis TPK;

assign a weight factor (WGHT) to each hypothesis TPK equal to;

the NS associated with each hypothesis TPK for which the associated R(H) corresponds to one detected TPK, and

zero (0) for each hypothesis TPK for which the associated R(H) value does not correspond to any detected TPK;

calculate a merit value (N) for each hypothesis TPK equal to the sum of the respective values of NM, N1_M, and WGHT, such that for each hypothesis TPK;

N=NM+N1_M+WGHT; and

wherein the fault distance calculation subsystem is configured to calculate the distance from the first location to the fault location based on the hypothesis TPK that has the highest N.

7. The system of claim 1, wherein the fault location determination subsystem is configured to determine a match when two numbers are within a threshold range of one another.

8. The system of claim 1, wherein the fault distance calculation subsystem is configured to calculate the distance from the first location to the fault location as one-half the length of the power line, multiplied by the selected hypothesis TPK, divided by the TWLPT.

9. The system of claim 1, wherein the selected hypothesis TPK is the hypothesis TPK having the highest NM+N1_M.

10. A method of calculating a relative location of a fault in a power line of an electric power delivery system comprising:

detecting, via at least one sensor in electrical communication with a first location of the power line of the electric power delivery system, an arrival time (TPK) of each of a plurality of traveling wave amplitude peaks (VPKs) at the first location on the power line;

identifying TPKs corresponding to VPKs detected after a first detected VPK that have the same polarity as the first detected VPK at the first location on the power line as hypotheses;

calculating, via a processor, a fault reflection value, F(H), for each hypothesis;

calculating a remote reflection value, R(H), for each hypothesis using the hypothesis TPK, a length of the power line between the first location and the remote location, and traveling wave propagation time (TWLPT);

determining a first number of matches (NM) of the F(H) of each hypothesis with a set of time differences (DT set) between pair-combinations of the detected TPKs received after the first detected VPK;

determining a second number of matches (N1_M) of the R(H) with the time differences in the DT set; and

calculating a location of the fault relative to the first location based on the length of the power line, the traveling wave line propagation time (TWLPT), and a selected hypothesis of the hypotheses.

11. The method of claim 10, further comprising:

determining a set of expected traveling wave arrival times (ETWs) for each hypothesis;

determining a third number of matches (NS) of the ETWs of each hypothesis with the detected TPKs; and

assigning a weight factor (WGHT) to each hypothesis equal to;

the NS associated with each hypothesis for which the associated R(H) corresponds to one detected TPK, and

zero (0) for each hypothesis for which the associated R(H) value does not correspond to any detected TPK,

wherein the selected hypothesis is the hypothesis determined to have the highest merit value (N) equal to the sum of the respective values of NM, N1_M, and WGHT, such that for each hypothesis;

N=NM+N1_M+WGHT.

12. The method of claim 10, wherein the selected hypothesis is the hypothesis determined to have a highest merit value (N), where N is calculated for each hypothesis as a function of the associated NM and N1_M.

13. The method of claim 10, wherein calculating the F(H) of each hypothesis comprises determining a time difference between the hypothesis and the TPK of the first detected VPK.

14. The method of claim 10, wherein calculating the R(H) of each hypothesis comprises determining the difference between twice the TWLPT and the F(H) value of the respective hypothesis, such that for each hypothesis:

R(H)=2*TWLPT F(H).

15. The method of claim 10, wherein determining a first number of matches (NM) and determining the second number of matches (N1_M) comprises identifying numbers that are within a threshold range of one another.

16. The method of claim 10, calculating the distance from the first location to the location of the fault comprises multiplying one-half the length of the power line by the selected hypothesis divided by the TWLPT.

17. The method of claim 10, detecting, the TPKs of the VPKs at the first location on the power line comprises only detecting those VPKs with TPKs within an observation window.

18. The method of claim 17, wherein the observation window includes TPKs received before the TPK of the first detected VPK+(2.4*TWLPT).

19. A computer-implemented method for calculating a relative location of a fault in a power line of an electric power delivery system comprising:

determining an initial guess of a relative location of a fault (m_ini) as being located within one of a first section of a power line, a second section of the power line, and a third section of a power line;

detecting, via at least one sensor in electrical communication with a first location on the power line of the electric power delivery system, an arrival time (TPK) of each of a plurality of traveling wave amplitude peaks (VPKs) at the first location on the power line; and

identifying TPKs corresponding to VPKs detected after a first detected VPK that have the same polarity as the first detected VPK as hypotheses;

calculating, via a processor, a fault reflection value, F(H), for each hypothesis;

calculating a remote reflection value, R(H), for each hypothesis using the hypothesis TPK, a length of the power line between the first location and the remote location, and traveling wave propagation time (TWLPT);

determining a first number of matches (NM) of the F(H) of each hypothesis with a set of time differences (DT set) between pair-combinations of the detected TPKs received after the first detected VPK;

determining a second number of matches (N1_M) of the R(H) with the time differences in the DT set;

determining a set of expected traveling wave arrival times (ETWs) for each hypothesis;

determining a third number of matches (NS) of the ETWs of each hypothesis with the detected TPKs;

assigning a weight factor (WGHT) to each hypothesis equal to;

the NS associated with each hypothesis for which the associated R(H) corresponds to one detected TPK, and

zero (0) for each hypothesis for which the associated R(H) value does not correspond to any detected TPK;

calculating a merit value (N) for each hypothesis that corresponds to the sum of the respective values of NM, N1_M, and WGHT, such that for each hypothesis;

N=NM+N1_M+WGHT;

sorting the hypothesis in a priority order based on;

NM from highest to lowest, where m_ini is determined to be within the first section of the power line,

N from highest to lowest, where m_ini is determined to be within the second section of the power line, and

N1_M from highest to lowest, where m_ini is determined to be within the third section of the power line; and

calculating a distance from the first location to the fault location based on a function of the length of the power line, the traveling wave line propagation time (TWLPT), and the hypothesis sorted to have the highest priority.

20. The computer-implemented method of claim 19, wherein the m_ini is determined to be within one of the first section of a power line, the second section of the power line, and the third section of a power line based on information received from one of:

a double-ended traveling wave fault location estimation system, a double-ended impedance fault location estimation system, and a single-ended impedance fault location estimation system.