Machine learning based method and device for disturbance classification in a power transmission line
The present specification provides a method and device for determining a disturbance condition in a power transmission line. The method may include obtaining a plurality of sample values corresponding to an electrical parameter measured in each phase. The method may further include determining a plurality of magnitudes of the electrical parameter corresponding to each phase based on the corresponding plurality of sample values and determining a plurality of difference values for each phase based on the corresponding plurality of magnitudes. The method may include processing the plurality of difference values using a machine learning technique to determine the disturbance condition. The disturbance condition may be one of a load change condition, a power swing condition, or an electrical fault condition. The method may also include performing at least one of a protection function or a control function based on the disturbance condition.
1. A method for determining a disturbance condition in a power transmission line, wherein the method is performed by an Intelligent Electronic Device (IED) communicatively coupled with a measurement equipment connected with the power transmission line, the method comprising:
obtaining a plurality of sample values corresponding to an electrical parameter measured in each phase among one or more phases, wherein the electrical parameter in the one or more phases is measured with the measurement equipment connected at one end of the power transmission line;
determining a plurality of magnitudes of the electrical parameter corresponding to each phase based on the corresponding plurality of sample values;
determining a plurality of difference values for each phase based on the corresponding plurality of magnitudes, wherein each of the plurality of difference values is representative of a rate of change of magnitudes for the corresponding phase, wherein determining the plurality of difference values comprises
detecting the disturbance condition based on comparing each of the plurality of difference values with a pre-determined threshold value, and
selecting a subset of the plurality of difference values subsequent to the detection of the disturbance condition;
processing only the selected subset of the plurality of difference values corresponding to the one or more phases using a machine learning technique to determine a type of the disturbance condition in the power transmission line, wherein the disturbance condition is one of a load change condition, a power swing condition, or an electrical fault condition; and
performing at least one of a protection function or a control function based on the disturbance condition.
2. The method of claim 1 , wherein the plurality of sample values corresponds to a plurality of voltage samples obtained from sampling a voltage signal.
3. The method of claim 1 , wherein determining the plurality of difference values comprises performing a smoothing operation on the plurality of magnitudes corresponding to each phase among the one or more phases by filtering the corresponding magnitudes.
4. The method of claim 3 , wherein the plurality of difference values is determined by computing difference between successive samples of the smoothed version of the plurality of magnitudes.
5. The method of claim 1 , wherein the one or more phases correspond to phases of a three-phase electrical system.
6. The method of claim 1 , wherein the plurality of sample values is obtained by sampling in the corresponding phase with one kilohertz sampling rate.
7. The method of claim 1 , wherein each of the plurality of difference values is a difference between two successive ones of the magnitudes.
8. The method of claim 1 , wherein the disturbance condition is the power swing condition.
9. The method of claim 1 , wherein the machine learning technique is based on an ensemble machine learning technique, wherein the ensemble machine learning technique comprises an extreme gradient boost classification model trained for determining the disturbance condition.
10. The method of claim 9 , wherein performing at least one of a protection function or a control function based on the disturbance condition comprises a blocking operation of a switching device using the IED during a self-restoring power swing condition or an unblocking operation of the switching device during fault conditions.
11. The method of claim 10 , wherein the extreme gradient boost classification model comprises a plurality of residual models.
12. An Intelligent Electronic Device (IED) for determining a disturbance condition in a power transmission line, the IED comprising:
a data acquisition unit communicatively coupled to a measurement equipment connected at one end of a power transmission line and configured to obtain a plurality of sample values corresponding to an electrical parameter measured in each phase among one or more phases, wherein the measurements are performed by the measurement equipment;
a signal processing unit communicatively coupled to the data acquisition unit and configured to:
determine a plurality of magnitudes of the electrical parameter corresponding to each phase among the one or more phases based on the corresponding plurality of sample values, and
determine a plurality of difference values for each phase among the one or more phases based on the corresponding plurality of magnitudes, wherein each of the plurality of difference values is representative of rate of change of magnitudes for the corresponding phase, wherein determining the plurality of difference values comprises
detecting the disturbance condition based on comparing each of the plurality of difference values with a pre-determined threshold value, and
selecting a subset of the plurality of difference values subsequent to the detection of the disturbance condition;
a machine learning unit communicatively coupled to the signal processing unit and configured to process only the selected subset of the plurality of difference values corresponding to one or more phases using a machine learning technique to determine a type of disturbance condition in the power transmission line, wherein the disturbance condition comprises one of a load change condition, a power swing condition, or a transmission line fault condition; and
a control unit communicatively coupled to the machine learning unit and configured to perform at least one of a protection function or a control function based on the disturbance condition.
13. The IED of claim 12 , wherein the plurality of sample values corresponds to a plurality of voltage samples obtained from a voltage signal.
14. The IED of claim 12 , wherein the signal processing unit is further configured to generate a smoothed version of the plurality of magnitudes corresponding to each of the plurality of phase voltage signals by filtering the corresponding magnitudes.
15. The IED of claim 14 , wherein the signal processing unit is configured to determine the plurality of difference values based on the smoothed version of the plurality of magnitudes.
16. The IED of claim 12 , wherein the data acquisition unit is configured to obtain the plurality of samples from one or more phases of a three-phase electrical system.
17. The IED of claim 12 , wherein the data acquisition unit is configured to sample the electrical parameter in each phase among the one or more phases with one kilohertz sampling rate.
18. The IED of claim 12 , wherein each of the plurality of difference values is a difference between two successive ones of the magnitudes.
19. The IED of claim 12 , wherein the disturbance condition is the power swing condition.
20. The IED of claim 12 , wherein:
the machine learning unit is further configured to train an ensemble machine learning model for determining the disturbance condition; and
the control unit is further configured to train an extreme gradient boost classification model for determining the disturbance condition.
21. The IED of claim 20 , wherein the control unit is configured to block operation of a switching device using the IED during a self-restoring power swing condition or unblock operation of the switching device during fault conditions.
22. The IED of claim 21 , wherein the machine learning unit is configured to determine a plurality of residual models.