Research Article

Utilization of Stockwell Transform and Random Forest Algorithm for Efficient Detection and Classification of Power Quality Disturbances

Table 10

Complete performance indices of all algorithms.

PQD classPrecisionRecallF-scoreAccuracyBest algorithm
KNNSVMDTRFKNNSVMDTRFKNNSVMDTRFKNNSVMDTRF

C11.001.001.001.001.001.001.001.001.001.001.001.00100100100100All
C20.930.860.981.000.920.950.981.000.920.900.981.0092.095.397.9100RF
C30.981.001.001.000.981.001.001.000.981.001.001.0098.1100100100DT and RF
C40.940.980.991.000.930.960.991.000.930.970.991.0093.096.099.1100RF
C50.900.970.950.960.870.930.930.980.880.950.940.9786.892.892.897.9RF
C60.951.000.981.000.960.980.971.000.950.990.971.0096.098.096.8100RF
C70.920.991.001.000.910.961.001.000.910.971.001.0090.996.1100100DT and RF
C80.871.001.001.000.921.001.001.000.891.001.001.0091.9100100100SVM, DT, and RF
C90.960.880.960.980.950.920.960.960.950.900.960.9794.992.096.096.0DT and RF
C100.890.911.001.000.900.970.991.000.890.940.991.0090.096.998.7100RF
C110.920.830.960.960.960.910.980.970.940.870.970.9696.090.598.096.8DT
C120.930.920.930.970.890.900.970.950.900.900.950.9689.090.996.994.8DT
C130.960.951.001.000.950.940.991.000.950.940.991.0095.094.298.9100RF
C140.920.940.971.000.940.850.981.000.930.890.971.0094.184.998.1100RF
C150.900.910.961.000.880.870.950.980.890.890.950.9987.986.995.098.0RF
C160.941.001.001.000.981.001001.000.961.001.001.0097.2100100100SVM, DT, and RF
C170.951.001.000.970.960.980.991.000.950.980.990.9895.297.799.499.7RF