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

Table 6

Confusion matrix of KNN.

True PQDs	Predicted PQDs by KNN
	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	C13	C14	C15	C16	C17
C1	353
C2		312		12		3					1				11
C3			368			60										1
C4		11		307		2					1						9
C5					291				5			9		30
C6		1	1		0	338									11	1
C7					14		351			7			14
C8								332		16					13
C9		3		1					359		15
C10			6				26			291
C11		1		1					10		336	2
C12								36				292
C13						1	5					12	342
C14					11			11						350
C15		3			8					3	10				304	9	9
C16										10						345
C17		5		6		6											340
Overall accuracy = 93.41%