A Novel Data-Driven Fault Diagnosis Algorithm Using Multivariate Dynamic Time Warping Measure

Table 1

FDRs (%) and FARs (%) comparison with classical PM-FD methods on TE data sets given in [24].

Fault (free)	PCA	DPCA	ICA	MICA	FDA	PLS	TPLS	MPLS	SAP	Proposed
IDV(1)	99.88	99.88	100	99.88	100	99.88	99.88	100	99.63	100
IDV(2)	98.75	99.38	98.25	98.25	98.75	98.63	98.88	98.88	97.88	100
IDV(4)	100	100	100	87.63	100	99.50	100	100	99.88	100
IDV(5)	33.63	43.25	100	100	100	33.63	100	100	100	89.47
IDV(6)	100	100	100	100	100	100	100	100	100	100
IDV(7)	100	100	100	100	100	100	100	100	99.88	100
IDV(8)	98.00	98.00	98.25	97.63	98.13	97.88	98.50	98.63	95.88	100
IDV(12)	99.13	99.25	99.88	99.88	99.75	99.25	99.63	99.88	99.88	99.15
IDV(13)	95.38	95.38	95.25	95.00	95.63	95.25	96.13	95.50	94.88	96.61
IDV(14)	100	100	100	99.88	100	100	100	100	97.63	100
IDV(17)	95.25	97.25	96.88	93.00	96.63	94.25	96.00	97.13	97.13	98.48
IDV(18)	90.50	90.88	90.50	89.75	90.75	90.75	91.88	91.25	91.00	94.87
IDV(10)	60.50	72.00	89.25	85.88	87.13	82.63	91.00	91.13	95.50	83.54
IDV(11)	78.88	91.50	78.88	61.63	73.38	78.63	86.13	83.25	84.75	89.47
IDV(16)	55.25	67.38	92.38	83.38	83.25	68.38	90.75	94.28	94.88	83.67
IDV(19)	41.13	87.25	92.38	80.25	87.88	26.00	82.88	94.25	88.50	77.55
IDV(20)	63.38	73.75	91.38	86.00	81.88	62.75	78.38	91.50	83.75	84.21
IDV(21)	52.13	61.00	56.38	70.75	52.75	59.88	66.38	72.75	38.63	89.39
IDV(3)	12.88	12.25	4.50	14.25	7.00	14.25	24.25	18.75	6.38	35.90
IDV(9)	8.38	12.88	4.75	8.88	6.25	14.50	23.50	12.13	0.88	48.72
IDV(15)	14.13	19.75	7.75	10.75	12.63	23.00	29.88	23.25	29.50	38.46
IDV(0)	6.13	10.13	2.75	1.63	6.38	10	19.62	10.75	1.5	3.32