A Multirate Sensor Information Fusion Strategy for Multitask Fault Diagnosis Based on Convolutional Neural Network

Table 6

Summary of components and their simulated fault conditions of the hydraulic dataset. A load cycle with 60 s duration is repeated 2205 times with the distribution of instances as indicated by the examples.

Component	Fault conditions	States	Abbreviation	Examples
Cooler (C1)	Cooling power decrease	3%: close to total failure	CTFc	732
		20%: reduced efficiency	RE	732
		100%: full efficiency	FE	741
Valve (V10)	Switching degradation	73%: close to total failure	CTFv	360
		80%: severe lag	Sel	360
		90%: small lag	SmL	360
		100%: optimal behaviour	OB	1125
Pump (MP1)	Internal leakage	0: no leakage	NL	1221
		1: weak leakage	WL	492
		2: severe leakage	SL	492
Accumulator (A1-A4)	Gas leakage	130 bar: optimal pressure	OP	599
		115 bar: slightly reduced	SlR	399
		100 bar: severely reduced	SeP	399
		90 bar: close to total failure	CTFa	80 8
Stable flag (S1)	——	0: conditions were stable	SC	1449
		1: nonstatic conditions	NSC	756