A Spatio-Temporal Attention Mechanism Based Approach for Remaining Useful Life Prediction of Turbofan Engine

Table 4

Comparison of experimental results.

Method	FD001		FD002		FD003		FD004
	RMSE	Score	RMSE	Score	RMSE	Score	RMSE	Score
Spatio-temporal attention	11.07	312.55	18.10	1339.5	10.73	329.95	17.00	2476.90
BiGRU-AS [6]	13.68	284	20.81	2454	15.53	428	27.31	4708
Ensemble ResCNN [9]	12.16	212.48	20.85	2087.77	12.01	180.76	24.97	3400.44
AdaBN-DCNN [10]	13.17	279	20.87	2020	14.97	817	24.57	3690
MS-DCNN [11]	11.44	196.22	19.35	3747	11.67	241.89	22.22	4844
Semi-supervised [23]	12.56	231	22.73	3366	12.1	251	22.66	2840
DCNN [24]	12.61	273.7	22.36	10412	12.64	284.1	23.31	12466
MODBNE [25]	15.04	334.23	25.05	5585.34	12.51	421.91	28.66	6557.62
DBN [26]	15.21	417.59	27.12	9031.64	14.71	442.43	29.88	7954.51
LSTM [25]	16.14	338	24.49	4450	16.18	852	28.17	5550
MLP [26]	16.78	560.59	28.78	14026.72	18.47	479.85	30.96	10444.35