Developing Deep Survival Model for Remaining Useful Life Estimation Based on Convolutional and Long Short-Term Memory Neural Networks

Table 6

Results of FD004 with two analyses and different TW.

	Loss function (regression)				Loss function (Weibull)
TW	MAE	RMSE	^2	Score	MAE	RMSE	^2	Score
20	27.81	36.39	0.54	58,522	28.69	38.56	0.78	345,533
30	27.44	35.89	0.52	65,319	27.36	36.75	0.76	580,301
40	22.77	30.32	0.64	17,827	23.16	31.59	0.84	41,486
50	21.16	29.17	0.68	13,305	20.87	28.67	0.84	10,035
60	19.41	26.37	0.73	11,064	21.59	29.99	0.84	51,971
70	19.83	27.56	0.73	6,999	18.11	26.05	0.88	11,148
80	18.59	24.55	0.78	4,652	18.85	27.68	0.86	120,754
90	18.41	24.93	0.77	4,832	16.90	24.36	0.89	75,231
100	17.78	23.69	0.77	3,448	15.71	23.05	0.89	36,726
110	16.11	21.97	0.81	2,726	16.95	24.98	0.87	96,205
120	16.29	21.92	0.82	2,260	16.75	25.09	0.88	122,440