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

Table 4

Results of FD002 with two analyses and different TW.

	Loss function (regression)				Loss function (Weibull)
TW	MAE	RMSE	^2	Score	MAE	RMSE	^2	Score
20	23.85	32.58	0.50	44,982	22.88	33.89	0.81	46,493
30	24.09	33.36	0.59	41,297	21.94	30.90	0.84	29,696
40	22.22	31.75	0.63	36,752	20.86	28.92	0.86	21,125
50	18.94	27.62	0.70	12,390	18.24	25.60	0.88	7,232
60	16.04	22.42	0.81	4,696	17.31	24.60	0.89	5,494
70	13.98	20.34	0.84	3,232	16.11	21.87	0.90	3,497
80	13.43	20.03	0.80	5,478	15.19	20.12	0.90	2,293
90	10.71	15.15	0.87	1,085	13.57	17.58	0.91	1,349
100	10.37	15.32	0.84	1,476	11.70	15.77	0.91	979
110	9.96	15.13	0.84	1,414	13.53	17.97	0.88	1,701
120	9.64	14.27	0.84	790	12.69	16.39	0.88	1,810