Discrete Cosine Transformation and Temporal Adjacent Convolutional Neural Network-Based Remaining Useful Life Estimation of Bearings

Table 4

RUL estimation results and comparison.

Testing bearing	Current time (s)	Actual RUL (s)	Er (%)
			LSTM [43]	MCNN [28]	Sparse R [44]	DBN-DP [45]	Proposed method
1–3	18010	5730	93.02	17.43	4.00	2.26	−1.80
1–4	11380	339	191.18	10.67	−10.00	23.31	20.35
1–5	23010	1610	503.11	−148.19	19.00	−83.23	13.66
1–6	23010	1460	542.47	−19.45	81.00	20.34	16.99
1–7	15010	7570	3.43	19.55	60.00	22.93	−2.10
Score			0.25	0.36	0.37	0.46	0.64