Remaining Useful Life Prediction of Rolling Bearings Using Electrostatic Monitoring Based on Two-Stage Information Fusion Stochastic Filtering

Table 4

RUL prediction results based on HSMM.

Sampling points	Degradation state	Identification of degradation state	Mean of the predicted result	Variance of the predicted result	Interval of the predicted result	Actual RUL	Predicted RUL
1	Initial degradation	Normal state	5238	1910	[3328, 7148]	910	Inaccurate
2	Initial degradation	Initial degradation	1227	633	[594, 1860]	805	Accurate
3	Initial degradation	Initial degradation	1227	633	[594, 1860]	785	Accurate
4	Initial degradation	Initial degradation	1227	633	[594, 1860]	753	Accurate
5	Initial degradation	Severe degradation	327	115	[212, 442]	717	Inaccurate
6	Initial degradation	Initial degradation	1227	633	[594, 1860]	648	Accurate
7	Initial degradation	Initial degradation	1227	633	[594, 1860]	599	Accurate
8	Severe degradation	Severe degradation	327	115	[212, 442]	303	Accurate
9	Severe degradation	Severe degradation	327	115	[212, 442]	210	Accurate
10	Severe degradation	Severe degradation	327	115	[212, 442]	116	Inaccurate