Research Article
An Improved Recursive ARIMA Method with Recurrent Process for Remaining Useful Life Estimation of Bearings
| Input: | (1) | | (2) | repeat | (3) | | (4) | | (5) | Calculate ACF of y | (6) | | (7) | until eps and ACF converge \(⊳\) the limit eps depends on practical situation. | (8) | First guess: | (9) | for training and testing set do | (10) | Calculate by \(⊳\) other prediction methods also work. | (11) | : end for | (12) | : k = 1 | (13) | repeat | (14) | for training and testing set do | (15) | Calculate RMA | (16) | | (17) | Construct feature | (18) | | (19) | end for | (20) | Train the Rotation Forest in training set | (21) | for training set do | (22) | | (23) | end for | (24) | Calculate | (25) | for training and testing set do | (26) | | (27) | end for | (28) | Calculate changes | (29) | | (30) | until or | (31) | Construct F<k> | (32) | forj from 1 to k do | (33) | Calculate | (34) | end for | (35) | | | Output: estimated output |
|