Health state assessment of bearing with feature enhancement and prediction error compensation strategy

Guo

et al. 2023

In this paper, a hybrid convolutional neural network (CNN)-Bidirectional gated recurrent unit (BiGRU) model is integrated with the bootstrap method to endow the deep learning (DL) based prognostic method with the quantification capability of the prognostic intervals. The proposed hybrid method contains three parts: (I) The complete ensemble empirical mode decomposition with adaptive noise and principal component analysis (CEEMDAN-PCA) and the CNN-BiGRU are utilized to automatically construct the health indicator (HI). (II) 3σ criterion is employed to detect the first predicting time based on the HIs of rolling bearings. (III) The bootstrap method is imposed to endow the proposed DL method with the quantification capability of the prognostic intervals. The experimental validation is carried out on the XJTU-SY bearing dataset and the proposed method outperforms the other four methods in the majority of cases. In addition, the proposed method not only comprehensively considers the fault prognosis error caused by model parameters and noise, but also considers the prediction error caused by different combinations of features on the model.

Section: Hi Construction Based On Ceemdan-pcamentioning

confidence: 86%

Section: Introductionmentioning

confidence: 99%

A deep learning based health indicator construction and fault prognosis with uncertainty quantification for rolling bearings

Guo

et al. 2023

“…To highlight the effectiveness of the proposed method, the methods of Ginart et al [35], Li et al [13], Zhang et al [36], and Wang et al [12] are also used for comparison. The comparison results are summarized in table 3.…”

Section: Rul Prediction Using Sliding Sir Filtering Methodsmentioning

confidence: 99%

A sliding sequence importance resample filtering method for rolling bearings remaining useful life prediction based on two Wiener-process models

Song,

Xu,

2023

The remaining useful life (RUL) prediction of rolling bearings is an important part of prognostic and health management of mechanical systems. The model based on Wiener process can describe the time variability in the degradation process of bearings. However, in practical engineering, the degradation trends of bearings are often inconsistent, and it is difficult to fit the actual degradation trends of bearings with a single Wiener process model-based filtering method. Therefore, to improve the generalization ability, this paper uses linear model and exponential model based on Wiener process to predict bearing RUL. A sliding sequence importance resample filtering algorithm is proposed to track the degradation state of the bearing and reduce the prediction error by combining the two degradation models. Last, the superiority and effectiveness of the proposed method are illustrated by comparing with other commonly used RUL prediction methods on the basis of PRONOSTIA dataset.

“…Time-domain features of bearings vibration signals are simple and easy to compute, and have clear physical significance. Actually, the RMS [34] can effectively demonstrate the escalation of vibration energy during the process of degradation. Therefore, the RMS feature is considered as a reliable health indicator for detecting the abnormal point time and predicting the RUL.…”

Section: Health Indicator Construction With Sdamentioning

confidence: 99%

Remaining useful life prediction of bearings using a trend memory attention-based GRU network

Li,

Fan

et al. 2024

Remaining useful life prediction of bearings holds significant importance in enhancing the reliability and durability of rotating machinery. Bearings undergo a gradual degradation process that unfolds over multiple stages. In this paper, a novel framework for forecasting the remaining useful life of bearings is put forward, which includes the construction of a health indicator with a stage division algorithm and the estimation of the health indicator using a new trend memory attention-based gated recurrent unit (TMAGRU). The stage division algorithm, based on the K-Means++ algorithm and angle recognition algorithm, is introduced to distinguish the degradation stage based on the health indicator. Inspired by the double exponential smoothing technique and attention mechanism, the proposed TMAGRU network effectively incorporates both the historical health information in the slow degradation stage and its trend. Experimental results conducted on IEEE PHM Challenge 2012 dataset and XJTU-SY dataset demonstrate the superior predictive performance of the proposed approach compared to several state-of-the-art predictive networks.