Sparse auto-encoder with regularization method for health indicator construction and remaining useful life prediction of rolling bearing

She, Daoming; Jia, Minping; Pecht, Michael

doi:10.1088/1361-6501/ab8c0f

Cited by 44 publications

(23 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There also exist studies using a multi-domain input. These studies [ 7 , 14 , 29 ] obtain features using statistics as a time domain feature, FT as a frequency domain feature, and WT as a time–frequency domain feature. One study [ 20 ] obtained features in three domains, reducing the dimension through restricted Boltzmann machine (RBM), and predicting HI through gated recurrent units (GRU).…”

Section: The Literaturementioning

confidence: 99%

See 1 more Smart Citation

A One-Stage Ensemble Framework Based on Convolutional Autoencoder for Remaining Useful Life Estimation

Park

Kim

2022

Sensors

View full text Add to dashboard Cite

As the legislative pressure to reduce energy consumption is increasing, data analysis of power consumption is critical in the production planning of manufacturing facilities. In legacy studies, a machine conducting a single continuous operation has been mainly observed for power estimation. However, the production machine of a modularized line, which conducts complex discrete operations, is more like the actual factory system than an identical simple machine. During the information collection of this kind of production line, it is important to interpret mixed signals from multiple machines to ensure that there is no reduction in the information quality due to noise and signal fusion and discrete events. A data pipeline—from data collection (from different sources) to preprocessing, data conversion, synchronization, and deep learning classification—to estimate the total power use of the future process plan, is proposed herein. The pipeline also establishes an auto-labeled data set of individual operations that contributes to building an power estimation model without manual data preprocessing. The proposed system is applied to a modular factory, connected with machine controllers, using standardized protocols individually and linked to a centralized power monitoring system. Specifically, a robot arm cell was investigated to evaluate the pipeline, with the result of the power profile being synchronized with the robot program.

show abstract

Section: The Literaturementioning

confidence: 99%

“…Some studies attempted to conduct both feature selection and extraction simultaneously. Hu et al and She et al performed feature extraction through RBM and then selected features based on the HI metric [ 29 , 31 ].…”

Section: The Literaturementioning

confidence: 99%

A One-Stage Ensemble Framework Based on Convolutional Autoencoder for Remaining Useful Life Estimation

Park

Kim

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The existing studies usually extracted the different fault characteristics from the original sensory signal. She et al [29] proposed a health indicator construction method based on a Sparse Auto-encoder with a Regularization (SAEwR) Model for rolling bearings. Zhang et al used the summation of the mean maximum radius of the different datasets divided by the k-means clustering algorithm as the health indicator, and then used the local outlier coefficient algorithm to eliminate the outliers' influence [30].…”

Section: Deep-learning-based Approaches For Rul Predictionmentioning

confidence: 99%

Hard Negative Samples Contrastive Learning for Remaining Useful-Life Prediction of Bearings

Qian

Chen³

et al. 2022

Lubricants

View full text Add to dashboard Cite

In recent years, deep learning has become prevalent in Remaining Useful-Life (RUL) prediction of bearings. The current deep-learning-based RUL methods tend to extract high dimensional features from the original vibration data to construct the Health Indicators (HIs), and then use the HIs to predict the remaining life of the bearings. These approaches ignore the sequential relationship of the original vibration data and seriously affect the prediction accuracy. In order to tackle this problem, we propose a hard negative sample contrastive learning prediction model (HNCPM) with encoder module, GRU regression module and decoder module, used for feature embedding, regression RUL prediction and vibration data reconstruction, respectively. We introduce self-supervised contrast learning by constructing positive and negative samples of vibration data rather than constructing any health indicators. Furthermore, to avoid the subtle variability of vibration data in the health stage to aggravate the degradation features learning of the model, we propose the hard negative samples by cosine similarity, which are most similar to the positive sample. Meanwhile, a novel infoNCE and MSE-based loss function is derived and applied to the HNCPM to simultaneously optimize a lower bound on mutual information of the positive and negative sample over life cycle, as well as the discrepancy between true and predicted values of the vibration data, such that the model can learn the fine-grained degradation representations by predicting the future without any HIs as labels. The HNCPM is validated on the IEEE PHM Challenge 2012 dataset. The results demonstrate that the prediction performance of our model is superior to the state-of-the-art methods.

show abstract

“…There are studies using a multi domain input. [25], [11], [26] make features using statistics as a time domain feature, FT as a frequency domain feature, and WT as a time frequency domain feature. [27] also make feature in 3 domains, reduces the dimension through RBM (Restricted Boltzmann machine), and predicts HI through GRU (Gated Recurrent Units).…”

Section: Domain Selectionmentioning

confidence: 99%

A One-Stage Ensemble Framework based on Convolutional Autoencoder for Remaining Useful Life Estimation

Park

Kim

2021

Preprint

View full text Add to dashboard Cite

The research on predictive maintenance of rotating machines, the most important element in manufacturing facilities, has been very active. The widespread availability of smart factory solutions has led to improved data collection from machines and processes and is able to provide key information. For our purpose, the collected information enables the maintenance system to predict the remaining useful life using deep learning models. The introduction of multi-layer perceptron of signal processing originating from bearings, in time series data, has been discussed in many publications. However, estimating accuracy for the remaining useful life is determined by the selection of the feature domain and the concatenation network model. Herein, we introduce a convolutional Autoencoder based on multi-domain ensemble learning in order to include various feature domains and a concatenation network operated by latent space into a single neural network. The performance of the proposed model is evaluated by using a simple health indicator and a PRONOSTIA dataset and compared with a simple concatenation model, 2-stage Autoencoder, and a recurrent neural network.

show abstract

Sparse auto-encoder with regularization method for health indicator construction and remaining useful life prediction of rolling bearing

Cited by 44 publications

References 46 publications

A One-Stage Ensemble Framework Based on Convolutional Autoencoder for Remaining Useful Life Estimation

A One-Stage Ensemble Framework Based on Convolutional Autoencoder for Remaining Useful Life Estimation

Hard Negative Samples Contrastive Learning for Remaining Useful-Life Prediction of Bearings

A One-Stage Ensemble Framework based on Convolutional Autoencoder for Remaining Useful Life Estimation

Contact Info

Product

Resources

About