Remaining useful life prediction of rolling bearings based on convolutional recurrent attention network

Zhang, Qiang; Ye, Zijian; Shao, Siyu; Niu, Tianlin; Zhao, Yuwei

doi:10.1108/aa-08-2021-0113

Cited by 12 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Attention mechanisms can ignore certain irrelevant regional information and focus on the key areas in the image through learning. Different from other methods, the proposed DMRFAB module includes a dense multi-receptive field module both introducing SAM 21 and CAM 22 , which helps multi- receptive field blocks better extract deep feature information, improve feature representation capabilities, and ultimately improve module deblurring performance. The DMRFAB module, illustrated in Fig.…”

Section: The Proposed Methodsmentioning

confidence: 99%

An image deblurring method using improved U-Net model based on multilayer fusion and attention mechanism

Lian,

Wang

2023

Sci Rep

View full text Add to dashboard Cite

The investigation of image deblurring techniques in dynamic scenes represents a prominent area of research. Recently, deep learning technology has gained extensive traction within the field of image deblurring methodologies. However, such methods often suffer from limited inherent interconnections across various hierarchical levels, resulting in inadequate receptive fields and suboptimal deblurring outcomes. In U-Net, a more adaptable approach is employed, integrating diverse levels of features effectively. Such design not only significantly reduces the number of parameters but also maintains an acceptable accuracy range. Based on such advantages, an improved U-Net model for enhancing the image deblurring effect was proposed in the present study. Firstly, the model structure was designed, incorporating two key components: the MLFF (multilayer feature fusion) module and the DMRFAB (dense multi-receptive field attention block). The aim of these modules is to improve the feature extraction ability. The MLFF module facilitates the integration of feature information across various layers, while the DMRFAB module, enriched with an attention mechanism, extracts crucial and intricate image details, thereby enhancing the overall information extraction process. Finally, in combination with fast Fourier transform, the FRLF (Frequency Reconstruction Loss Function) was proposed. The FRLF obtains the frequency value of the image by reducing the frequency difference. The present experiment results reveal that the proposed method exhibited higher-quality visual effects. Specifically, for the GoPro dataset, the PSNR (peak signal-to-noise ratio) reached 31.53, while the SSIM (structural similarity index) attained a value of 0.948. Additionally, for the Real Blur dataset, the PSNR achieved 31.32, accompanied by an SSIM score of 0.934.

show abstract

Section: The Proposed Methodsmentioning

confidence: 99%

An image deblurring method using improved U-Net model based on multilayer fusion and attention mechanism

Lian,

Wang

2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Therefore, to establish a predictive model with accuracy and robustness, it is necessary to develop a nonlinear data fusion model that can capture the degradation of complex machinery. Zhang Qiang [68] proposed a new RUL prediction model CRAN, a model based on CNN-LSTM, in order to achieve the high-precision prediction of rolling bearing end-to-end RUL, which effectively combines the powerful feature extraction abilities of CNN and LSTM, has a time-series-processing capability and has higher RUL prediction accuracy than CNN-and LSTM-based models. Nistane Vinod [69] proposed a fault prediction method that integrates an optimization health indicator (OHIs) and bearing RUL by using a genetic algorithm.…”

Section: The Mainshaft Bearing Life Analysis Based On Digital Twinmentioning

confidence: 99%

Application of Life Cycle of Aeroengine Mainshaft Bearing Based on Digital Twin

Yan³

et al. 2023

Processes

View full text Add to dashboard Cite

Aeroengine mainshaft bearings are key components in modern aeroengines, and their main functions are to support the rotation of the main shaft of the aeroengine in harsh environments, such as high temperature, heavy load, high speed and oil break; reduce the friction coefficient during the high-speed rotation of the main shaft; and reliably ensure the rotation accuracy and power transmission of the aeroengine’s main shaft during operation. The manufacture of aeroengine mainshaft bearings requires complex processes and precise machining to ensure high performance and reliability, and how to intelligently complete the production and manufacture of mainshaft bearings and ensure the strength and accuracy of the bearings, quickly distinguish the fault types of the bearings and efficiently calculate, analyze and predict the life of the bearings are the current research hotspots. Therefore, building a high-fidelity and computationally efficient digital twin life cycle of aeroengine mainshaft bearings is a valuable solution. This paper summarizes the key manufacturing technology, manufacturing mode and manufacturing process based on digital twins in the life cycle of aeroengine mainshaft bearings, including the metallurgical process, heat treatment process and grinding process of aeroengine mainshaft bearings. It presents a fault diagnosis and life analysis of mainshaft bearings of aeroengines, discussing the key technologies and research directions of the life cycle of mainshaft bearings based on digital twins.

show abstract

“…Yin used Bi-RNN method to predict the residual life of main bearing of wind turbine, and achieved high accuracy [8]. Han proposed a residual life prediction method for rolling bearings based on BiLSTM [9]. The above research is to predict the remaining life of the bearing by establishing a deep learning model, and has achieved good prediction results.…”

Section: Introductionmentioning

confidence: 99%

Residual Life Prediction of Rolling Bearing based on AOA-LSTM

Wang¹,

Xi²

2023

SJT

View full text Add to dashboard Cite

Aiming at the problem of predicting the remaining life of rolling bearings, a method for predicting the remaining life of bearings based on the arithmetic optimization algorithm (AOA) and the long-short-term memory network (LSTM) fusion algorithm is proposed. First, use the random forest algorithm to analyze the importance of the extracted time-domain and frequency-domain feature indicators, and build a degradation feature decision table; then, use the AOA optimization algorithm to optimize the hyperparameters in the LSTM, and select the optimal hyperparameters to establish predictions model; finally, the degradation features are input into the prediction model for prediction, and the prediction model is evaluated by root mean square error (RMSE) and mean absolute error (MAE). The proposed research method is experimentally verified by the XJTU-SY dataset. The experimental results show that the RMSE and MAE of the proposed research method are 5.56% and 4.37%, respectively. Compared with the MLP model, the RMSE and MAE are reduced by 31.58% and 29.61%, respectively. Compared with the RNN model, the RMSE and MAE are reduced by 24.66% and 25.49%, respectively, which verifies the effectiveness of the research method.

show abstract

Remaining useful life prediction of rolling bearings based on convolutional recurrent attention network

Cited by 12 publications

References 34 publications

An image deblurring method using improved U-Net model based on multilayer fusion and attention mechanism

An image deblurring method using improved U-Net model based on multilayer fusion and attention mechanism

Application of Life Cycle of Aeroengine Mainshaft Bearing Based on Digital Twin

Residual Life Prediction of Rolling Bearing based on AOA-LSTM

Contact Info

Product

Resources

About