Remaining useful life prediction method based on gated dilation causal convolution

Read full text Summarise with VIRAT
Authors: Jun He, Wei Sun, Changfan Zhang, Yue Yu
Cited by: 0
References: 44

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Rolling bearings are a crucial element of rotating mechanical equipment. Therefore, predicting the remaining useful life (RUL) of this equipment is vital. Monitoring data for rolling bearing operation usually consists of a long-life cycle sequence covering the entire life cycle of the bearing. This process requires deep learning models to model global and local modeling capabilities. However, traditional convolutional neural networks are not very good at modeling global features. We proposed a novel RUL prediction framework based on gating inflation causal convolution to address the above shortcomings. This framework includes a multi-feature squeeze excitation unit that adaptively proofreads feature responses from local and global perspectives. A sinusoidal position coding is also designed to allow the network to obtain information between the distant time steps. The framework also features a gated dilated causal convolution (GDCC) network to reduce the vanishing gradient probability, retain the nonlinear ability of the network, and dynamically realize information changes. Finally, the effectiveness of this method is verified through comparative experimental analysis, ablation study experiments, and visual analysis.

Fields of Study & Subjects

Machine Fault Diagnosis Techniques Reliability and Maintenance Optimization Gear and Bearing Dynamics Analysis

Publication details

Journal / source
Scientific Reports
Volume
16
Issue
1
Publication date
28 March 2026
Language
EN