Abstract:Vibration-based condition monitoring techniques are commonly used to identify faults in rolling element bearings. Accuracy and speed of fault detection procedures are critical performance measures in condition monitoring. Delay is especially important in remote condition monitoring and time-sensitive industrial applications. While most existing methods focus on accuracy, little attention has been given to the inference time in the fault identification process. In this paper, we address this gap by presenting a Convolutional Neural Network (CNN) based approach for real-time fault identification in rolling element bearings. We encode raw vibration signals into two-dimensional images using various encoding methods and use these with a CNN to classify several categories of bearing fault types and sizes. We analyse the interplay between fault identification accuracy and processing time. For training and evaluation we use a bearing failure CWRU dataset.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: in the condition monitoring of rolling bearings, although the existing fault identification methods focus on accuracy, they pay less attention to the inference time (i.e., processing time). This is especially important for remote condition monitoring and time - sensitive industrial applications, because the delay will significantly affect the timeliness of maintenance decisions. Specifically, the goal of the paper is to develop a deep - learning - based method, especially the Convolutional Neural Network (CNN), for real - time fault identification. By encoding the original vibration signals into two - dimensional images and using these images for classification, fast and accurate identification of fault types and sizes can be achieved. In addition, the paper also explores the trade - off between fault identification accuracy and processing time. ### Main problem summary: 1. **Limitations of existing methods**: Most of the existing methods mainly focus on the accuracy of fault identification and ignore the inference time, which has an adverse impact on time - sensitive application scenarios such as remote monitoring and industrial automation. 2. **Real - time and accuracy**: How to reduce the processing time of fault identification while ensuring high accuracy, so as to achieve real - time fault detection. 3. **Choice of data encoding techniques**: Select an appropriate encoding method to convert the time - series vibration signals into two - dimensional images, so that the CNN can effectively extract features and perform classification. ### Solutions: - Use the Convolutional Neural Network (CNN) for fault identification. - Encode the original vibration signals into two - dimensional images, using multiple encoding methods (such as pixel intensity, Gramian Angular Field (GAF), Markov Transition Field (MTF), Recurrence Plot, etc.). - Evaluate the performance of different encoding methods in terms of fault identification accuracy and processing time, and find the optimal balance point. Through this method, the paper aims to provide a fault identification scheme that can ensure both high accuracy and fast response, which is suitable for industrial application scenarios that require immediate decision - making.

Deep learning-based fault identification in condition monitoring

Real-Time Vibration-Based Bearing Fault Diagnosis Under Time-Varying Speed Conditions

Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings

A Vision Transformer-Based Approach to Bearing Fault Classification via Vibration Signals

Condition Monitoring of Ball Bearings Based on Machine Learning with Synthetically Generated Data

Investigation of Rolling Bearing Weak Fault Diagnosis Based on CNN with Two-Dimensional Image

A Deep Learning Method for Bearing Fault Diagnosis Based on Time-Frequency Image

Bearing Fault Diagnosis under Transient Conditions: Using Variational Mode Decomposition and the Symmetrized Dot Pattern‐Based Convolutional Neural Network Model

FaultNet: A Deep Convolutional Neural Network for Bearing Fault Classification

Rolling Bearing Fault Diagnosis Based on Convolutional Neural Network and Multi-sensor Information Fusion

Review of the U.S. military's human immunodeficiency virus program: a legacy of progress and a future of promise.

Intelligent Defect Diagnosis of Rolling Element Bearings under Variable Operating Conditions Using Convolutional Neural Network and Order Maps

An Intelligent Fault Diagnosis Method of Rolling Bearings Based on Short-Time Fourier Transform and Convolutional Neural Network

Bearing Fault Diagnosis Based on Convolutional Neural Network Learning of Time-Domain Vibration Signal Imaging

An intelligent of fault diagnosis and predicting remaining useful life of rolling bearings based on convolutional neural network with bidirectional LSTM

Multitask Convolutional Neural Network With Information Fusion for Bearing Fault Diagnosis and Localization

Rolling Bearing Fault Diagnosis Based on Convolutional Neural Network and Support Vector Machine

2D CNN-Based Multi-Output Diagnosis for Compound Bearing Faults under Variable Rotational Speeds

Application of Multiscale Learning Neural Network Based on CNN in Bearing Fault Diagnosis

Fault diagnosis of rolling bearing of wind turbines based on the Variational Mode Decomposition and Deep Convolutional Neural Networks

Bearing fault diagnosis base on multi-scale CNN and LSTM model