A Multiple Attention Convolutional Neural Networks for Diesel Engine Fault Diagnosis

Xiao Yang,Fengrong Bi,Jiangang Cheng,Daijie Tang,Pengfei Shen,Xiaoyang Bi

DOI: https://doi.org/10.3390/s24092708

IF: 3.9

2024-04-25

Sensors

Abstract:Fault diagnosis can improve the safety and reliability of diesel engines. An end-to-end method based on a multi-attention convolutional neural network (MACNN) is proposed for accurate and efficient diesel engine fault diagnosis. By optimizing the arrangement and kernel size of the channel and spatial attention modules, the feature extraction capability is improved, and an improved convolutional block attention module (ICBAM) is obtained. Vibration signal features are acquired using a feature extraction model alternating between the convolutional neural network (CNN) and ICBAM. The feature map is recombined to reconstruct the sequence order information. Next, the self-attention mechanism (SAM) is applied to learn the recombined sequence features directly. A Swish activation function is introduced to solve "Dead ReLU" and improve the accuracy. A dynamic learning rate curve is designed to improve the convergence ability of the model. The diesel engine fault simulation experiment is carried out to simulate three kinds of fault types (abnormal valve clearance, abnormal rail pressure, and insufficient fuel supply), and each kind of fault varies in different degrees. The comparison results show that the accuracy of MACNN on the eight-class fault dataset at different speeds is more than 97%. The testing time of the MACNN is much less than the machine running time (for one work cycle). Therefore, the proposed end-to-end fault diagnosis method has a good application prospect.

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation

What problem does this paper attempt to address?

The paper proposes a method called Multi-Attention Convolutional Neural Network (MACNN) for diesel engine fault diagnosis. Existing methods have limitations in diesel engine fault diagnosis, such as relying on manual feature extraction and insufficient utilization of time series information. MACNN improves feature extraction capability by optimizing the arrangement and kernel size of channel attention and spatial attention modules, and introduces an improved Convolutional Block Attention Module (ICBAM). Vibration signal features are extracted by alternately using CNN and ICBAM, and then the feature maps are recombined to preserve the order information. Next, the recombined sequential features are directly learned using self-attention mechanism (SAM). The paper also introduces the Swish activation function to address the "dead" problem of ReLU and improves accuracy. A dynamic learning rate curve is designed to improve the convergence ability of the model. Experiments show that MACNN achieves an accuracy of over 97% on eight types of fault datasets at different speeds, and the testing time is much shorter than the time for the machine to complete a working cycle. This demonstrates the promising application prospects of this end-to-end fault diagnosis method. The study focuses on diesel engine fault diagnosis, especially diagnosis based on vibration signals, aiming to improve the accuracy and efficiency of diagnosis.

A Multiple Attention Convolutional Neural Networks for Diesel Engine Fault Diagnosis

Fault Diagnosis of Diesel Engine Based on Marginal Spectrums and BP Neural Network

Diesel Engine Fault Diagnosis Method Based on Optimized VMD and Improved CNN

Intelligent Fault Diagnosis of Marine Diesel Engines Based on Efficient Channel Attention-Improved Convolutional Neural Networks

MITDCNN: A multi-modal input Transformer-based deep convolutional neural network for misfire signal detection in high-noise diesel engines

A Multidepth Step-Training Convolutional Neural Network for Power Machinery Fault Diagnosis Under Variable Loads

A digital twin auxiliary approach based on adaptive sparse attention network for diesel engine fault diagnosis

Fault Diagnosis of Marine Diesel Engine Based on Multi-scale Time Domain Decomposition and Convolutional Neural Network

Misfire detection of diesel engine based on convolutional neural networks

A New Fault Diagnosis Method for a Diesel Engine Based on an Optimized Vibration Mel Frequency under Multiple Operation Conditions

Research on fault diagnosis of automobile engines based on the deep learning 1D-CNN method

Knowledge features enhanced intelligent fault detection with progressive adaptive sparse attention learning for high-power diesel engine

Fault detection and diagnostic method of diesel engine by combining rule-based algorithm and BNs/BPNNs

Fault detection and diagnosis of a 12-cylinder trainset diesel engine based on vibration signature analysis and neural network

Fusion of Multi-Layer Attention Mechanisms and CNN-LSTM for Fault Prediction in Marine Diesel Engines

Fault Diagnosis of Diesel Engine Based on Neural Network

A novel intelligent fault diagnosis method based on dual convolutional neural network with multi-level information fusion

Multiattention-Based Feature Aggregation Convolutional Networks with Dual Focal Loss for Fault Diagnosis of Rotating Machinery under Data Imbalance Conditions

TDANet: A Novel Temporal Denoise Convolutional Neural Network With Attention for Fault Diagnosis

Priori-distribution-guided adaptive sparse attention for cross-domain feature mining in diesel engine fault diagnosis

Combination of VMD Mapping MFCC and LSTM: A New Acoustic Fault Diagnosis Method of Diesel Engine