Abstract:With outstanding deep feature learning and nonlinear classification abilities, Convolutional Neural Networks (CNN) have been gradually applied to deal with various fault diagnosis tasks. Affected by variable working conditions and strong noises, the empirical datum always has different probability distributions, and then different data segments may have inconsistent contributions, so more attention should be assigned to the informative data segments. However, most of the CNN-based fault diagnosis methods still retain black-box characteristics, especially the lack of attention mechanisms and ignoring the special contributions of informative data segments. To address these problems, we propose a new intelligent fault diagnosis method comprised of an improved CNN model named Efficient Convolutional Neural Network (ECNN). The extensive view can cover the special characteristic periods, and the small view can locate the essential feature using Pyramidal Dilated Convolution (PDC). Consequently, the receptive field of the model can be greatly enlarged to capture the location information and excavate the remarkable informative data segments. Then, a novel residual network feature calibration and fusion (ResNet-FCF) block was designed, which uses local channel interactions and residual networks based on global channel interactions for weight-redistribution. Therefore, the corresponding channel weight is increased, which puts more attention on the information data segment. The ECNN model has achieved encouraging results in information extraction and feature channel allocation of the feature. Three experiments are used to test different diagnosis methods. The ECNN model achieves the highest average accuracy of fault diagnosis. The comparison results show that ECNN has strong domain adaptation ability, high stability, and superior diagnostic performance.

Analog Circuit Fault Diagnosis with Echo State Networks Based on Corresponding Clusters

An Constraint-based Clustering Integrated BP Networks Approach for Analog Circuit Fault Diagnose

Fault diagnosis model based on Granular Computing and Echo State Network

A novel compound neural network for fault diagnosis

Algorithms Comparation Of Feature Extraction And Multi-Class Classification For Fault Diagnosis Of Analog Circuit

A Novel Approach of Feature Extraction for Analog Circuit Fault Diagnosis Based on WPD-LLE-CSA

Analog Circuit Fault Diagnosis via Joint Cross-Wavelet Singular Entropy and Parametric t-SNE

A fault diagnosis method for analog circuits based on EEMD-PSO-SVM

Attention Based Echo State Network

Research on Electronic Circuit Fault Diagnosis Method Based on SWT and DCNN-ELM

A novel analog circuit fault diagnosis method based on multi-channel 1D-resnet and wavelet packet transform

Analog fault diagnosis of actual circuits using neural networks

Application of a Pattern-Recognition Neural Network for Detecting Analog Electronic Circuit Faults

Strategy for Identifying Analog Circuit Faults Using Improved Neural Network Algorithms

Fault diagnosis of analog circuits based on multifrequency test and neural networks

Application of multi-frequency test and neural network to fault diagnosis in analog circuits

Fault diagnosis method of analog circuits based on temporal CNN

Method for Fault Diagnosis of Track Circuits Based on a Time–Frequency Intelligent Network

A Fuzzy Classification Approach for Analog Fault Diagnosis Applying Immune Algorithm.

ECNN: Intelligent Fault Diagnosis Method Using Efficient Convolutional Neural Network

An echo state network with interacting reservoirs for modeling and analysis of nonlinear systems