Abstract:Blind deconvolution (BD) has been demonstrated to be an efficacious approach for extracting bearing fault-specific features from vibration signals under strong background noise. Despite BD’s appealing feature in adaptability and mathematical interpretability, a significant challenge persists: How to effectively integrate BD with fault-diagnosing classifiers? This issue is intricate to be tackled because the traditional BD method is solely designed for feature extraction with its own optimizer and objective function. When BD is combined with the downstream deep learning classifier, the different learning objectives easily get in conflict. To address this problem, this paper introduces classifier-guided BD (ClassBD) for joint learning of BD-based feature extraction and deep learning-based fault diagnosis. Towards this goal, we first develop a time and frequency neural BD that employs neural networks to implement conventional BD, thereby facilitating seamless integration of BD and the deep learning classifier for co-optimization of model parameters. In the neural BD, we incorporate two filters: (i) a time domain quadratic filter to utilize quadratic convolutional networks for extracting periodic impulses; (ii) a frequency domain linear filter composed of a fully-connected neural network to amplify discrete frequency components. Next, we develop a unified framework built upon a deep learning classifier to guide the learning of BD filters. In addition, we devise a physics-informed loss function composed of kurtosis, l2/l4 norm, and a cross-entropy loss to jointly optimize the BD filters and deep learning classifier. In so doing, the fault labels are fully exploited to direct BD to extract features in distinguishing classes amidst strong noise. To the best of our knowledge, this is the first of its kind that BD is successfully applied to bearing fault diagnosis. Experimental results from three different datasets highlight that ClassBD outperforms other state-of-the-art methods under noisy conditions. The source codes of this paper are available at https://github.com/asdvfghg/ClassBD.

Nonlinear Blind Deconvolution Based on Generalized Normalized l p / l q Norm for Early Fault Detection

Multi-frame Blind Deconvolution Using Sparse Priors

Blind Deconvolution Using an Improved L0 Sparse Representation

Lp (1≤p≤∞) BLIND ADAPTIVE DECONVOLUTION AND ITS APPLICATIONS

Identifiability of Multichannel Blind Deconvolution and Nonconvex Regularization Algorithm

Robust Sparse Threshold Optimization for Impulsive Blind Deconvolution

Convolutional Sparse Learning for Blind Deconvolution and Application on Impulsive Feature Detection.

Non-blind Deconvolution with ℓ 1 -Norm of High-Frequency Fidelity

A geometrical investigation on the generalized lp/lq norm for blind deconvolution.

Blind Deconvolution Technique Based on Improved Correlated Generalized <i>Lp</i>/<i>Lq</i> Norm for Extracting Repetitive Transient Feature

Sparse Blind Deconvolution with Nonconvex Optimization for Ultrasonic NDT Application

Blind Deconvolution Technique Based on Improved Correlated Generalized Lp / Lq Norm for Extracting Repetitive Transient Feature

Multi-task neural network blind deconvolution and its application to bearing fault feature extraction

An Adaptive Multi-D-Norm-Driven Sparse Unfolding Deconvolutional Network for Bearing Fault Diagnosis

Generalized nonlinear hybrid-norm parallel sparse filtering for bearing fault diagnosis under complex interference

Convolutional plug-and-play sparse optimization for impulsive blind deconvolution

Nonnegative bounded convolutional sparsity learning algorithm for envelope blind deconvolution

Deep Non-Blind Deconvolution Via Generalized Low-Rank Approximation.

Generalized Orthogonal Locality Preserving Projections for Nonlinear Fault Detection and Diagnosis

Classifier-guided Neural Blind Deconvolution: A Physics-Informed Denoising Module for Bearing Fault Diagnosis under Noisy Conditions

Training Very Deep CNNs for General Non-Blind Deconvolution.