Abstract:Generally, the health conditions of rotating machinery are complicated and changeable. Meanwhile, its fault labeled information is mostly unknown. Therefore, it is man-sized to automatically capture the useful fault labeled information from the monitoring raw vibration signals. That is to say, the intelligent unsupervised learning approach has a significant influence on fault diagnosis of rotating machinery. In this study, a span-new unsupervised deep learning network can be constructed based on the proposed feature extractor (L12 sparse filtering (L12SF)) and the designed clustering extractor (Weighted Euclidean Affinity Propagation) for resolving the issue that the acquisition of fault sample labeled information is burdensome, yet costly. Naturally, the novel intelligent fault diagnosis method of rotating machinery based on unsupervised deep learning network is first presented in this study. Thereinto, the proposed unsupervised deep learning network consists of two layers of unsupervised feature extractor (L12SF) and one layer of unsupervised clustering (Weighted Euclidean Affinity Propagation). L12SF can improve the regularization performance of sparse filtering, and Weighted Euclidean Affinity Propagation can get rid of the traditional Euclidean distance in affinity propagation that cannot highlight the contribution of different features in fault clustering. To make a long story short, the frequency spectrum signals are first entered into the constructed unsupervised deep learning network for fault feature representation; afterward, the unsupervised feature learning and unsupervised fault classification of rotating machinery can be implemented. The superiority of the proposed algorithms and method is validated by two cases of rolling bearing fault dataset. Ultimately, the proposed unsupervised fault diagnosis method can provide a theoretical basis for the development of intelligent unsupervised fault diagnosis technology for rotating machinery.

Self-supervised signal representation learning for machinery fault diagnosis under limited annotation data

A Few-Shot Machinery Fault Diagnosis Framework Based on Self-Supervised Signal Representation Learning

Fault Diagnosis of Rotating Machinery Based on Improved Self-Supervised Learning Method and Very Few Labeled Samples

Self-supervised Learning Advance Fault Diagnosis of Rotating Machinery

Self-Supervised Simple Siamese Framework for Fault Diagnosis of Rotating Machinery with Unlabeled Samples.

Self-Supervised Learning with Signal Masking and Reconstruction for Machinery Fault Diagnosis under Limited Labeled Data and Varying Working Condition

Unsupervised Mechanical Fault Feature Learning Based on Consistency Inference-Constrained Sparse Filtering

A novel self-training semi-supervised deep learning approach for machinery fault diagnosis

Self-supervised Learning for Intelligent Fault Diagnosis of Rotating Machinery with Limited Labeled Data

A Universal Feature Extractor Based on Self-Supervised Pre-Training for Fault Diagnosis of Rotating Machinery under Limited Data

A Novel Cross-Sensor Self-Supervised Learning Method for Rotating Machinery Fault Diagnosis

Design and application of unsupervised convolutional neural networks integrated with deep belief networks for mechanical fault diagnosis

An Intelligent Fault Diagnosis Based on Adversarial Generating Module and Semi-supervised Convolutional Neural Network

Temporal Self-supervised Domain Adaptation Network for Machinery Fault Diagnosis under Multiple Non-ideal Conditions

Few-shot Learning for Rolling Bearing Fault Diagnosis Via Siamese Two-dimensional Convolutional Neural Network

Unsupervised Learning for Machinery Adaptive Fault Detection Using Wide-Deep Convolutional Autoencoder with Kernelized Attention Mechanism

A Novel Unsupervised Deep Learning Network for Intelligent Fault Diagnosis of Rotating Machinery

Semi-supervised meta-learning networks with squeeze-and-excitation attention for few-shot fault diagnosis

A self-supervised contrastive learning framework with the nearest neighbors matching for the fault diagnosis of marine machinery

An Intelligent Fault Diagnosis Method Using Unsupervised Feature Learning Towards Mechanical Big Data

Self-Supervised Learning Denoising Network For Intelligent Fault Diagnosis With Limited Labeled Data