← Back to all news
arXiv AI

Spectral Aliasing Pretext: A novel task for Self-Supervised fault diagnosis in rotating machinery

arXiv:2608. 05705v1 Announce Type: cross Abstract: Deep learning is a new way for machinery fault diagnosis but requires extensive labeled data, a scarce resource in industrial settings.

Related stories

arXiv Machine Learning
Sep 24

When Labels Are Scarce: An Oscillatory State Space Model for Vibration Diagnosis

The paper introduces DualRes, a compact oscillatory state‑space model designed for vibration‑based fault diagnosis when labeled data are scarce and computational resources are limited. DualRes integrates two spectral views of vibration and employs selective oscillatory memory to learn how long to retain temporal patterns, resulting in a lightweight encoder with only 39,528 parameters. Evaluations on six bearing datasets and a gearbox benchmark show that DualRes outperforms nine competing methods across most label budgets, achieving significant gains in macro‑F1, faster inference, and reduced storage requirements.

By Mainak Mallick, Seung-Kyum Choi
benchmarks
More like this →
Hugging Face Trending Papers
Jun 25

A Multi-Fidelity Convolutional Autoencoder-Transfer Learning Framework for Guided-Wave-Based Damage Diagnosis Using Large Simulated and Limited Experimental Datasets

Guided wave-based structural health monitoring (GWSHM) with onboard transducers offers significant potential for the early diagnosis of damage in engineering structures. However, the practical deployment of deep learning models is often hindered by the limited availability of labelled experimental data and the high computational cost of generating large-scale high-fidelity simulation datasets.

fine-tuning
More like this →
arXiv AI
Sep 18

FreqCondNorm: Towards Cross-domain Predictive Maintenance through a Frequency-Conditioned Transformer Foundation Model

The paper introduces FreqCondNorm, a Transformer-based architecture that adds a frequency-conditioned normalization layer to unify heterogeneous time-series data for predictive maintenance. The model is pretrained on five public datasets using masked auto‑encoding and contrastive learning, achieving 99.2% accuracy on CWRU and 82.1% zero‑shot accuracy on MFPT, showing strong transfer across sampling frequencies. However, it does not improve remaining useful life prediction, indicating a mismatch between pretraining and RUL objectives that requires further study.

By Zaynab Raounak, Camille LHermine, Zhiguo Zeng
llms
More like this →
arXiv Machine Learning
Sep 10

Deep Learning Approach to Bearing and Induction Motor Fault Diagnosis via Data Fusion

The paper presents a deep learning framework that uses Convolutional Neural Networks to analyze accelerometer and microphone data for diagnosing bearing and induction motor faults. It further employs a Long Short-Term Memory network to fuse these sensor streams, demonstrating the advantages of data fusion. The authors advocate for multi‑model diagnosis and encourage the collection of diverse multi‑sensor datasets, such as acoustic and accelerometer recordings, for constant‑speed data collection.

By Mert Sehri, Merve Ertargin, Ozal Yildirim, Ahmet Orhan, Patrick Dumond
More like this →