Logistic-ELM: A Novel Fault Diagnosis Method for Rolling Bearings

Tan, Zhenhua; Ning, Jingyu; Peng, Kai; Xia, Zhenche; Wu, Danke

doi:10.48550/arxiv.2204.11845

Cited by 1 publication

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, Magar et al [ 7 ] introduced FaultNet, a deep convolutional neural network that combined various signal processing techniques and machine learning techniques for feature extraction and fault classification. Tan et al [ 8 ] introduced fault classification methods employing sequential forward selection to obtain features from vibration datasets, followed by training an integrated model comprising an extreme learning machine and logistic mapping. Other approaches focus on feature decomposition, with an emphasis on effective feature extraction.…”

Section: Introductionmentioning

confidence: 99%

Failure Mode Classification for Rolling Element Bearings Using Time-Domain Transformer-Based Encoder

Vu,

Hiraga,

Miura

et al. 2024

Sensors

View full text Add to dashboard Cite

In this paper, we propose a Transformer-based encoder architecture integrated with an unsupervised denoising method to learn meaningful and sparse representations of vibration signals without the need for data transformation or pre-trained data. Existing Transformer models often require transformed data or extensive computational resources, limiting their practical adoption. We propose a simple yet competitive modification of the Transformer model, integrating a trainable noise reduction method specifically tailored for failure mode classification using vibration data directly in the time domain without converting them into other domains or images. Furthermore, we present the key architectural components and algorithms underlying our model, emphasizing interpretability and trustworthiness. Our model is trained and validated using two benchmark datasets: the IMS dataset (four failure modes) and the CWRU dataset (four and ten failure modes). Notably, our model performs competitively, especially when using an unbalanced test set and a lightweight architecture.

show abstract

Section: Introductionmentioning

confidence: 99%