A novel asymmetrical autoencoder with a sparsifying discrete cosine Stockwell transform layer for gearbox sensor data compression

TL;DR

This paper introduces a novel asymmetrical autoencoder with a sparsifying discrete cosine Stockwell transform layer for efficient gearbox sensor data compression, enhancing reconstruction accuracy with fewer training samples.

Contribution

It proposes a new transform domain layer within an autoencoder that reduces parameters and improves data reconstruction for gearbox sensor data.

Findings

Outperforms existing autoencoder methods on gearbox datasets.

Requires fewer training samples for effective compression.

Achieves up to 32.35% improvement in quality score.

Abstract

The lack of an efficient compression model remains a challenge for the wireless transmission of gearbox data in non-contact gear fault diagnosis problems. In this paper, we present a signal-adaptive asymmetrical autoencoder with a transform domain layer to compress sensor signals. First, a new discrete cosine Stockwell transform (DCST) layer is introduced to replace linear layers in a multi-layer autoencoder. A trainable filter is implemented in the DCST domain by utilizing the multiplication property of the convolution. A trainable hard-thresholding layer is applied to reduce redundant data in the DCST layer to make the feature map sparse. In comparison to the linear layer, the DCST layer reduces the number of trainable parameters and improves the accuracy of data reconstruction. Second, training the autoencoder with a sparsifying DCST layer only requires a small number of datasets. The proposed method is superior to other autoencoder-based methods on the University of Connecticut (UoC) and Southeast University (SEU) gearbox datasets, as the average quality score is improved by 2.00% at the lowest and 32.35% at the highest with a limited number of training samples