Evolutionary manifold regularized stacked denoising autoencoders for gearbox fault diagnosis

作者：

Highlights：

• A new deep learning model called MRSDAE is proposed for feature learning.

• MRSDAE is very effective to learn features from vibration signals.

• PSO is used to evolve structure and parameters of MRSDAE simultaneously.

• The experimental results illustrate that MRSDAE outperforms other DNNs.

摘要

•A new deep learning model called MRSDAE is proposed for feature learning.•MRSDAE is very effective to learn features from vibration signals.•PSO is used to evolve structure and parameters of MRSDAE simultaneously.•The experimental results illustrate that MRSDAE outperforms other DNNs.

论文关键词：Gearbox fault diagnosis,Deep learning,Stacked denoising autoencoders,Manifold regularization,Particle swarm optimization

论文评审过程：Received 10 November 2018, Revised 23 April 2019, Accepted 25 April 2019, Available online 7 May 2019, Version of Record 4 June 2019.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.04.022