Differentiable neural architecture search augmented with pruning and multi-objective optimization for time-efficient intelligent fault diagnosis of machinery

Abstract Intelligent fault diagnosis, which is mainly based on neural network, has been widely used in machinery monitoring. Although such deep learning methods are effective, the new architectures are mainly handcrafted by series of experiments that require ample time and substantial efforts. To automate process of building neural networks and save designing time, a novel differentiable neural architecture search method is proposed. By gradually reducing candidate operations while retaining trained parameters during pruning, computation consumed by each stage of neural architecture search is decreased, which accelerates search process. To improve inferential efficiency of subnetworks, specially designed penalty terms are introduced into the objective function for searching optimal numbers of layers and nodes, which can reduce complexity of subnetworks and save calculation time of signal analysis. In addition, exclusive competition between candidate operations is broken by changing discretization and selection methods of operations, which provides a basis for channel fusion. Effectiveness of the proposed method is verified by two datasets. Experiments show that this method can generate subnetworks of lower complexity and less computational cost than other state-of-art neural architecture search techniques, while achieving competitive result.

[1]  Peng Liu,et al.  Deep Evolutionary Networks with Expedited Genetic Algorithms for Medical Image Denoising , 2019, Medical Image Anal..

[2]  Jong-Myon Kim,et al.  Automated bearing fault diagnosis scheme using 2D representation of wavelet packet transform and deep convolutional neural network , 2019, Comput. Ind..

[3]  Biao Wang,et al.  LiftingNet: A Novel Deep Learning Network With Layerwise Feature Learning From Noisy Mechanical Data for Fault Classification , 2018, IEEE Transactions on Industrial Electronics.

[4]  Xiangdong Wang,et al.  Multiscale local features learning based on BP neural network for rolling bearing intelligent fault diagnosis , 2020, Measurement.

[5]  Yi Lu,et al.  MixPath: A Unified Approach for One-shot Neural Architecture Search , 2020, ArXiv.

[6]  Yongsheng Zhu,et al.  A novel model with the ability of few-shot learning and quick updating for intelligent fault diagnosis , 2020 .

[7]  Hanzhang Wang,et al.  Evolutionary recurrent neural network for image captioning , 2020, Neurocomputing.

[8]  Xi Li,et al.  A lightweight neural network with strong robustness for bearing fault diagnosis , 2020 .

[9]  Wentao Mao,et al.  Online detection of bearing incipient fault with semi-supervised architecture and deep feature representation , 2020 .

[10]  Giuseppe De Pietro,et al.  Deep neural network for hierarchical extreme multi-label text classification , 2019, Appl. Soft Comput..

[11]  Michael Carbin,et al.  The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks , 2018, ICLR.

[12]  Xiang Li,et al.  Deep residual learning-based fault diagnosis method for rotating machinery. , 2019, ISA transactions.

[13]  Gurpreet Singh,et al.  A novel method to classify bearing faults by integrating standard deviation to refined composite multi-scale fuzzy entropy , 2020, Measurement.

[14]  Zitong Zhou,et al.  A Compact Convolutional Neural Network Augmented with Multiscale Feature Extraction of Acquired Monitoring Data for Mechanical Intelligent Fault Diagnosis , 2020 .

[15]  Xiaodong Wang,et al.  Incipient fault feature extraction of rolling bearings based on the MVMD and Teager energy operator. , 2018, ISA transactions.

[16]  Konstantinos Gryllias,et al.  A deep learning method for bearing fault diagnosis based on Cyclic Spectral Coherence and Convolutional Neural Networks , 2020 .

[17]  Xiao-Sheng Si,et al.  A rotating machinery fault diagnosis method based on multi-scale dimensionless indicators and random forests , 2020 .

[18]  Ameet Talwalkar,et al.  Random Search and Reproducibility for Neural Architecture Search , 2019, UAI.

[19]  Qi Tian,et al.  Progressive Differentiable Architecture Search: Bridging the Depth Gap Between Search and Evaluation , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[20]  Yu Zhou,et al.  Application of neural network algorithm in fault diagnosis of mechanical intelligence , 2020, Mechanical Systems and Signal Processing.

[21]  Vicente Climente-Alarcon,et al.  Time-frequency vibration analysis for the detection of motor damages caused by bearing currents , 2017 .

[22]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[23]  Hee-Jun Kang,et al.  A survey on Deep Learning based bearing fault diagnosis , 2019, Neurocomputing.

[24]  Jun Pan,et al.  A Deep Learning Network via Shunt-Wound Restricted Boltzmann Machines Using Raw Data for Fault Detection , 2020, IEEE Transactions on Instrumentation and Measurement.

[25]  Ruixin Wang,et al.  A Deep Transfer Nonnegativity-Constraint Sparse Autoencoder for Rolling Bearing Fault Diagnosis With Few Labeled Data , 2019, IEEE Access.

[26]  Shuilong He,et al.  A Novel Deep Learning Network via Multiscale Inner Product With Locally Connected Feature Extraction for Intelligent Fault Detection , 2019, IEEE Transactions on Industrial Informatics.

[27]  Fei Han,et al.  Efficient network architecture search via multiobjective particle swarm optimization based on decomposition , 2019, Neural Networks.

[28]  Xu Li,et al.  Deep representation clustering-based fault diagnosis method with unsupervised data applied to rotating machinery , 2020 .

[29]  Xuefeng Chen,et al.  Reweighted generalized minimax-concave sparse regularization and application in machinery fault diagnosis. , 2020, ISA transactions.

[30]  Yong Yu,et al.  Efficient Architecture Search by Network Transformation , 2017, AAAI.

[31]  Ruixin Wang,et al.  A reinforcement neural architecture search method for rolling bearing fault diagnosis , 2020 .

[32]  Jaskaran Singh,et al.  Multisensor data fusion for gearbox fault diagnosis using 2-D convolutional neural network and motor current signature analysis , 2020 .

[33]  Ning Ding,et al.  Journal bearing seizure degradation assessment and remaining useful life prediction based on long short-term memory neural network , 2020 .