A novel fusion approach of deep convolution neural network with auto-encoder and its application in planetary gearbox fault diagnosis

Chen, Fafa; Liu, Huan; Tang, Baoping; Chen, Baojia; Zhang, Fajun

doi:10.1177/1748006x20964614

Cited by 10 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the authors found that the Recurrent Neural Network, such as 3D Convolutional Neural Network [63,64] or Long and Short Term Memory [65,66] with recursive structures can be used to extract temporal features from historical data and to simulate the temporal relationships between individual data points clearly. Therefore, the recursive structures will be considered and involved in the future to make the hidden layers of DNN model to achieve self-invocation across time nodes.…”

Section: Discussionmentioning

confidence: 99%

An Advanced Operation Mode with Product-Service System Using Lifecycle Big Data and Deep Learning

Ren

Zhang

Sakao

et al. 2021

Int. J. of Precis. Eng. and Manuf.-Green Tech.

View full text Add to dashboard Cite

As a successful business strategy for enhancing environmental sustainability and decreasing the natural resource consumption of societies, the product-service system (PSS) has raised significant interests in the academic and industrial community. However, with the digitisation of the industry and the advancement of multisensory technologies, the PSS providers face many challenges. One major challenge is how the PSS providers can fully capture and efficiently analyse the operation and maintenance big data of different products and different customers in different conditions to obtain insights to improve their production processes, products and services. To address this challenge, a new operation mode and procedural approach are proposed for operation and maintenance of bigger cluster products, when these products are provided as a part of PSS and under exclusive control by the providers. The proposed mode and approach are driven by lifecycle big data of large cluster products and employs deep learning to train the neural networks to identify the fault features, thereby monitoring the products’ health status. This new mode is applied to a real case of a leading CNC machine provider to illustrate its feasibility. Higher accuracy and shortened time for fault prediction are realised, resulting in the provider’s saving of the maintenance and operation cost.

show abstract

Section: Discussionmentioning

confidence: 99%

An Advanced Operation Mode with Product-Service System Using Lifecycle Big Data and Deep Learning

Ren

Zhang

Sakao

et al. 2021

Int. J. of Precis. Eng. and Manuf.-Green Tech.

View full text Add to dashboard Cite

show abstract

“…Lei Yaguo et al of Xi'an Jiaotong University [6] solved the problem of selecting the kernel functions of the correlation vector machine by adaptive combination of multiple kernel functions, and achieved good recognition results in the prediction of the remaining life of rolling bearings. Chen Fafa et al [7] designed a new deep convolution automatic coding neural network, which automatically learns the characteristics of the Hilbert envelope spectral spatial data of the gearbox vibration signal with the network, and verifies the effectiveness and practicability of the method in equipment fault diagnosis through the analysis of gearbox fault experiments.…”

Section: Introductionmentioning

confidence: 99%

Improvement of one-dimensional convolution neural network and its application in gearbox fault diagnosis

Yang,

Jiao,

Yang

et al. 2023

Third International Conference on Advanced Algorithms and Neural Networks (AANN 2023)

View full text Add to dashboard Cite

Gearbox is a key component of large rotating machinery. To ensure the safe operation of large rotating machinery and to meet the demand for automatic and intelligent gearbox fault diagnosis, this paper investigates the use of the frequency spectrum of gearbox vibration signals in an improved One-dimensional Convolution Neural Network (1D-CNN) to diagnose early weak faults in gearboxes. In view of the problem that the maximum pooling algorithm loses too much information and the average pooling algorithm blurs important information in the common convolution neural network (CNN), a small-scale convolution kernel with a moving step of 3 is proposed to replace pooling operation, so as to realize the role of automatically adjusting the weight extraction characteristics in the process of down-sampling and model training. Also, the global max pooling is used to extract the effective information of the final convolution layer instead of the global average pooling, and the regularized Dropout technique is used to further optimize the deep convolution neural network (DCNN), significantly reducing the number of network parameters and minimizing the phenomenon of network over-fitting. The research is carried out on an experimental platform for gearbox failures, the results show that the improved 1D-CNN has a correct fault identification rate of over 98%, which is 20% higher than that of the previous CNN, and the network has better noise immunity and stability. The method in this paper can provide a basis for researchers in the construction of DCNN and has certain practical engineering application value for fault diagnosis of gearboxes.

show abstract

“…The approach is to train the model in the source domain and fine-tune the model with a small number of labeled samples in the target domain [23]. Chen et al [24] used a small number of labeled samples to fine-tune the improved CNN to solve the gearbox fault diagnosis problem in non-stationary conditions and verified the method's robustness. Zhang et al [25] fine-tuned the improved deep neural network (DNN) model with a small number of samples in the target domain and achieved high fault diagnosis accuracy.…”

Section: Introductionmentioning

confidence: 99%

Rotating machinery fault diagnosis based on transfer learning and an improved convolutional neural network

Jiang¹,

Zheng²,

Li³

2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Deep learning-based methods have been widely used for rotating machinery fault diagnosis. However, they would exhibit poor performance due to the severe data distribution difference under variable working conditions. Therefore, we first develop an improved convolutional neural network, consisting of multi-scale convolutional layer (MSC), channel attention layer (CA), and inception network structure (INS). Compared with other models, our model has strong feature extraction ability, fewer parameters and less training cost. Subsequently, based on transfer learning (TL), we propose the MSC-CA-INS-TL method. In order to improve the model’s generalization ability, we propose an appropriate fine-tuning strategy to coordinate with the model and pay attention to the accuracy of both source and target domains during migration. The bearing datasets and gear experimental platforms are used to verify the proposed method, and high fault diagnosis accuracy and stability are achieved under variable working conditions and small samples.

show abstract

A novel fusion approach of deep convolution neural network with auto-encoder and its application in planetary gearbox fault diagnosis

Cited by 10 publications

References 35 publications

An Advanced Operation Mode with Product-Service System Using Lifecycle Big Data and Deep Learning

An Advanced Operation Mode with Product-Service System Using Lifecycle Big Data and Deep Learning

Improvement of one-dimensional convolution neural network and its application in gearbox fault diagnosis

Rotating machinery fault diagnosis based on transfer learning and an improved convolutional neural network

Contact Info

Product

Resources

About