Deep Fault Diagnosis for Rotating Machinery with Scarce Labeled Samples

Zhang, Jing; Tian, Jing; Wen, Tao; Yang, Xiaohui; Rao, Yong; Xu, Xiaobin

doi:10.1049/cje.2020.05.016

Cited by 5 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, challenges of IFD were also discussed. Zhang et al [31] proposed a novel deep CNN method which was based on knowledge transferring from shallow models for rotating machinery fault diagnosis with scarce labeled samples. In their work, they first applied short-time Fourier transform (STFT) to extract integral features.…”

Section: Related Workmentioning

confidence: 99%

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

Gupta

Rajnish

Bhattacherjee

2021

Scientific Programming

View full text Add to dashboard Cite

Deep neural network models built by the appropriate design decisions are crucial to obtain the desired classifier performance. This is especially desired when predicting fault proneness of software modules. When correctly identified, this could help in reducing the testing cost by directing the efforts more towards the modules identified to be fault prone. To be able to build an efficient deep neural network model, it is important that the parameters such as number of hidden layers, number of nodes in each layer, and training details such as learning rate and regularization methods be investigated in detail. The objective of this paper is to show the importance of hyperparameter tuning in developing efficient deep neural network models for predicting fault proneness of software modules and to compare the results with other machine learning algorithms. It is shown that the proposed model outperforms the other algorithms in most cases.

show abstract

Section: Related Workmentioning

confidence: 99%

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

Gupta

Rajnish

Bhattacherjee

2021

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…The central fusion algorithm of the existing FL mechanism is not effective enough to deal with the phenomena such as time delay and packet loss in time. Therefore, how to technically solve the problem of data islands under the premise of protecting data privacy, how to make the system use their respective data more efficiently and accurately under the requirements of privacy, security and supervision, and how to acquire better fault diagnostic result under the conditions of fewer samples and weak supervision (with few labeled samples) [ 17 ] , which have become urgent matters.…”

Section: Preliminary Workmentioning

confidence: 99%

An Asynchronous Quasi‐Cloud/Edge/Client Collaborative Federated Learning Mechanism for Fault Diagnosis

Wen

2021

Chinese J of Electronics

View full text Add to dashboard Cite

Although the federated learning method has the ability to balance data and protect data privacy by means of model aggregation, while the existing methods are difficult to achieve the effectiveness of centralized learning under data sharing. The existing federated structure only has a certain degree of confidentiality for data privacy, that is to say, each client can reconstruct a part of the information of other clients based on the model parameters shared between the server and the clients under certain conditions. In order to make the federated learning mechanism more confidential, we breaks the existing mechanism that the parameters between the federated model and the client model are completely shared, and establishes a new asynchronous quasi‐cloud/edge/client collaborative federated learning mechanism. We construct a hierarchical multi‐level confidential communication network, where the network parameters are shared in a way of quasi‐cloud/edge/client coordination without data communication. The cloud and the edges respectively use the sequential Kalman filter algorithm to perform an asynchronous fusion of the network parameters uploaded in their respective fusion centers for the next round of updates; The effectiveness of the proposed algorithm is verified on the data of a type of rotating machinery

show abstract

“…At present, extremely variants with LSTM have been widely used in various fields that need to process continuous data. [17][18][19] In order to improve the performance of neural networks, scholars have made various improvements to the architecture of LSTM in recent years. Zhou et al proposed MGU (Minimal Gated Unit), 20 which achieves the functionality of LSTM by containing only one gate structure effectively reducing the number of unit parameters.…”

Section: Introductionmentioning

confidence: 99%

“…At present, extremely variants with LSTM have been widely used in various fields that need to process continuous data. 17–19…”

Section: Introductionmentioning

confidence: 99%

Study on observation method of injection pressure curve in high pressure common rail system

2022

International Journal of Engine Research

View full text Add to dashboard Cite

The performance of the diesel engine was determined by the injection pressure of the high pressure common rail system. The observation method of the injection pressure curve was of great value in evaluating the operating conditions of the diesel engine. However, the observation method could make it impossible to directly observe the injection pressure data on a high pressure common rail system in operation. In this paper, common rail pressure signals and control pulse widths, which were easy to collect, were used as the basis for proposing a GRU (Gated Recurrent Unit) based model for the observation of injection pressure curves of high pressure common rail system in order to solve the above problem. The accuracy and precision of the model was verified by the experiments and compared with commonly used simulation models. From the results, it was concluded that the overall accuracy of the injection pressure curve observation was similar to that of the simulation model, but the accuracy was higher around the peak of the injection pressure, as well as the equipment requirement was low and the calculation speed was fast. The mean EMD of the curvilinear observation model of injection pressure of high pressure common rail system is similar to that of the 1D simulation model, with 37.88 and 33.62 respectively. The overall difference in injection pressure between the two models is similar with 53.62 and 54.23 MPa respectively. On the same computing platform, the average computation time of the test set containing five states is 0.0113 s, and the simulation time of each running state is 12951.6 s using AMESim software.

show abstract

Deep Fault Diagnosis for Rotating Machinery with Scarce Labeled Samples

Cited by 5 publications

References 21 publications

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

An Asynchronous Quasi‐Cloud/Edge/Client Collaborative Federated Learning Mechanism for Fault Diagnosis

Study on observation method of injection pressure curve in high pressure common rail system

Contact Info

Product

Resources

About