Jinfeng Gao

Prof. Dr. Jinfeng Gao

Zhejiang Sci-Tech University, Hangzhou, Zhejiang, China
orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2022 in ISA Transactions
Verified
Journal Article
Closed access
JWJiaqi Wang
Jiaqi Wang
Jinfeng GaoJinfeng Gao
Jinfeng Gao
PWPing Wu
Ping Wu

3 November 2021 in Sensors
Verified
Journal Article
Open access
PDF file
JHJiajun He
Jiajun He
Ping WuPing Wu
Ping Wu
Yizhi TongYizhi Tong
Yizhi Tong
Xujie ZhangXujie Zhang
Xujie Zhang
MLMeizhen Lei
Meizhen Lei
Jinfeng GaoJinfeng Gao
Jinfeng Gao
Bearings are the key and important components of rotating machinery. Effective bearing fault diagnosis can ensure operation safety and reduce maintenance costs. This paper aims to develop a novel bearing fault diagnosis method via an improved multi-scale convolutional neural network (IMSCNN). In traditional convolutional neural network (CNN), a fixed convolutional kernel is often employed in the convolutional layer. Thus, informative features can not be fully extracted for fault diagnosis. In the proposed IMSCNN, a 1D dimensional convolutional layer is used to mitigate the effect of noise contained in vibration signals. Then, four dilated convolutional kernels with different dilation rates are integrated to extract multi-scale features through the inception structure. Experimental results from the popular CWRU and PU datasets show the superiority of the proposed method by comparison with other related methods.

1 July 2020 in Journal of the Franklin Institute
Verified
Journal Article
Closed access
Jinfeng GaoJinfeng Gao
Jinfeng Gao
ZZZhen Zhao
Zhen Zhao
JWJinxia Wang
Jinxia Wang
TTTian Tan
Tian Tan
MMMiao Ma
Miao Ma

10 September 2019 in IEEE Access
Verified
Journal Article
Open access
SMShan Miao
Shan Miao
Jinfeng GaoJinfeng Gao
Jinfeng Gao
In this paper, two inverting buck-boost converters which possess wide conversion ratios are proposed. At the proper duty cycles, the inverting buck 2 -boost converter (IB2BC) achieves high step-down gain while the inverting buck-boost 2 converter (IBB2C) realizes high step-up gain. Advantages of the IB2BC and the IBB2C also include fourth-order simple common-ground structures to increase the power density and operability, low components stress to decrease the corresponding losses and then to improve the system efficiency, no sudden changings on capacitors voltage to prevent the instantaneous overcurrent phenomenon. Owing to the special structures of the IB2BC and the IBB2C, their improved single-switch topologies with reduced number of switch and extensional topologies with wider conversion ratios are constructed, respectively. Operational principles and corresponding comparisons of the above mentioned buck-boost converters are deeply studied and analyzed. As the representative, the single-switch inverting buck-boost 2 converter (SIBB2C) is designed and experimented in step-down/up modes for verifying the family of buck-boost converters performance.

1 September 2020 in ISA Transactions
Verified
Journal Article
Closed access
Jinfeng GaoJinfeng Gao
Jinfeng Gao
JLJiahao Li
Jiahao Li
HPHaipeng Pan
Haipeng Pan
ZWZhengguang Wu
Zhengguang Wu
XYXiuxing Yin
Xiuxing Yin
HWHuijiao Wang
Huijiao Wang

12 October 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Ping WuPing Wu
Ping Wu
YWYixuan Wang
Yixuan Wang
XZXujie Zhang
Xujie Zhang
Jinfeng GaoJinfeng Gao
Jinfeng Gao
LWLin Wang
Lin Wang
Yichao LiuYichao Liu
Yichao Liu
Wind turbines often work in harsh environmental conditions. The risk of blade breakage by hitting the tower has increased because of the decreased bending stiffness of the wind turbine blades with the increasing sizes of wind turbines. To address the system dynamics of wind turbines, this paper proposes a Mogrifier Long Short-Term Memory Autoencoder (MLSTM-AE) method to monitor blade breakage using the supervisory control and data acquisition (SCADA) data. Firstly, the Pearson correlation coefficient is calculated for variables selection. Secondly, the time-lagged multivariate variables are augmented and taken as input for the encoder which consists of a Mogrifier Long Short-Term Memory (MLSTM) layer to learn the deep features. And another MLSTM layer runs as the decoder to reconstruct the input. The proposed MLSTM-AE model can extract spatial-temporal information more effectively than traditional Long Short-Term Memory (LSTM) and autoencoder (AE). Next, kernel density estimation is applied to develop boundaries for generating blade breakage alerts based on reconstruction errors, which can reflect changes in system dynamics caused by blade breakage. The advantages of the proposed MLSTM-AE based monitoring method are illustrated by employing real blade breakage cases from serval wind farms located in China by comparing with other related methods in terms of warning time, false alarm rate and accuracy.