Research on feature fusion strategy for gear states diagnosis based on fusion assessment

ABSTRACT

Gear is commonly applied as a vital component that connects and transmits power, which plays a crucial role in mechanical system. In order to represent the running states of the gear effectively and improve the diagnosis accuracy further, a feature fusion strategy based on fusion assessment to obtain optimum feature fusion pattern is proposed in this research, including two-direction feature fusion strategy and three-direction feature fusion strategy. Firstly, different feature extraction methods are employed to extract the features of vibration signal in each direction (X, Y and Z axis), respectively. Then, the best fusion mode is determined by fusion assessment mechanism based on fuzzy logic in two directions and vibration intensity of signal in three directions. Finally, support vector machine (SVM) and decision tree (DT) are selected to verify the validity and universality of the proposed method. Experimental studies show that feature fusion strategy for gear states diagnosis based on fusion assessment can take full advantage of the complementary performances of different feature extraction methods and signal characteristics in different directions, which can fully represent the health states of the running gear and effectively improve the diagnosis accuracy.

KEYWORDS:

• Gear
• health states diagnosis
• feature fusion strategy
• fusion assessment
• mode classification

Acknowledgments

This work was funded by National Natural Science Foundation of China (61773078, 61701006), Industrial Technology Project Foundation of Changzhou Government (CE20175040), the Natural Science Foundation of Anhui Province (1708085QF147), Jiangsu Graduate Scientific Research Innovation Program (SJCX20_0885, KYCX20_2533). The authors would like to offer special thanks to Yan Ruqiang team of Southeast University for providing the free open experimental data. The authors would also like to thank the anonymous reviewers for their very useful suggestions and comments.

Disclosure statement

The authors declare that there is no conflict of interests regarding the publication of this paper.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [61701006]; National Natural Science Foundation of China [61773078]; Natural Science Foundation of Anhui Province [1708085QF147]; Industrial Technology Project Foundation of Changzhou Government [CE20175040]; Jiangsu Graduate Scientific Research Innovation Program [SJCX20_0885]; Jiangsu Graduate Scientific Research Innovation Program [KYCX20_2533].

Notes on contributors

Lizheng Pan

LIZHENG PAN is currently a lecturer in the school of mechanical engineering, Changzhou University, China. His research interests include mechanical fault diagnosis, rehabilitation robot and man-machine interaction.

Lu Zhao

LU ZHAO is a master degree candidate in the study of mechanical fault diagnosis in the school of mechanical engineering, Changzhou University, China.

Dashuai Zhu

DASHUAI ZHU received the M.S. degree in mechanical engineering from the school of mechanical engineering, Changzhou University, China, in 2019. His research interests include mechanical fault diagnosis.

Shigang She

SHIGANG SHE is a professor in the school of mechanical engineering, Changzhou University, China. His research mainly focuses on precision measurement and control, and intelligent equipment design.

Xianchuan Shi

XIANCHUAN SHI is an associate professor in the school of mechanical engineering, Changzhou University, China. His research mainly focuses on digital equipment technology and precision electrochemical machining.

Zhu Zhu

ZHU ZHU is currently an associate professor in the school of electronic engineering and intelligent manufacturing, Anqing Normal University, China. His research mainly focuses on image processing and pattern recognition.