Abstract：To solve the problem of traditional fault diagnosis for tea garden tractor with serious delay during collecting fault information, a remote fault diagnosis system(RFDS) of tea garden tractor was proposed based on fuzzy neural network(FNN). Combining the advantages of fuzzy algorithm and neural network, through the real-time monitoring of vehicle operation information, the vehicle information was remotely processed to obtain vehicle potential fault and real-time fault information for avoiding a large number of detections in traditional methods, and more time was gained for maintenance and improving production efficiency. Based on the simulation software of Carsim, the vehicle operation was simulated in real time,and the algorithm model was trained and verified by training set and test set. The results show that FNN diagnosis algorithm can meet the performance requirements of the system, and the accuracy can reach more than 90%. The tractor fault in tea garden can be diagnosed remotely. RFDS technology also plays an important role in the development of vehicles. The performance of vehicles can be evaluated through remote fault diagnosis system in the R & D stage of vehicles, which saves a lot of human resources.
WE X B,QIAO W G,DU C Q,et al.Design and test of tractor electrical fault detection system based on LabVIEW [J]. Journal of Chinese Agricultural Mechanization, 2020, 41(7): 143-149.（in Chinese）
WANG Y Q,GAO J,ZHANG J H, et al.Research on the development status of turbine fault diagnosis techno-logy[J]. Mechanical and Electrical Information, 2020 (21): 30-32. （in Chinese）
FISCHER F, DEUTSCHER J. Flatness-based algebraic fault diagnosis for distributed-parameter systems [J]. Automatica, doi：10.1016/j.automatica.2020.108987.
［5］
LIU Z, JIN Y, ZUO M J, et al. ACCUGRAM: a novel approach based on classification to frequency band selection for rotating machinery fault diagnosis [J]. ISA Transactions, 2019, 95：346-357.
［6］
BHANDARI G P, RATNESHWER R. Dependency-based fault diagnosis approach for SOA-based systems using Colored Petri Nets [J]. Journal of King Saud University-Computer and Information Sciences, doi:org/10.1016/j.jksuci.2018.12.002.
［7］
ZHANG J. Research of remote fault diagnosis system of electric locomotive based on 4G network [J]. Advances in Engineering Research，2018，163：1777-1780.
CHEN H D,ZHU J B.The development status and trend of remote monitoring technology[J]. Information Recording Materials, 2019, 20(8): 42-43. （in Chinese）
LYU H W,ZHAO H,WANG H Z, et al. Fault diagnosis algorithm for MVB based on fuzzy neural network[J]. Journal of Jilin University(Science Edition), 2020, 58(1): 104-108. （in Chinese）
LU Y H,PENG H P. Circuit fault diagnosis based on wavelet packet analysis and relevance vector machine[J]. Journal of Jilin University(Science Edition), 2015, 53(5): 981-986. （in Chinese）
ELNASHAR G, BHAT R B, SEDAGHATI R. Modeling and dynamic analysis of a vehicle-flexible pavement coupled system subjected to road surface excitation [J]. Journal of Mechanical Science and Technology, 2019, 33(7): 3115-3125.
[14]
WANG T, LIU W, ZHAO J, et al. A rough set-based bio-inspired fault diagnosis method for electrical substations [J]. International Journal of Electrical Power & Energy Systems, doi:10.1016/j.ijepes.2020.105961.
[15]
ZHANG X, CHEN G, HAO T, et al. Rolling bearing fault convolutional neural network diagnosis method based on casing signal [J]. Journal of Mechanical Science and Technology, 2020, 34(2):1-10.
[16]
BOUM A, YOME N, LEANDRE N. Fault diagnosis of an induction motor based on fuzzy logic, artificial neural network and hybrid system [J]. International Journal of Control Science and Engineering ，2018, 8(2): 42-51.
2021, Vol. 42 
