%A Shuo Meng, Jianshe Kang, Kuo Chi, and Xupeng Die %T Intelligent Fault Diagnosis of Gearbox based on Multiple Synchrosqueezing S-Transform and Convolutional Neural Networks %0 Journal Article %D 2020 %J Int J Performability Eng %R 10.23940/ijpe.20.04.p4.528536 %P 528-536 %V 16 %N 4 %U {https://www.ijpe-online.com/CN/abstract/article_4383.shtml} %8 2020-04-30 %X In order to solve the problem of gearbox fault diagnosis, we proposed a gearbox fault diagnosis method based on multiple synchrosqueezing S-transform (MSSST) and convolutional neural network (CNN). Firstly, the time-frequency analysis method of MSSST is used to solve the problem that general time-frequency analysis methods cannot obtain good time-frequency aggregation when processing the strong time-varying signals. Then, convolutional neural network is used to extract the image features of time-frequency graphs and classify the faults to diagnosis. Finally, we set up the combination of different time-frequency analysis methods and different deep learning models to compare and analyze the advantages and disadvantages among different methods and verify the effectiveness of the MSSST and CNN methods. The test results show that the fault diagnosis method based on MSSST and CNN has the highest accuracy and shortest operation time compared with other methods, which verifies the effectiveness and superiority of this method.