Research on Open Circuit Fault Diagnosis of Inverter Circuit Switching tube Based on Machine Learning Algorithm

The inverter circuit is widely used, and the switching tube has the highest incidence of open circuit failure. In order to improve the recognition rate of the open circuit fault of the inverter circuit, this paper proposes a method to extract the output voltage, output current and input current time domain features and then use the Random Forest and K-Nearest Neighbor to identify the fault. In this paper, through the simulation of single-phase full-bridge inverter circuit, the open-circuit fault test of the switch tube is simulated, and the output voltage, output current and DC-side input current of each switch tube open-circuit fault are obtained, and then the time domain feature extraction is performed. The Random Forest and K-Nearest Neighbor are used for diagnostic comparison. The results show that the single-tube fault recognition rate can reach more than 96% and the highest fault recognition rate can reach 99.77%.

[1]  S. Ding,et al.  Fault classification method for inverter based on hybrid support vector machines and wavelet analysis , 2011 .

[2]  Furqan Asghar,et al.  A Novel Three-Phase Inverter Fault Diagnosis System Using Three-dimensional Feature Extraction and Neural Network , 2019 .

[3]  Won-Sul Shim,et al.  A real-time fault diagnosis system for UPS based on FFT frequency analysis , 2010, 2010 Twenty-Fifth Annual IEEE Applied Power Electronics Conference and Exposition (APEC).

[4]  Jian Sun,et al.  Face Alignment at 3000 FPS via Regressing Local Binary Features , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Dong-Choon Lee,et al.  Fault diagnosis of three-phase PWM inverters using wavelet and SVM , 2008, 2008 IEEE International Symposium on Industrial Electronics.

[6]  Antanas Verikas,et al.  Mining data with random forests: A survey and results of new tests , 2011, Pattern Recognit..

[7]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[8]  Jian Sun,et al.  Global refinement of random forest , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Juergen Gall,et al.  Class-specific Hough forests for object detection , 2009, CVPR.

[10]  Andrew W. Fitzgibbon,et al.  Real-time human pose recognition in parts from single depth images , 2011, CVPR 2011.

[11]  Giorgio Rizzoni,et al.  Mechanical signature analysis using time-frequency signal processing: application to internal combustion engine knock detection , 1996, Proc. IEEE.