| Title |
Study on Development of Deep Learning Fault Diagnosis Algorithm Considering Induction Motor Speed and Load Condition |
| Authors |
한지훈(Ji-Hoon Han) ; 최동진(Dong-Jin Choi) ; 홍선기(Sun-Ki Hong) |
| DOI |
http://doi.org/10.5370/KIEE.2019.68.3.423 |
| Keywords |
Deep learning ; Motor fault diagnosis ; CNN ; Data analysis ; Induction motor ; FFT ; Frequency domain |
| Abstract |
The motor mechanical fault has been diagnosed under fixed driving conditions. The induction motor speed is affected not only by the input frequency but also by the load. In addition, the vibration generated by the induction motor is affected by the speed as well as the input frequency. For these reasons, a data preprocessing algorithm has been developed that shifts the measured data in the frequency domain based on motor speed. The algorithm also takes the input frequency as an input variable and removes the vibration component by the power source frequency. The data processed by the above procedure are classified through the deep learning algorithm based on CNN. As a result, a fault diagnosis system that can be applied to the industrial field has been developed by considering the motor driving conditions using the proposed algorithms. |