This system presents an effective diagnosis algorithm for permanent magnet synchronous motors running with an array of faults of varying severity over a wide speed range.

與傳統的小波包轉換(wavelet packet transform, WPT)作為馬達定子電流特徵提取的工具相比較，並說明深度學習對於傳統特徵提取方法的優勢。
本系統不只提出1D卷積神經網路一種網路架構，還另外提出了堆疊自編碼器的架構對此進行比較。

Diagnosis system can be applied to industrial plants. Through the industrial network of Industry 4.0, we can instantly collect the stator current data of any faulty motor and continuously update the prediction model through the diagnostic system to monitor the status of the motor in real time.