深層学習を用いた機器・プラントの異常検知と識別

Anomaly detection and identification of equipment and plants using deep learning

出町 和之*; Dong, F.*; 阿部 哲*; Chen, S.*; 高屋 茂  ; 関 暁之   ; 吉川 雅紀 ; 三木 大輔*

Demachi, Kazuyuki*; Dong, F.*; Abe, Toru*; Chen, S.*; Takaya, Shigeru; Seki, Akiyuki; Yoshikawa, Masanori; Miki, Daisuke*

In this study, a method was developed to detect anomalies using a deep learning model and further identify the types of anomalies. In order to verify the performance of this method, the calculation results of the plant state time series data by the plant simulator and the benchmark data set of the vibration sensor values in the rotating equipment were analyzed.