| 학회 | 한국공업화학회 |
| 학술대회 | 2022년 봄 (05/11 ~ 05/13, 제주국제컨벤션센터(ICC JEJU)) |
| 권호 | 26권 1호 |
| 발표분야 | 포스터-화학공정 |
| 제목 | Abnormality Detection using Time-frequency features and machine learning for Gas Blending Process |
| 초록 | The high-resolution gas industries (i.e., semiconductor and display) need specially purposed gas. Special gas supplied to these industries must keep the quality of ultra-high purity. To meet customer needs, a system is needed that can predict gas purity in advance, and take precautions against operational abnormalities through real-time process data-based machine learning. In this study, past normal/abnormal time-frequency data are classified and machine-learned to check whether it is normal in real time. Through this, productivity can be improved by preventing product quality abnormalities. |
| 저자 | 이주성, 이철진, 이가영, 임준영 |
| 소속 | 중앙대 |
| 키워드 | Abnormality detection; Time-frequency; Time-series; Gas blending process |
