통계적기법을 이용한 폴리스티렌 반응기 모니터링 시스템 개발

Abstract

In chemical processes, faults can trigger serious problems and deviations which occur accidents. Fault diagnosis models should quickly identify the root cause of faults to mitigate the loss. Most previous researches in the field of fault diagnosis model just handle the data set of benchmark process simulated on MATLAB. To design a fault diagnosis model, the analysis of a process and its data should be performed. In this study, a polystyrene process is tested. In this process, a runaway reaction occurred and this caused a large loss since operators were late aware of the occurrence of this accident. To design a proper fault diagnosis model, we analyzed the process and tested a real accident data set. At first, a mode classification model based on support vector machine (SVM) was trained and principal component analysis (PCA) model for each mode was constructed under normal operation conditions. The results show that a proposed model can quickly diagnose the occurrence of a fault and they indicate that this model is able to reduce the potential loss.