On the use of DPCA in process fault detection and identification

This paper proposes an innovative Dynamic Principal Component Analysis (DPCA) scheme to perform fault detection and identification (FDI) for systems affected by process faults. In this scheme, a new modeling method is proposed to fix the data lag and the number of principal components to retain. A new structuration method is introduced to identify the process fault. It is based on the computation of common angles between residual subspaces of different modes. This structuration yields a concise set of detection indices, specifically sensitive to certain process faults while remaining unsensitive to others. The proposed FDI scheme is successfully applied to a three tank affected by several process faults.