An Intelligent Process Fault Diagnosis System Integrating Andrews Plot, PCA and Neural Networks

With industrial production processes becoming more and more sophisticated, traditional fault diagnosis systems may not achieve reliable diagnosis performance. In order to improve fault diagnosis performance, this paper proposes an enhanced fault diagnosis system by integrating neural networks with Andrews plot and principal component analysis (PCA). On-line measurements are first processed by Andrews plot and then fed to a neural network directly or via PCA for fault classification. Application to a simulated CSTR process indicates that the proposed method can give more reliable and earlier diagnosis than the traditional neural network based fault diagnosis method.