Fault Detection and Diagnosis for Plasticizing Process of Single-Base Gun Propellant Using Mutual Information Weighted MPCA under Limited Batch Samples Modelling

Aiming at the lack of reliable gradual fault detection and abnormal condition alarm evaluation ability in plasticizing process single-base gun propellant, a diagnosis method based on normalized mutual information weighted multiway principal component analysis (NMI-WMPCA) under limited batch samples modelling was proposed. In this method, differences coupling correlation among multi-dimensional variables characteristics linear nonlinear relationships are considered. NMI-WMPCA utilizes generalization multi-model to establish an accurate model samples, adopts methods SPE statistic contribution plot identify source. The experimental results demonstrate that proposed is effective, which can realize rapid multiple faults process.