Design of Fault Diagnosis System of FPSO Production Process Based on MSPCA

Based on the theory of wavelet analysis and principal component analysis, Multi-scale PCA is introduced which combines the ability of PCA to decorrelate the variables by extracting a linear relationship, with that of wavelet analysis to extract deterministic features and approximately decorrelate autocorrelated measurements to improve the performance of PCA whose modeling is limited to a single scale. It is applied to the fault monitor and diagnose of Floating Production Storage and Off-loading System. The result show: the fault diagnose method based on multi-scale principal components analysis can realized FPSO earlier period fault monitor and diagnose accurately, and the capability of multi-scale principal components analysis fault diagnosis is better than the principal components analysis for the small disturbance.