Process Disturbances Detection via Spectral Graph Analysis

MUSULIN, E.

24 th European Symposium on Computer Aided Process Engineering

Elsevier

Lugar: Amsterdam; Año: 2014; p. 1885 - 1890

In this work, Spectral Graph Analysis Monitoring (SGAM) is introduced as a method for  process  disturbance  detection.  It  is  shown  that  processes  can  be  monitored  by analysing the spectral properties of a properly defined weighted graph. The developed technique can be used as an alternative or complement to PCA, LPP and other on-line statistical  monitoring  approaches.  SGAM  has  been  illustrated  in  an  autocorrelated synthetic  case,  where  several  types  of  process  disturbances  have  been  evaluated, including steps, drifts and random variations.