CH5440 Multivariate Data Analysis for Process Modeling


Use of data analysis for process performance enhancementMathematical background: Sources and characterization of errors in data,Random variables, probability density functions, estimation, confidenceintervals and hypothesis testing, stochastic signals, frequency domainanalysis of signals, measures of nongaussianity.Process Modeling: Model structures, linear regression, Nonlinearregression, principla Component Analysis, Independent Component Analysis.Applications: Parameter estimation in linear and nonlinear processes,Data Reconcilation, Continuous/Batch process monitoring using MSPC,controller performance monitoring, fault diagonis, chemometrics,biomedical and speedh signal processing.

Text Books:

  1. Johnson R A and D W Wichern, Applied Multivariate StatisticalAnalysis, Prentice Hall, 2002.
  2. Montgomery, D C and G C Runger, Applied Statitics andProbability for engineers, Wiley, 2003