Improved Principal Component Analysis Models for Fault Detection Using Delay Adjustment

Abstract

Principal component analysis (PCA) has been widely and successfully used with process monitoring purposes. However, it has an important drawback as it does not account for time-delays present in data thus causing an inefficient dimensionality reduction of process variables and the subsequent poor monitoring and disturbance detection performance. In this chapter, a new method, the genetic algorithm based delay adjusted PCA (GA-DAPCA), based on genetic algorithm optimization is proposed to improve the PCA performance in the presence of time delays between process signals. The optimization is performed in two loops. The first loop finds the shift between variables that minimize the number of principal components to be considered as common cause variance, the second loop maximizes the variance contained in the previously selected principal component dimensions. The methodology is demonstrated through motivating case studies in the bio-ethanol processor system (see Chap.  9). The obtained results are presented with the same set of faults considered in Chap.  13.