Fault detection for distillation column using multivariate stastistical process control (MSPC)

Abstract

Chemical process is inclined to be a large-scale, complex and having stringent requirements on the desired quality. It also utilizes a lot of energy, must be environmentally friendly and fulfill safety requirements. Accurate process fault detection at an early stage of the process is important to modern chemical plant in achieving the above requirements. This paper focuses on the application of Multivariate Statistical Process Control (MSPC) as a fault detection tool. An industrial distillation column is modelled and chosen as the case study for this research. Principal Component Analysis (PCA) and Partial Correlation Analysis (PCorrA) are used to develop the correlation coefficients between the variables of the process. Faults considered in the research are sensor failures, valve failures and controller malfunctions. Shewhart Control Chart with the developed correlation coefficients are used for detecting the faults. Results show that both methods based on PCorrA and PCA are able to detect the pre-designed faults.