(371s) A MSPC Technique for Identifying Biases in Industrial Processes

Measurement bias is one type of systematic error that can be caused by many sources, such as poorly calibrated or malfunctioning instruments. Several model-based approaches have been proposed for bias detection and identification, which work comparing the actual operation of the plant with that predicted by a mathematical model using hypothesis statistical tests. A good survey of these techniques is available in the books by Narasimhan and Jordache (2000) and Romagnoli and Sánchez (2000).

Regarding Multivariate Statistical Process Control (MSPC) techniques, the -statistic is widely used in SPC to reliably detect the out of control status, but by itself it offers no assistance as fault identification tool. Different strategies have been proposed to calculate the contribution of each process variable to the inflated statistic. They work in the original or in the latent variable space.

A straightforward method to decompose the -statistic as a unique sum of each variable contribution was recently developed by Alvarez et al. (2007), which is called OSS (Original Space Strategy). This decomposition was succesfully applied to detect and identify biases for steady state processes (Sanchez et al., 2008). Later on Alvarez et al. (2008) proposed a new strategy to estimate the influence of a given variable on the final value of the inflated statistic's value. In this approach, the contribution of each variable is measured in terms of the distance between the current observation and its Nearest In Control Neighbour (NICN). The detection and identification capabilities of the NICN technique are evaluated and compared with those corresponding to the most commonly used systematic error detection and identification techniques for some benchmarks (Cedeño et al., 2009). As results indicated the technique succeeds in identifying single and multiple biases, it was applied for monitoring a petrochemical process. In this work, the results of this industrial application of the methodology are presented. Also a discussion is provided about the problems that arise during its implementation and how they are solved.

References

Alvarez, R.; Brandolin, A.; Sánchez, M. (2007) On the Variable Contributions to the D-statistic. Chemometr. and Intell. Lab. System., 88, 89-196.

Alvarez, R.; Brandolín, A.; Sánchez, M.; Puigjaner, L. (2008) A Nearest In Control Neighbour Based Method to Estimate Variable Contributions to the Hotelling's Statistic. Proceedings of 2008 AIChE Annual Meeting, Philadelphia, USA, November 16?21.

Cedeño, Marco; Galdeano, Rubén; Elwart, Juan; Alvarez, Rodrigo; Mabel Sánchez; ?Bias Detection and Identification Using Historical Data?, Proceedings of 2009 AIChE Annual Meeting, Nashville, TN, USA, November 8?13 2009.

Narasimhan, S.; Jordache, C. (2000) Data Reconciliation and Gross Error Detection; Gulf Publishing Company, Houston. Romagnoli, J.; Sánchez M. (2000) Data Processing and Reconciliation for Chemical Process Operations; Academic Press: San Diego.

Sánchez, M.; Alvarez, R.; Brandolin, A. (2008) A MSPC procedure for bias identification in steady state processes. AIChE Journal, 54, 8, 2082-2088.