Fault Detection and Diagnosis in High Dimensional Processes Using Model Driven Multivariate Control Charts

• Jianfeng Ding (JMP)
• Jeremy Ash (JMP)

The Model Driven Multivariate Control Chart (MDMCC) is a new platform in JMP that allows users to efficiently monitor many correlated process variables. The platform can interface with the PCA and PLS platforms to monitor multivariate process variation over time, give advanced warnings of process shifts, and suggest probable causes of process changes. We will demonstrate several use cases. First, we demonstrate fault diagnosis in an offline setting with an example from polyethylene manufacturing. Offline diagnosis often involves switching between many multivariate control charts, univariate control charts, and diagnostic plots. MDMCC provides a user-friendly way to move between these plots. Next, we demonstrate online monitoring of a PLS model using a simulator of a real world industrial chemical process — the Tennessee Eastman Process. We demonstrate how MDMCC can perform online monitoring by connecting JMP to an external database. Measuring product quality variables often involves a time delay before measurements are available, which can delay fault detection substantially. When MDMCC monitors a PLS model, the variation of product quality variables is monitored as a function of process variables. Since process variables are often more readily available, this can aide in the early detection of faults.