Adaptability of statistical process control tools to modifications in plant operation strategies

The influence of modifications in the operation strategy of a chemical plant on the output of statistical process control (SPC) tools developed for fault detection is examined. Particularly, the square prediction error (SPE) and the Hotelling T2 charts are analyzed for reported problems. The effect...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Maestri, M.L., Cassanello, M.C., Horowitz, G.I., Atanor; Compania Mega S.A.; Consejo Federal de Inversiones (CFI); Dow; et al.; Petroquimica Rio Tercero SA (PRIII)
Formato: CONF
Materias:
Fault detection
Fault diagnosis
Process control
Chemical detection
Chemical modification
Chemical plants
Failure analysis
Data bank
Hotelling T
In-plant operations
Operation strategy
Square prediction errors
Statistical process controls (SPC)
Statistical process control
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_NIS12852_v_n_p_Maestri
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
Descripción
Sumario:The influence of modifications in the operation strategy of a chemical plant on the output of statistical process control (SPC) tools developed for fault detection is examined. Particularly, the square prediction error (SPE) and the Hotelling T2 charts are analyzed for reported problems. The effect of training the tools either with an extended historic databank obtained under standard operation or including also non-conventional conditions is studied. The ability of the tools to provide a specific alarm is examined in both cases. In addition, the sensitivity for diagnosing a given problem by analyzing the contributions per variable to the SPE and the T2 is addressed.

Ejemplares similares