Performance assessment and diagnosis of refinery control loops

This paper discusses the application of control loop performance assessment (Desborough and Harris, 1992) in a refinery setting. In a large process it is not feasible to tailor the parameters of the algorithm to every individual control loop. A procedure is illustrated for selecting default values which make it possible to implement the technology on a refinery-wide scale. For instance, it is shown that the prediction horizon perameter in the CLPA algorithm can be set so that the analysis is sensitive to the persistent signals that cause loss of performance. Default values are suggested for refinery applications.A frequent cause of loss of performance in a control loop is a persistent oscillation due to a valve nonlinearity or a tuning fault. The paper presents an operational signatures in the form of an estimate of the closed loop impulse response that suggest the causes of such oscillations.

[1]  Karl Johan Åström,et al.  Assessment of Achievable Performance of Simple Feedback Loops , 1988 .

[2]  G. D. Martin,et al.  Estimating control function benefits , 1991 .

[3]  J. Macgregor,et al.  Monitoring and Diagnosing Process Control Performance: The Single-Loop Case , 1991, 1991 American Control Conference.

[4]  Manfred Morari,et al.  Performance monitoring of control systems using likelihood methods , 1996, Autom..

[5]  Tore Hägglund Automatic Monitoring of Control Loop Performance , 1994 .

[6]  Matthew L. Tyler,et al.  Performance monitoring of control systems using likelihood methods , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[7]  Thomas E. Marlin,et al.  Monitoring and Diagnosing Process Control Performance: The Single-Loop Case , 1991 .

[8]  Tore Hägglund,et al.  A Control-Loop Performance Monitor , 1995 .

[9]  Brian Surgenor,et al.  Plant-Wide Feedback Control Performance Assessment using an Expert System Framework , 1995 .

[10]  George J Thaler Automatic Control Systems , 1989 .

[11]  Thomas E Marlin,et al.  Process Control , 1995 .

[12]  G.A. Dumont,et al.  Control loop performance monitoring , 1996, IEEE Trans. Control. Syst. Technol..

[13]  T. Harris,et al.  Performance assessment measures for univariate feedback control , 1992 .

[14]  P. Welch The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms , 1967 .

[15]  Tore Hägglund,et al.  Detection and Diagnosis of Oscillation in Control Loops , 1997 .

[16]  Nina F. Thornhill,et al.  Refinery-wide control loop performance assessment , 1999 .

[17]  T. Harris Assessment of Control Loop Performance , 1989 .