Detection and Diagnosis of Plant-wide Oscillations: An application study

This paper presents an overview of the emerging techniques for oscillation detection and diagnosis and investigate their efficiency through an industrial case study. The recently proposed autocorrelation function based method [15] is used for detection of multiple oscillations in process measurements and identifying signals having common oscillations. The signals having common oscillatory behavior are analyzed for the possible presence of valve stiction using the Higher Order Statistical method [1]. This method is helpful in identifying the key variables that are most likely to be the root cause of oscillations. We also present some issues pertinent to the diagnosis of oscillations, which are potential directions for future research.

[1]  Sirish L. Shah,et al.  Detection and Quantification of Control Valve Stiction , 2004 .

[2]  Guy A. Dumont,et al.  Detection and diagnosis of oscillations in control loops , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[3]  T. Miao,et al.  Automatic detection of excessively oscillatory feedback control loops , 1999, Proceedings of the 1999 IEEE International Conference on Control Applications (Cat. No.99CH36328).

[4]  Nina F. Thornhill,et al.  Path analysis for process troubleshooting , 2002 .

[5]  N. F. Thornhilla,et al.  Spectral principal component analysis of dynamic process data , 2002 .

[6]  A qualitative shape analysis formalism for monitoring control loop performance , 2001 .

[7]  Alexander Horch.  A simple method for detection of stiction in control valves , 1999 .

[8]  Nina F. Thornhill,et al.  Detection of multiple oscillations in control loops , 2003 .

[9]  Chunming Xia,et al.  Loop status monitoring and fault localisation , 2003 .

[10]  L. Desborough,et al.  Increasing Customer Value of Industrial Control Performance Monitoring—Honeywell’s Experience , 2002 .

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

[12]  Nina F. Thornhill,et al.  Diagnosis of poor control-loop performance using higher-order statistics , 2004, Autom..

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

[14]  Nina F. Thornhill,et al.  Diagnosis of plant-wide oscillation through data-driven analysis and process understanding , 2003 .

[15]  Dale E. Seborg,et al.  Monitoring Model Predictive Control Systems Using Pattern Classification and Neural Networks , 2003 .