As the noise in industrial processes have remarkable amplitude, frequent happening and timevariant characteristics, an improved performance assessment method based on minimum variance control (MVC) benchmark was presented, in which forgetting factor was introduced to identification process for the performance assessment of control loops. Compared with the traditional approach, the identification method with forgetting factor could not only consider the influence of historical data, but also emphasize the information included in the new data. Through introducing the forgetting factor parameters, data saturation was prevented and the accuracy and stability of control performance assessment was improved in the industrial process with timevariant disturbance. The proposed method was demonstrated by the simulation and industrial PTA process application. The results showed that the method can provide an accurate performance assessment index for industrial control loop, effectually guide the operators to optimize and retune the underlying loops that have potential performance problem, and stabilize the process operation.Key words: performance assessment; PID controller; minimum variance control; timevariant disturbance; forgetting factor
CHEN Chao-Mian, DIAO Jun, LI Hua-Yin, JIAN Ji-Xin. Control loop performance assessment method with applications in
PTA production equipment. J4, 2010, 44(8): 1460-1465.
［1］ ASTROM K J, HAGGLUND T. PID controllers: theory, design, and tuning ［M］. 2nd ed. New York: ISA 1995: 5970．
［2］ BIALKOWSKI W L. Dreams vs. reality: a view from both sides of the gap ［J］. Pulp and Paper Canada, 1993, 94(11): 1927．
［3］ MITCHELL W, SHOOK D, SHAH S L. A picture worth a thousand control loops: an innovative way of visualizing controller performance data ［C］∥ Control Systems Conference. Quebec City: ［s. n.］, 2004: 147158．
［4］ HAGGLUND T. Industrial implementation of online performance monitoring tools ［J］. Control Engineering Practice, 2005, 13(11): 13831390．
［5］ INGIMUNDARSON A, HAGGLUND T. Closedloop performance monitoring using loop tuning ［J］. Journal of Process Control, 2005, 15(2): 127133．
［6］ LYNCH C B, DUMONT G A. Control loop performance monitoring ［J］. IEEE Transactions on Control Systems Technology, 1996, 4(2): 185192．
［7］ KOZUB D J. Controller performance monitoring and diagnosis: experiences and challenges ［C］∥ Proceedings of the 5th International Conference on Chemical Process Control. Tahoe: AIChE and CACHE, 1996: 8396．
［8］ VISHNUBHOTLA A, SHAH S L, HUANG B A. Feedback and feedforward performance analysis of the Shell industrial closed loop data set ［C］∥ IFAC Symposium Advanced Control of Chemical Processes. Banff Alberta: Pergamon Press Oxford, 1997: 313318．
［9］ THORNHILL N F, OETTINGER M, FEDENCZUK P. Refinerywide control loop performance assessment ［J］. Journal of Process Control, 1999, 9(2): 109124．
［10］ THORNHILL N F, OETTINGER M, FEDENCZUK P. Performance assessment and diagnosis of refinery control loops ［J］. AIChE Symposium Series, 1998, 94(320): 373379．
［11］ YUAN Q L, LENNOX B. Control performance assessment for multivariable systems based on a modified relative variance technique ［J］. Journal of Process Control, 2009, 19(3): 489497．
［12］ XU F W, HUANG B, TAMAYO E C. Performance assessment of MIMO control systems with timevariant disturbance dynamics ［J］. Computers and Chemical Engineering, 2008, 32(9): 21442154．
［13］ QIN S J. Control performance monitoring: a review and assessment ［J］. Computer and Chemical Engineering, 1998, 23(2):173186．
［14］ HARRIS T J, SEPPALA C T. DESBOROUGH L. A review of performance monitoring and assessment techniques for univariate and multivariate control systems ［J］. Journal of Process Control, 1999, 9(1): 117．
［15］ DESBOROUGH L, HARRIS T. Performance and assessment measures for univariate feedback control ［J］. Canadian Journal of Chemical Engineering, 1992, 70(6): 11861197．
［16］ HAN Kai, ZHAO Jun, XU Zhuhua, et al. A closedloop particle swarm optimizer for multivariable process cntrollers design ［J］. Journal of Zhejiang University:Science A, 2008, 9(8): 10501060．
