Porównanie różnych sposobów optymalizacji nastaw regulacji procesów przemysłowych z uwzględnieniem wpływu wskaźników oceny ich jakości

pol Artykuł w języku polskim DOI: 10.14313/PAR_233/27

Konrad Bogusz, Bartosz Rajkowski, wyślij Paweł D. Domański Politechnika Warszawska, Instytut Automatyki i Informatyki Stosowanej

Pobierz Artykuł

Streszczenie

Jakość regulacji jest kluczowym zagadnieniem w nowoczesnym przemyśle. Zadaniem inżyniera jest nie tylko dobór parametrów regulacji, ale również bieżące nadzorowanie jej jakości tak, aby maksymalizować wydajność procesu a także dbać o stan aparatury i urządzeń wykonawczych. W zdecydowanej większości rozwiązań praktycznych stosowany jest kwadratowy wskaźnik jakości, zarówno w procesie strojenia jak i oceny jakości regulacji. W artykule zostały zaproponowane inne wskaźniki, cechujące się większą odpornością. Zostały one zastosowane do dwóch zadań: projektowania nastaw regulatora oraz oceny jakości już pracującej struktury sterowania. W pracy uwzględnione zostały rozwiązania dla różnych wersji podstawowego algorytmu regulacji w przemyśle procesowym, tj. PID. Różne strategie regulacji PID zostały również porównane z algorytmem sterowania predykcyjnego typu MPC. Analiza symulacyjna wykorzystuje przemysłowy benchmark układu sterowania systemem chłodniczym wykorzystującym zjawisko kompresji pary.

Słowa kluczowe

analiza R/S, MPC, ocena jakości sterowania, pid, wykładnik Hursta

Comparison of Various Controller Parameters Optimization Strategies for Industrial Processes Using Different Control Performance Assessment Indexes

Abstract

Control quality is a crucial issue in modern industry. The engineer’s goal is to set control parameters and assess it constantly in order to maximize the efficiency of the process and watch the condition of actuators. The article describes a number of Control Performance Assessment indexes. They are applied to two tasks: controller tuning and the assessment of the quality of an already tuned control strategy. It presents a comparison between different indexes applied to various structures of the PID control algorithm. Finally the results are compared with Model Predictive Control. The analysis uses well known nonlinear industrial benchmark of a refrigeration system based on vapour compression.

Keywords

Control Performance Assessment, Hurst exponent, MPC, pid, R/S analysis

Bibliografia

  1. Ordys A., Uduehi D., Johnson M.A., Process Control Performance Assessment – From Theory to Implementation. Springer-Verlag, London 2007.
  2. Bauer M., Horch A., Xie L., Jelali M., Thornhill N., The current state of control loop performance monitoring – a survey of application in industry. “Journal of Process Control”, Vol. 38, 2016, 1–10, DOI: 10.1016/j.jprocont.2015.11.002.
  3. Domański P.D., Golonka S., Jankowski R., Kalbarczyk P., Moszowski B., Control rehabilitation impact on production neffciency of ammonia synthesis installation . “Industrial & Engineering Chemistry Research”, Vol. 55, No. 39, 2016, 10366–10376, DOI: 10.1021/acs.iecr.6b02907.
  4. Häggblom K.E., Process control laboratory. [www.users.abo.fi/khaggblo/PDC/PDC7.pdf] 2017. Accessed: 2019-05-05.
  5. Ziegler J.G., Nichols N.B., Optimum settings for automatic controllers. “Transactions ASME”, Vol. 64, 759–768, 1942.
  6. Cohen G.H., Coon G.A., Theoretical consideration of retarded control. “Transactions ASME”, Vol. 75, 827, 1953.
  7. Åström K.J., Hägglund T., Automatic tuning of simple regulators with specifications on phase and amplitude margins. “Automatica”, Vol. 20, No. 5, 645–651, 1984, DOI: 10.1016/0005-1098(84)90014-1.
  8. McMillan G.K., Good tuning: A pocket guide. ISA, 2 edition, 2006.
  9. Visioli A., Practical PID control. Springer Science & Business Media, 2006.
  10. O'Dwyer A., Handbook of PI and PID controller tuning rules. Imperial College Press, 2003.
  11. Seborg D.E., Edgar T.F., Mellichamp D.A., Doyle F.J., Process Dynamics and Control, 4th ed. Wiley, 2016.
  12. Rousseeuw P.J., Robust regression and outlier detection. Wiley Series in Probability and Mathematical Statistics. John Wiley, New York 1987.
  13. Huber P.J., Ronchetti E.M., Robust Statistics. Wiley, Hoboken, 2009.
  14. Verboven S., Hubert M., LIBRA: A MATLAB library for robust analysis. “Chemometrics and Intelligent Laboratory Systems”, Vol. 75, No. 2, 127–136, 2005, DOI: 10.1016/j.chemolab.2004.06.003.
  15. Smuts J.F., Hussey A., Requirements for successfully implementing and sustaining advanced control applications. Proceedings of the 54th ISA POWID Symposium, 89–105, 2011.
  16. Jelali M., Control performance management in industrial automation: assessment, diagnosis and improvement of control loop performance. Springer-Verlag, London 2013.
  17. Domański P.D., Non-Gaussian properties of the real industrial control error in SISO loops. Proceedings of the 19th International Conference on System Theory, Control and Computing, 877–882, 2015, DOI: 10.1109/ICSTCC.2015.7321405.
  18. Tatjewski P., Sterowanie zaawansowane obiektów przemysłowych. IBUK Libra, 2016.
  19. MathWorks. MATLAB Documentation. The MathWorks Inc., 1994–2018.
  20. Cajo R., Zhao S., Ionescu C.M., De Keyser R., Plaza D., Liu S., IMC based PID control applied to the benchmark PID18. “IFAC-PapersOnLine”, Vol. 51, No. 4, 2018, 728–732, DOI: 10.1016/j.ifacol.2018.06.210.
  21. Spinner T., Srinivasan B., Rengaswamy R., Data-based automated diagnosis and iterative retuning of proportional-integral (PI) controllers. “Control Engineering Practice”, Vol. 29, 23–41, 2014, DOI: 10.1016/j.conengprac.2014.03.005.
  22. Hägglund T., Automatic detection of sluggish control loops. “Control Engineering Practice”, Vol. 7, No. 12, 1505–1511, 1999, DOI: 10.1016/S0967-0661(99)00116-1.
  23. Visioli A., Method for proportional-integral controller tuning assessment. “Industrial & Engineering Chemistry Research”, Vol. 45, No. 8, 2741–2747, 2006, DOI: 10.1021/ie0508482.
  24. Salsbury T.I., A practical method for assessing the performance of control loops subject to random load changes. “Journal of Process Control”, Vol. 15, No. 4, 393–405, 2005, DOI: 10.1016/j.jprocont.2004.08.004.
  25. Zhao Y.M., Xie W.F., Tu X.W., Performance-based parameter tuning method of model-driven PID control systems. “ISA Transactions”, Vol. 51, No. 3, 393–399, 2012, DOI: 10.1016/j.isatra.2012.02.005.
  26. Zheng B., Analysis and auto-tuning of supply air temperature PI control in hot water heating systems. PhD thesis, Dissertation of University of Nebraska, 2007.
  27. Yu Z., Wang J., Performance assessment of static leadlag feedforward controllers for disturbance rejection in PID control loops. “ISA Transactions”, Vol. 64, 67–76, 2016, DOI: 10.1016/j.isatra.2016.04.016.
  28. Horch A., A simple method for detection of stiction in control valves. “Control Engineering Practice”, Vol. 7, No. 10, 1221–1231, 1999, DOI: 10.1016/S0967-0661(99)00100-8.
  29. Howard R., Cooper D., A novel pattern-based approach for diagnostic controller performance monitoring. “Control Engineering Practice”, Vol. 18, No. 3, 279–288, 2010, DOI: 10.1016/j.conengprac.2009.11.005.
  30. Choudhury M.A.A.S., Shah S.L., Thornhill N.F., Diagnosis of poor control-loop performance using higher-order statistics. “Automatica”, Vol. 40, No. 10, 1719–1728, 2004, DOI: 10.1016/j.automatica.2004.03.022.
  31. Li Y., O’Neill Z., Evaluating control performance on building HVAC controllers. International Building Performance Simulation Association, 962–967, Hyderabad, India, 2015.
  32. Domański P.D., Statistical measures for proportional–integral–derivative control quality: Simulations and industrial data. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering,  232(4):428–441, 2018.
  33. Harris T.J., Assessment of closed loop performance. “Canadian Journal of Chemical Engineering”, Vol. 67, No. 5, 856–861, 1989, DOI: 10.1002/cjce.5450670519.
  34. Meng Q.W., Gu J.Q., Zhong Z.F., Ch S., Niu Y.G., Control performance assessment and improvement with a new performance index. 25th Chinese Control and Decision Conference (CCDC), 4081–4084, 2013.
  35. Afshar Khamseh S., Khaki Sedigh A., Moshiri B., Fatehi A., Control performance assessment based on sensor fusion techniques. “Control Engineering Practice”, Vol. 49, 14–28, 2016, DOI: 10.1016/j.conengprac.2016.01.008.
  36. Salsbury T.I., Alcala C.F., Two new normalized EWMAbased indices for control loop performance assessment. 2015 American Control Conference (ACC), 962–967, 2015, DOI: 10.1109/ACC.2015.7170858.
  37. Shardt Y., Zhao Y., Qi F., Lee K., Yu X., Huang B., Shah S., Determining the state of a process control system: Current trends and future challenges. “The Canadian Journal of Chemical Engineering”, Vol. 90, No. 2, 217–245, 2012, DOI: 10.1002/cjce.20653.
  38. Schlegel M., Skarda R., Cech M., Running discrete Fourier transform and its applications in control loop performance assessment. 2013 International Conference on Process Control (PC), 113–118, 2013, DOI: 10.1109/PC.2013.6581393.
  39. Nesic Z., Dumont G.A., Davies M.S., Brewster D., CD control diagnostics using a wavelet toolbox. Proceedings CD Symposium, IMEKO, Vol. XB, 120–125, 1997.
  40. Lynch C.B., Dumont G.A., Control loop performance monitoring. “IEEE Transactions on Control Systems Technology”, Vol. 4, No. 2, 185–192, 1996.
  41. Zhou Y.F., Wan F., A neural network approach to control performance assessment. “International Journal of Intelligent Computing and Cybernetics”, Vol. 1, No. 4, 617–633, 2008, DOI: 10.1108/17563780810919159.
  42. Pillay N., Govender P., A data driven approach to performance assessment of PID controllers for setpoint tracking. “Procedia Engineering”, Vol. 69, 1130–1137, 2014, DOI: 10.1016/j.proeng.2014.03.101.
  43. Domański P.D., Non-Gaussian and persistence measures for control loop quality assessment. “Chaos”, 322(043105), 2016, DOI: 10.1063/1.4946772.
  44. Zhang J., Jiang M., Chen J., Minimum entropy-based performance assessment of feedback control loops subjected to non-Gaussian disturbances. “Journal of Process Control”, Vol. 24, No. 11, 1660–1670, 2015, DOI: 10.1016/j.jprocont.2014.08.003.
  45. Knierim-Dietz N., Hanel L., Lehner J., Definition and verification of the control loop performance for different power plant types. Technical report, Institute of Combustion and Power Plant Technology, University of Stuttgart, 2012.
  46. Borak S., Misiorek A., Weron R., Models for heavy-tailed asset returns. [In:] Cizek P., Härdle, W. K., Weron R. (eds), Statistical tools for finance and insurance, 21–56. Springer, New York, 2nd edition, 2011.
  47. Koutrouvelis I.A., Regression-type estimation of the parameters of stable laws. “Journal of the American Statistical Association”, Vol. 75, No. 372, 918–928, 1980, DOI: 10.2307/2287182.
  48. Peters E.E., Teoria chaosu a rynki kapitałowe. WIG Press, Warszawa, 1997.
  49. Spinner T., Srinivasan B., Rengaswamy R., Data-based automated diagnosis and iterative retuning of proportional-integral (PI) controllers. “Control Engineering Practice”, Vol. 29, 23–41, 2014, DOI: 10.1016/j.conengprac.2014.03.005.
  50. BENCHMARK PID 2018, Benchmark for PID control of refrigeration systems based on vapour compression (version: july 2017). [http://servidor.dia.uned.es/~fmorilla/benchmarkPID2018]. Accessed: 2019-05-05.
  51. Bejarano G., Alfaya J.A., Rodriguez D., Morilla F., Ortega M.G., Benchmark for PID control of refrigeration systems based on vapour compression. “IFAC-PapersOnLine”, Vol. 51, No. 4, 497–502, 2018, DOI: 10.1016/j.ifacol.2018.06.144.
  52. Pudlik W., Termodynamika, wydanie III – cyfrowe. [pbc.gda.pl/Content/4098/pbc_termodynamika.pdf]. Accessed: 2019-05-05.