Abstract.

Nowadays, as far as nonlinear industrial plants control is concerned, in most cases the same set of PIcontroller parameters is used both to follow a setpoint schedule and reject disturbances. This set is usually found to solve the first of the mentioned tasks. Not only does that lead to disturbances rejection time increase, but also it increases energy consumption. A neural tuner is developed in this research in order to adjust PI-controller parameters should step-like limited disturbances acting on a plant output emerge. An additional neural network is added to its structure to solve this task. Tuner structure, rule base and algorithm are also improved. The rule base is used to define both moments when to train the neural network and the learning rate values. Experiments are conducted with the help of heating furnace and rolling mill DC drive mathematical models and a real muffle electroheating furnace. Having analyzed obtained results, a conclusion could be made that the tuner usage allows to achieve 20% time decrease to reject disturbances in comparison with a conventional PI-controller. Moreover, energy consumption is decreased by 13.8% for furnace, and amount of rejected products is decreased by 1.5% for the DC drive.

Keywords:

disturbance, neural tuner, adaptive control, neural network.

PP. 83-94.

References

1. Astrom, K.J., and T. Hagglund. 2006. Advanced PID Control. Research Triangle Park: ISA. 461 p.
2. Mimura, K., and T. Shiotsuki. 2009. Experimental study of PID auto-tuning for unsymmetrical processes. Proceedings of ICCAS-SICE 2009. IEEE. 2967-2971.
3. Stashinov, Ju.P. 2016. K voprosu o nastrojke sistemy upravlenija jelektroprivoda postojannogo toka na modul'nyj optimum. Chast' 1 [On the issue of control system adjustment of a direct current drive on the modular optimum. Part 1]. Jelektrotehnika [Russian Electrical Engineering]. 1: 2-7.
4. Pfeiffer, B.–M. 2000. Towards «plug and control»: self–tuning temperature controller for PLC. International journal of  Adaptive Control and Signal Processing. 14: 519-532.
5. Tjukin, I.Ju., and V.A. Terehov. 2008. Adaptacija v nelinejnyh dinamicheskih sistemah [Adaptation in nonlinear dynamical systems]. Moscow: LKI. 384 p.
6. Li, Y., K. Ang, and C. Chong. 2006. Patents, software, and hardware for PID control – an overview and analysis of the  current art. IEEE Control Systems Magazine. 26(1): 42–54.
7. Aleksandrov, A.G., and M.V. Palenov. 2014. Sostojanie i perspektivy razvitija adaptivnyh PID regulja-torov v tehnicheskih sistemah [Adaptive PID Controllers: State-of-the-Art and Future Developments]. Avtomatika i telemehanika [Automation and remote control]. 2: 16-30.
8. Shubladze, A.M., and S.I. Kuznecov. 2007. Avtomaticheski nastraivajushhiesja promyshlennye PI i PID reguljatory  [Automatically tunable industrial PI and PID controllers]. Avtomatizacija v promyshlennosti [Industrial automation]. 2:15–17.
9. Schaedel, H.M. 1997. A new method of direct PID controller design based on the principle of cascaded damping ratios.  European Control Conference (ECC). IEEE. 1265–1271.
10. Alexandrov, A.G., and M.V. Palenov. 2011. Self–tuning PID–I controller. Proceedings of the 18th IFAC World Congress. FAC. 3635-3640.
11. Rotach, V.Ja., V.F. Kuzishhin, and S.V. Petrov. 2009. Nastrojka reguljatorov po perehodnym harakteristikam sistem  upravlenija bez ih approksimacii analiticheskimi vyrazhenijami [Controllers tuning based on control systems transients analysis  without their  approximation with analytical dependencies]. Avtomatizacija v promyshlennosti [Industrial automation]. 11: 9-12.
12. Allaoua, B., B. Gasbaoui, and B. Mebarki. 2009. Setting up PID DC motor speed control alteration parameters using particle warm optimization strategy. Leonardo Electronic Journal of Practices and Technologies. 14: 19–32.
13. Bindu, R., and M.K. Namboothiripad. 2012. Tuning of PID controller for DC servo motor using genetic algorithm. Emerging Technology and Advanced Engineering. 3(2): 310–314.
14. Erenoglu, I., I. Eksin, E. Yesil, and M. Guzelkaya. 2006. An intelligent hybrid fuzzy PID controller. European Conference on Modelling and Simulation. Bonn: ECMS. 62–67.
15. Kudinov, Ju.I., and A.Ju. Kelina. 2013. Uproshhennyj metod opredelenija parametrov nechetkih PID regu-ljatorov [The  Simplified Method of Definition of Parameters Fuzzy Pid Regulators]. Mehatronika, avtomatizacija, upravlenie [Mechatronics, automation, control]. 1: 12–22.
16. Ahmed, H., and A. Rajoriya. 2014. Performance Assessment of Tuning Methods for PID Controller Parameter used for  Position Control of DC Motor. International Journal of u- and e-Service, Science and Technology. 5(7): 139-150.
17. Anderson, K.L., G.I. Blankenship, and L.G. Lebow. 1988. A rule–based adaptive PID controller. Proc. 27th IEEE Conf.  Decision. Control. IEEE. 564-569.
18. Omatu S., M. Khalid, and R. Yusof. 1995. Neuro–Control and its Applications. London: Springer. – 255 p.
19. Chen, J., and T. Huang. 2004. Applying neural networks to on–line updated PID controllers for nonlinear process control. Journal of Process Control. 14: 211-230.
20. Reyes, J., C. Astorga, M. Adam, and G. Guerrero. 2010. Bounded neuro–control position regulation for a geared DC motor. Engineering Applications of Artificial Intelligence. 23: 1398-1407.
21. Makarov, I.M., and V.M. Lohin. 2001. Intellektual'nye sistemy avtomaticheskogo upravlenija [Intelligent automatic control
systems]. Moscow: FIZMATLIT. 576 p.
22. Eremenko, Y.I., A.I. Glushchenko, and V.A. Petrov. 2016. O nejrosetevoj adaptacii parametrov PI-reguljatora kontura toka sistemy upravlenija prokatnoj klet'ju v real'nom vremeni [Rolling mill current control loop PI-controller parameters adaptation based on neural tuner]. Sistemy upravlenija i informacionnye tehnologii [Control systems and information technologies]. 3(65): 62-68.
23. Eremenko, Y.I., D.A. Poleshchenko, and A.I. Glushchenko. 2015. O primenenii nejrosetevogo optimizatora parametrov  PIreguljatora dlja upravlenija nagrevatel'nymi pechami v razlichnyh rezhimah raboty [About usage of PI-controller parameters neural tuner for control of heating furnaces functioning in different modes]. Upravlenie bol'shimi sistemami [Large scale systems control]. 56: 143-175.
24. Song, Y., J. Guo, and X. Huang. 2016. Smooth Neuroadaptive PI Tracking Control of Nonlinear Systems With Unknown and Nonsmooth Actuation Characteristics. IEEE Transactions on neural networks and learning systems. 99: 1-13.
25. Eremenko, Y.I., and A.I. Glushchenko. 2016. O razrabotke metoda vybora struktury nejronnoj seti dlja reshenija zadachi adaptacii parametrov linejnyh reguljatorov [On development of neural network structure selection method to solve linear controllers parameters adjustment problem]. Upravlenie bol'shimi sistemami [Large scale systems control]. 62: 75-123.
26. Huang, G.B., D.H. Wang, and Y. Lan. 2011. Extreme learning machines: a survey. International Journal of Machine Learning Cybernetics. 2: 107-122.