Authors A. I. Glushchenko
Month, Year 09, 2017 @en
Index UDC 004.89 + 681.5.037
Abstract A neural network tuner allows tuning in real time the P and PI controllers to improve the quality of transient processes for electromechanical objects of control in the process of their func-tioning. A main part of this tuner is a neural network. It is learned online with the help of the backpropagation method. A learning rate value for such learning is calculated by a rule base con-taining knowledge of a process automation engineer. This value may be too high for current situation. It might result in too quick controller parameters change made by the network. In its turn, this can lead to instability of the control system under consideration. Trying to avoid such situation, we propose to estimate sustainability of the control system with the tuner using Lyapunov functions. In contrast to the existing methods, our approach does not require the plant model and functions online. This allows limiting the current learning rate for the neural network. Proposed method is applied to adjust speed P-controller parameter of a two-high reverse rolling mill model. Experiments, which main aim is to follow the speed setpoint graph, are conducted under the following conditions. The rolling mill model is used with its nominal parameters values, whereas the speed P-controller is non optimally tuned. The tuner task is to adjust the speed P-controller pa-rameter back to the optimal value calculated in accordance with the technical optimum require-ments. Obtained results allow concluding that proposed sustainability estimation method is ade-quate, since stability sufficiency condition has not been broken during experiments.

Download PDF

Keywords Adaptive control; sustainability; Lyapunov second method; neural tuner; DC drive; speed control loop.
References 1. Astrom K.J., Wittenmark B. Adaptive Control. New York: Dower Publications, 2008, 573 p.
2. Aleksandrov A.G., Palenov M.V. Sostoyanie i perspektivy razvitiya adaptivnykh PID-regulyatorov v tekhnicheskikh sistemakh [Status and prospects of the development of adaptive PID controllers in technical systems], Avtomatika i telemekhanika [Automation and Remote Control], 2014, No.. 2, pp. 16-30.
3. Novikov N.I., Novikova G.V. Toplivo-energeticheskaya sostavlyayushchaya chernoy metallurgii: problemy i tendentsii [The fuel energy component of ferrous metallurgy: problems and trends], Vestnik Kemerovskogo gosudarstvennogo universiteta [Bulletin of Kemerovo State University], 2013, Vol. 1, No. 4 (56), pp. 257-263.
4. Rotach V.Ya. Teoriya avtomaticheskogo upravleniya [Theory of automatic control.]. 5 ed.
– Moscow: Izd-vo MEI, 2008, 396 p.
5. Astrom K.J., Hagglund T. Advanced PID Control. Research Triangle Park: ISA – The Instrumentation, Systems, and Automation Society, 2006, 461 p.
6. Li Y., Ang K., Chong C. Patents, software, and hardware for PID control – an overview and analysis of the current art, IEEE Control Systems Magazine, 2006, No. 26 (1), pp. 42-54.
7. Calvo-Rolle J.L. et al. A hybrid intelligent system for PID controller using in a steel rolling process, Expert Systems with Applications, 2013, Vol. 40, No. 13, pp. 5188-5196.
8. Kudinov Yu.I. i dr. Postroenie i nastroyka nechetkogo adaptivnogo PID-regulyatora [The construction and tuning of a fuzzy adaptive PID controller], Informatika i sistemy upravleniya [Informatics and control systems], 2016, No. 3 (49), pp. 86-96.
9. Erenoglu I., Eksin I., Yesil E. and etc. An intelligent hybrid fuzzy PID controller. European Conference on Modelling and Simulation. Bonn: European Council for Modelling and Simulation, 2006, pp. 62-67.
10. Chen J., Huang T. Applying neural networks to on-line updated PID controllers for nonlinear process control, Journal of Process Control, 2004, No. 14, pp. 211-230.
11. Song Y., Guo J., Huang X. Smooth Neuroadaptive PI Tracking Control of Nonlinear Systems with Unknown and Nonsmooth Actuation Characteristics, IEEE Transactions on neural net-works and learning systems, 2016, Vol. 99, pp. 1-13.
12. Eremenko Yu.I., Poleshchenko D.A., Glushchenko A.I. O primenenii neyrosetevogo optimizatora parametrov PI-regulyatora dlya upravleniya nagrevatel'nymi pechami v razlichnykh rezhimakh raboty [On the application of neural network optimizer parameters
13. PI-controller for heating furnaces in different operation modes], Upravlenie bol'shimi sistemami [Managing large systems], 2015, Issue 56, pp. 143-175.
14. Eremenko Yu.I., Glushchenko A.I. O razrabotke metoda vybora struktury neyronnoy seti dlya resheniya zadachi adaptatsii parametrov lineynykh regulyatorov [On the development of a selection method of a neural network structure for solving the problem of adapting the parameters of the linear regulators], Upravlenie bol'shimi sistemami [Managing large systems], 2016, Issue 62, pp. 75-123.
15. Lyapunov A.M. The general problem of stability of motion, International journal of control, 1992, Vol. 55, pp. 531-534.
16. Makarov I.M., Lokhin V.M., Man'ko S.V., Romanov M.P., Sitnikov M.S. Ustoychivost' intellektual'nykh sistem avtomaticheskogo upravleniya. Prilozhenie k zhurnalu [The stability of the intelligent systems of automatic control], Informatsionnye tekhnologii [Information technology], 2013, No. 2, 32 p.
17. Cong S., Liang Y. PID-like neural network nonlinear adaptive control for uncertain multivariable motion control systems, IEEE Transactions on Industrial Electronics, 2009, Vol. 56,
No. 10, pp. 3872-3879.
18. Kang J. et al. An adaptive PID neural network for complex nonlinear system control, Neurocomputing, 2014, Vol. 135, pp. 79-85.
19. Stashinov Yu.P. K voprosu o nastroyke sistemy upravleniya elektroprivoda postoyannogo toka na modul'nyy optimum [To the question about the configuration of the control system of DC electric drive on the modular optimum], Elektrotekhnika [Electrical engineering], 2016, No. 1, pp. 2-7.
20. Eremenko Yu.I., Glushchenko A.I., Petrov V.A. Ob ispol'zovanii neyrosetevogo nastroyshchika dlya adaptatsii P-regulyatora skorosti elektroprivoda prokatnoy kleti [On the use of neural network Adjuster to adapt N-controller speed of the electric drive of rolling mill], Mekhatronika, avtomatizatsiya, upravlenie [Mechatronics, automation, control], 2017, No. 10 (18), pp. 685-692.
21. Huang G.B., Wang D.H., Lan Y. Extreme learning machines: a survey, International Journal of Machine Learning Cybernetics, 2011, No. 2, pp. 107-122.

Comments are closed.