«Automation and Informatization of the fuel and energy complex»

Synthesis of PID-controller for a separator of HWD type based on principles of hybrid parallel neural control (HPNC)

UDC: 681.51
DOI: -

Authors:

¹ Tyumen Industrial University, Tyumen, Russia
² NOVATEK-TARKOSALENEFTEGAZ, Тarko-Sale, Russia

Keywords: PID controller, hybrid control, neural network technologies, adaptive control, stability, parallel control, automatic regulation, synthesis, dynamic systems

Annotation:

The authors of the article consider the method applied for synthesizing a proportional-integral-differentiative (PID) controller based on the principles of hybrid parallel neural control (HPNC). Traditional PID controllers are widely used in automatic control systems due to their versatility, simplicity and clarity of adjustment. However, their efficiency is significantly limited when working with nonlinear dynamic systems, rapidly changing parameters of the control object and external disturbances. Such limitations are caused by the fixed setting of the controller coefficients, which reduces its adaptability. Under these conditions, to ensure higher control accuracy and improve adaptability to various disturbances, an approach is proposed that combines the classical PID control algorithm with the capabilities of neural network technologies that perform dynamic adjustment of the control action in real time. The mathematical foundations of the method are described in detail, including the formalization of the control object, the structure of the neural network component and learning algorithms. The hybrid controller relative to the classical one is comparatively analyzed as well. The results obtained demonstrate the advantages of the hybrid approach to automatic control problems.

Bibliography:

1. Spasibov V.M., Kabeeva N.V. Tsifrovizatsiya neftegazovogo mestorozhdeniya i kadrovyy potentsial // Neft'. Gaz. Novatsii. – 2018. – № 12. – S. 24–28.
2. Zhilov R.A. Postroyka PID-regulyatora s ispol'zovaniem neyronnykh setey // Izv. Kabardino-Balkarskogo nauch. tsentra RAN. – 2022. – № 5(109). – S. 38–47. – DOI: 10.35330/1991-6639-2022-5-109-38-47
3. Fliess M., Join C. An alternative to proportional-integral and proportional-integral-derivative regulators: Intelligent proportional-derivative regulators // Int. J. of Robust and Nonlinear Control. – 2021. – Vol. 32, Issue 18. – P. 9512–9524. – DOI: 10.1002/rnc.5657
4. Chertilin K.E., Ivchenko V.D. Nastroyka adaptivnykh proportsional'no-integral'no-differentsial'nykh regulyatorov sistemy avtomaticheskogo regulirovaniya chastoty vrashcheniya gazoturbinnogo dvigatelya // Rossiyskiy tekhnolog. zhurn. – 2020. – T. 8, № 6(38). – S. 143–156. – DOI: 10.32362/2500-316X-2020-8-6-143-156
5. Xu Yuanyuan, Chen Bing. Adaptive neural network control for nonlinear non-strict feedback time-delay systems // Systems Science & Control Engineering. – 2020. – Vol. 9, Issue Sup2. – P. 81–92. – DOI: 10.1080/21642583.2020.1833787
6. Metod gibridnogo upravleniya v intellektual'nykh sistemakh na osnove PID i PID-FUZZY-regulyatorov / V.V. Ignat'ev, O.B. Spiridonov, V.M. Kureychik [i dr.] // Vestn. RGRTU. – 2017. – № 62. – S. 110–118. – DOI: 10.21667/1995-4565-2017-62-4-110-118
7. Andriyanov A.M. Standartizatsiya i unifikatsiya kak osnova vnedreniya novykh informatsionnykh tekhnologiy v neftegazovuyu otrasl' // Novye inform. tekhnologii v neftegazovoy otrasli i obrazovanii: materialy XI Mezhdunar. nauch.-prakt. konf.-konkursa, Tyumen', 18–19 apr. 2024 g. – Tyumen': TIU, 2024. – S. 94–98.
8. Performance of Neural Network Based Controllers and ΔΣ-Based PID Controllers for Networked Control Systems: A Comparative Investigation / C. Wanigasekara, D. Almakhles, A. Swain [et al.] // 2019 IEEE Int. Conf. on Environment and Electrical Engineering and 2019 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe), Genova, Italy, June 11–14, 2019. – P. 1–6. – DOI: 10.1109/EEEIC.2019.8783404