Research on neural adaptive PID control based on RFNN identifier applied to stable liquid level control
Abstract
In industries such as paper, energy, chemical processing, water treatment, and oil refining, achieving extremely high accuracy in liquid level and flow control is essential for ensuring process stability and improving production efficiency. Consequently, a key challenge lies in precisely regulating water flow to maintain stable liquid levels. In this paper, a single neural adaptive PID controller based on a recurrent fuzzy neural network identifier (SNA-PID-RFNNI) is proposed and compared with a fuzzy PID controller (FUZZY-PID) and a conventional PID controller. The controllers are designed and evaluated through simulations in MATLAB/Simulink and experiments on an interacting two-tank water system model. Simulation results show that the SNA-PID-RFNNI achieves a rise time of 0.389 s, an overshoot of 1.68%, and a steady-state error of 0.05 cm. Experimental results further demonstrate that the SNA-PID-RFNNI achieves a settling time of approximately 345 s with only 0.1 cm steady-state error, while maintaining stable responses under disturbance conditions.
Tóm tắt
Trong các ngành công nghiệp giấy, năng lượng, hóa chất, xử lý nước, lọc dầu... Để giải quyết vấn đề ổn định mực nước và lưu lượng nước, cần có độ chính xác cực cao. Quy trình sản xuất đang ngày càng hiệu quả hơn. Do đó, thách thức đặt ra là phải điều chỉnh chính xác lưu lượng nước để ổn định mức chất lỏng. Trong bài báo này bộ điều khiển PID thích nghi đơn nơ-ron dựa trên bộ nhận dạng mạng nơ-ron mờ hồi quy (SNA-PIDRFNNI) được đề xuất và so sánh với bộ điều khiển mờ PID (FUZZY-PID) và PID cổ điển. Các bộ điều khiển được thiết kế và đánh giá thông qua mô phỏng trên MATLAB/ Simulink và thực nghiệm trên mô hình hai bồn nước đôi tương tác. Kết quả mô phỏng cho thấy bộ SNA-PID-RFNNI có thời gian tăng 0,389 giây, độ vọt lố 1,68% và sai số xác lập 0,05 cm. Kết quả thực nghiệm SNA-PID-RFNNI đạt thời gian xác lập khoảng 345 giây với sai số chỉ 0,1 cm và duy trì đáp ứng định trong điều kiện có nhiễu
References
[1]. P. K. Choudhary, P. Raj, and D. K. Das, “Controller Design for Decoupled TwoInput Two-Output Coupled Tank System,” 2024 IEEE International Conference on
Smart Power Control and Renewable Energy (ICSPCRE), pp. 1–6, Jul. 2024, doi:
10.1109/ICSPCRE62303.2024.10675217.
[2]. S. Manna, D. K. Singh, Y. Y. Ghadi, A. Yousef, H. Kotb, and K. M. AboRas,
“Probabilistic Bi-Level Assessment and Adaptive Control Mechanism for Two-Tank
Interacting System,” IEEE Access, vol. 11, pp. 118268–118280, 2023, doi: 10.1109/
ACCESS.2023.3326727.
[3]. Miral Changela and Ankit Kumar, “Designing a Controller for Two Tank
Interacting System,” International Journal of Science and Research, vol. 4, no. 5, pp.
2319–7064, 2015, doi: 10.21275/SUB154198.
[4]. D. Mukherjee, P. K. Kundu, and A. Ghosh, “PID Controller Design for An Interacting
Tank Level Process with Time Delay Using MATLAB FOMCON Toolbox,” 2016
2nd International Conference on Control, Instrumentation, Energy & Communication
(CIEC), Kolkata, India, pp. 1–5, 2016, doi: 10.1109/CIEC.2016.7513803.
[5]. L. Wang and H. Wang, “Study on the Optimization for Reactive Power Regulation
of Synchronous Condenser Based on Single Neuron Adaptive PID,” Chinese
Journal of Electrical Engineering, vol. 11, no. 1, pp. 184–193, 2025, doi: 10.23919/
CJEE.2025.000109.
[6]. K. Jia, S. Lin, Y. Du, C. Zou, and M. Lu, “Research on Route Tracking Controller
of Quadrotor UAV Based on Fuzzy Logic and RBF Neural Network,” IEEE Access,
vol. 11, pp. 111433–111447, 2023, doi: 10.1109/ACCESS.2023.3322944.
[7]. F. Ma, “An Improved Fuzzy PID Control Algorithm Applied in Liquid Mixing
System,” 2014 IEEE International Conference on Information and Automation
(ICIA), Hailar, China, pp. 587–591, 2014, doi: 10.1109/ICInfA.2014.6932722.
[8]. Wen-Shyong Yu, “Adaptive Fuzzy PID Control for Nonlinear Systems with H
∞ Tracking Performance,” 2006 IEEE International Conference on Fuzzy Systems,
Vancouver, BC, Canada, pp. 1010–1015, 2006, doi: 10.1109/FUZZY.2006.1681834.
[9]. G. Jie and T. Shengjing, “Application of parameter self-tuning fuzzy pid controller
in guidance loop of unmanned aircraft,” in 6th International Conference on Fuzzy
Systems and Knowledge Discovery, 2009, pp. 57–61. doi: 10.1109/FSKD.2009.374.
[10]. J. Z. Shi, “A Fractional Order General Type-2 Fuzzy PID Controller Design
Algorithm,” IEEE Access, vol. 8, pp. 52151–52172, 2020, doi: 10.1109/
ACCESS.2020.2980686.
[11]. Y. Y. Lin, J. Y. Chang, and C. T. Lin, “Identification and prediction of dynamic
systems using an interactively recurrent self-evolving fuzzy neural network,” IEEE
Trans Neural Netw Learn Syst, vol. 24, no. 2, pp. 310–321, 2013, doi: 10.1109/
TNNLS.2012.2231436[12]. T. Liu, Y. Chen, Z. Chen, and H. Wu and L. Cheng, “Adaptive fuzzy fractional
order PID control for 6-DOF quadrotor,” 2020 39th Chinese Control Conference
(CCC), pp. 2158–2163, 2020, doi: 10.23919/CCC50068.2020.9188677.
[13]. R. Tipsuwanpom, T. Runghimmawan, S. Intajag, and V. Krongratana, “Fuzzy
Logic PID controller based on FPGA for process control,” 2004 IEEE International
Symposium on Industrial Electronics, Ajaccio, France, vol. 2, pp. 1495–1500, 2004,
doi: 10.1109/ISIE.2004.1572035.
[14]. A. Zribi, M. Chtourou, and M. Djemel, “A New PID Neural Network Controller
Design for Nonlinear Processes,” Journal of Circuits, Systems, and Computers, vol.
27, no. 10, 2018, doi: DOI:10.1142/S0218126618500652.
[15]. S. Kamalasadan and A. A. Ghandakly, “A Neural Network Parallel Adaptive
Controller for Dynamic System Control,” in IEEE Transactions on Instrumentation
and Measurement, vol. 56, no. 5, pp. 1786–1796, Oct. 2007, doi: 10.1109/
TIM.2007.895674.
[16]. C. Yildirim and G. Böcekçi, “Neuro-Based Adaptive PID Controller for Marine
Satellite Tracking Systems,” IEEE Access, vol. 13, pp. 60424–60439, 2025, doi:
10.1109/ACCESS.2025.3558032.
[17]. Jinkun Liu, “Radial Basis Function (RBF) Neural Network Control for Mechanical
Systems,” Tsinghua University Press; Springer-Verlag Berlin Heidelberg, 2013, pp.
3–4. doi: 10.1007/978-3-642-34816-7.
[18]. L. M. Thanh, L. H. Thuong, P. T. Loc, and C. N. Nguyen, “Delta Robot Control
Using Single Neuron PID Algorithms Based on Recurrent Fuzzy Neural Network
Identifiers,” International Journal of Mechanical Engineering and Robotics Research,
vol. 9, no. 10, pp. 1411–1418, Oct. 2020, doi: 10.18178/ijmerr.9.10.1411-1418.
[19]. X. W. and M. L. Ming-guang Zhang, “Adaptive PID Control Based on RBF
Neural Network Identification,” 17th IEEE International Conference on Tools with
Artificial Intelligence (ICTAI’05), Hong Kong, China, pp. 3–683, 2005, doi: 10.1109/
ICTAI.2005.26.
