Comparative study of neural networks used in modeling and control of dynamic systems

• Autorzy: Tiliouine H.

Warianty tytułu

PL

Studium porównawcze sieci neuronowych używanych w modelowaniu i sterowaniu systemami dynamicznymi

• Języki publikacji: EN

Abstrakty

EN

In this paper, a diagonal recurrent neural network that contains two recurrent weights in the hidden layer is proposed for the designing of a synchronous generator control system. To demonstrate the superiority of the proposed neural network, a comparative study of performances, with two other neural networks, has been made. The investigated neural networks were: a feedforward neural network (FFNN), a first-order diagonal recurrent neural network (1_DRNN) and the proposed second-order diagonal recurrent neural network (2_DRNN). Moreover, to confirm the superiority of the proposed 2_DRNN, the results obtained with this network were also compared with those of the IEEE recommended conventional excitation control system (CECS).

PL

W artykule przedstawiono diagonalną rekurencyjną sieć neuronową do projektowania układu regulacji generatora synchronicznego. Proponowana sieć posiada dwie rekurencyjne wagi w warstwie ukrytej. Aby wykazać wyższość proponowanej sieci dokonano analizy porównawczej efektywności z dwoma innymi sieciami neuronowymi. Badanymi sieciami neuronowymi były: jednokierunkowa sieć neuronowa (FFNN), diagonalna rekurencyjna sieć neuronowa pierwszego rzędu (1_DRNN) oraz proponowana diagonalna rekurencyjna sieć neuronowa drugiego rzędu (2_DRNN). Ponadto, aby potwierdzić wyższość proponowanej sieci (2_DRNN), uzyskane wyniki dla tej sieci porównano z wynikami uzyskanymi dla konwencjonalnego układu regulacji (CECS) zalecanego przez IEEE.

Słowa kluczowe

EN

feedforward neural network; diagonal recurrent neural network; neurocontrol; turbogenerator; electrical technology

PL

sieci neuronowe; modelowanie; elektrotechnika

• Wydawca: Wydawnictwo SIGMA-NOT
• Czasopismo: Przegląd Elektrotechniczny
• Rocznik: 2011
• Tom: R. 87, nr 7
• Strony: 104--109
• Opis fizyczny: Bibliogr. 17 poz., rys., tab., wykr.

Twórcy

autor

Tiliouine H.

• Gdansk Technical Uniwersity Faculty of Electrical and Control Engineering, ul. Narutowicza 11/12, 80-233 Gdańsk,

Bibliografia

• [1] Mohammadzaheri M., Member L.C.: A design approach to feedback-feedforward control systems, American Contro Conference (ACC), 2010.
• [2] Ting-Li C., Chung-Cheng C., Yi-Chieh H., Wen-Jiun L.: Stabilit) and almost disturbance decoupling analysis of nonlineat system subject to feedback linearization and feedforwarc neural network controller, Neural Networks, IEEI Transactions on, volume 19, 2008.
• [3] Min H., Wei G., Jincheng W.: Predictive control based or feedforward neural network for strong nonlinear system Neural Networks, 2005 IJCNN'05, Proceedings, 2005 IEEI Internationa Joint Conference, Volume 4, 2005, 2266-2271.
• [4] Parisini T., Zoppoli R.: Neural networks for feedbact feedforward nonlinear control systems, Neural Networks, IEEt Transactions on, Volume 5, issue 3, 1994,436^449.
• [5] Kuschewski, J.G., Hui, S., Żak S.H.: Application of feedforwarc neural networks to dynamical system identification anc control, Control Systems Technology, IEEE Transactions on Volume 1, Issue 1, 1993, 37-49.
• [6] Thomas R.J., Sakk E.: On using an artificial feedforward neural network as a controller in large -scale power systems, Circuits and Systems, IEEE International Symposium on, vol. 2, 1991. 1133-1136.
• [7] Tiliouine H.: A modified neural network controller configuration 14th International Conference on Methods and Models r Automation and Robotics, Międzyzdroje, Poland, Volume 14 part l, August 2009.
• [8] Sumina, D., Bulić, N., Erceg, G.: Simulation model of neural network based synchronous generator excitation control, Power Electronics and Motion Control Conference, 2008. EPE-PEMC 2008. 13th, 556-560.
• [9] Venayagamoorthy G.K., Harley R.G.: A continually online trained neuro-controller for excitation and turbinę control of a turbogenerator, IEEE Transactions on Energy Conversion, 16 (16), 2001, 261-269.
• [10] Salem M.M, Żaki A.M., Mahgoub O.A., Abu EI_Zahab E., Malik O.P.: Studies on a multi-machine power system with a neural network based excitation controller, Power Engineering Society Summer Meeting, 2000 IEEE, Volume 1, 105 - 110.
• [11] Lubośny Z.: Adaptacyjny neuronowy regulator generator synchronicznego, Przegląd Elektrotechniki R.80, nr 10, 2004, 971-974.
• [12] Chun-Jung C., Tien-Chi C., Hung-Jung H. and Chin-Chih O.: PSS design using adaptive recurrent neural network controller, Fifth International Conference on Natural Computation, Volume 2, 2009, 277-281.
• [13] Yingyi J., Chengli S.: Adaptive model predictiye control using diagonal recurrent neural network, Natural Computation, ICNC '08. Fourth International Conference on, Volume 2, 2008, 276-280.
• [14] Xiaochen H., Yen, G.G.: Local signal based supplementary excitation controller for damping inter-area oscillations through recurrent neural networks, Neural Networks, IJCNN 2007. International Joint Conference on, 2007, 2498-2503.
• [15] Jeen-Shing W., Yen-Ping C.: A fully automated recurrent neural network for unknown dynamie system Identification and control. Circuits and Systems I: Regular Papers, IEEE Transactions on, Volume 53, 2006, 1363-1372.
• [16] Chi-Huang L., Ching-Chih T.: Design and experimental evaluation of an adaptive predictive controller using recurrent neural network, Systems, Mań and Cybernetics, 2005 IEEE International Conference on, Volume 1, 2005, 690-695.
• [17] IEEE Recommended practice for excitation system models for power system stability studies, IEEE Std 421.5-2005 (Revision of IEEE Std 421.5-1992), 2006, 1-85.