نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه مهندسی کامپیوتر، دانشگاه آزاد اسلامی واحد قوچان، ایران.

2 گروه مهندسی کامپیوتر، دانشگاه آزاد اسلامی، واحد نیشابور، ایران.

چکیده

در این مقاله، یک سیستم فازی-عصبی خود‌سازمانده برای یادگیری تطبیقی برخط  برای شناسایی و مدل‌سازی  سیستم‌های دینامیکی غیر‌خطی معرفی شده است. در این سیستم، در ابتدا هیچ نودی در لایه‌ی‌‌ پنهان وجود ندارد و چنان‌چه معیارهای تولید قوانین در طی فرآیند آموزش برآورده شود نرون RBF به لایه‌ی‌‌ پنهان اضافه می‌شود. از الگوریتم آموزش حداقل مربعات بازگشتی وزن‌دار (WRLS) برای قابلیت یادگیری برخطو افزایش سرعت همگرایی،در فاز یادگیری پارامترهای قسمت تالی قوانین نوع تاکاگی سوگنو استفاده شده است. در فاز یادگیری، ساختار برای تولید تعداد قوانین مناسب، معیار جدید درجه‌ی تطبیق و معیار متداول خطا به‌کار گرفته شده است. بعد از ایجاد قانون جدید، کارایی سیستم محاسبه شده و  برای ایجاد شبکه‌ای با ساختار فشرده‌تر قوانینی که تاثیر کم‌تری در کارایی سیستم  دارند با یک الگوریتم هرس جدید هرس می‌شوند. در پایان، برای بهینه‌سازی ساختار توابع عضویت مشابه‌با یکدیگر ترکیب می‌شوند. برای بررسی عملکرد سیستم، دو سیستم دینامیک غیرخطی مبنا، در دو حالت نویزی و بدون نویز در محیط Matlab مدل‌سازی شده‌اند. دقت این مدل‌سازی برمبنای دو معیار تعداد نرون ها (قوانین) و ریشه‌ی‌‌ میانگین مربعات خطا با سایر روش‌ها مقایسه شده است. با‌توجه به نتایج به‌دست‌آمده، میانگین درصد بهبود جواب‌ها در تعداد قوانین به‌دست‌آمده نسبت‌به روش مبنای انتخاب‌شده در مدل‌سازی این دو سیستم در دو حالت نویزی و بدون نویز در مثال اول 42.35% و در مثال دوم 29% می باشد.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Dynamical nonlinear systems modeling and identifying using a self-organized NFS

نویسندگان [English]

  • Hamid Tabatabaee 1
  • Shirin Rikhtegar Mashhad 2

1 Department of Computer Engineering, Islamic Azad University, Quchan, Iran.

2 Department of Computer Engineering, Islamic Azad University, Neyshabur branch, Iran.

چکیده [English]

Nonlinear dynamical systems modeling is one of the real challenges of the real world due to the nonlinear and variable nature of time. In this paper, an Online Self-organizing Takagi-SugenoNeuro-Fuzzy System(OSO-NFS) for dynamic Nonlinear System Identification is proposed. OSO-NFS is built based on radial basis function(RBF). The algorithm has the ability to adaptive adjustment of the system’s parameter and continuous evolution of the system’s structure. Structure identification and parameters estimation are performed simultaneously. The OSO-NFS starts with no hidden neuron. In structural learning, the proposed OSO-NFS uses a two-step algorithm to create a suitable number of rules. A pruning algorithm is used for detecting inactive hidden units and removing them as learning progresses. The weighted recursive least square (WRLS) algorithm is used to adjust all the consequent parameters. Finally, two benchmark examples of nonlinear system identification are demonstrated to show the effectiveness of the proposed method, compared with the other methods. The accuracy of this modeling has been compared with the other methods according to two criteria of the number of neurons (rules) and the root mean square error. According to the results, the average percentage of improvement of the answers in the number of rules obtained in comparison to the chosen method in the modeling of these two systems in both the noise and non-noise modes in the first example is 42.35% and in the second example is 29 %.

کلیدواژه‌ها [English]

  • Nonlinear system identification
  • Online self-organizing learning
  • Takagi-sugeno fuzzy reasoning
  • noise
Rikhtegar Mashhad, Sh; Akbarzadeh Totouchi, M. U (2013). Designing a self-organized neural fuzzy system to identify nonlinear dynamical systems in the presence of noise. 21st Iranian electrical engineering conference. Mashhad, Ferdowsi University of Mashhad.
Wu, S., & Er, M. J. (2000). Dynamic fuzzy neural networks-a novel approach to function approximation. IEEE transactions on systems, man, and cybernetics, part B (cybernetics)30(2), 358-364.
Wu, S., Er, M. J., & Gao, Y. (2001). A fast approach for automatic generation of fuzzy rules by generalized dynamic fuzzy neural networks. IEEE transactions on fuzzy systems9(4), 578-594.
Wang, N., Er, M. J., & Meng, X. (2009). A fast and accurate online self-organizing scheme for parsimonious fuzzy neural networks. Neurocomputing72(16-18), 3818-3829.
Wang, N. (2011). A generalized ellipsoidal basis function based online self-constructing fuzzy neural network. Neural processing letters34(1), 13-37.
de Jesús Rubio, J. (2009). SOFMLS: online self-organizing fuzzy modified least-squares network. IEEE transactions on fuzzy systems17(6), 1296-1309.
Han, H., & Qiao, J. (2010). A self-organizing fuzzy neural network based on a growing-and-pruning algorithm. IEEE transactions on fuzzy systems18(6), 1129-1143.
Kao, C. H., Hsu, C. F., & Don, H. S. (2012). Design of an adaptive self-organizing fuzzy neural network controller for uncertain nonlinear chaotic systems. Neural computing and applications21(6), 1243-1253.
 Hsu, C. F. (2012). Intelligent tracking control of a DC motor driver using self-organizing TSK-type fuzzy neural networks. Nonlinear dynamics67(1), 587-600.
Chen, C. S. (2011). Robust self-organizing neural-fuzzy control with uncertainty observer for MIMO nonlinear systems. IEEE transactions on fuzzy systems19(4), 694-706.
Leng, G., McGinnity, T. M., & Prasad, G. (2006). Design for self-organizing fuzzy neural networks based on genetic algorithms. IEEE transactions on fuzzy systems14(6), 755-766.
Alcalá-Fdez, J., Alcalá, R., Gacto, M. J., & Herrera, F. (2009). Learning the membership function contexts for mining fuzzy association rules by using genetic algorithms. Fuzzy Sets and Systems160(7), 905-921.
Khayat, O., Ebadzadeh, M. M., Shahdoosti, H. R., Rajaei, R., & Khajehnasiri, I. (2009). A novel hybrid algorithm for creating self-organizing fuzzy neural networks. Neurocomputing73(1-3), 517-524.
Chen, C. H., Lin, C. J., & Liao, Y. Y. (2011). A rule-based symbiotic modified differential evolution for self-organizing neuro-fuzzy systems. Proceedings 2011 international conference on system science and engineering (pp. 165-170). IEEE.
Lin, S. F., Chang, J. W., & Hsu, Y. C. (2012). A self-organization mining based hybrid evolution learning for TSK-type fuzzy model design. Applied intelligence36(2), 454-471.
Juang, C. F., Chiu, S. H., & Chang, S. W. (2007). A self-organizing TS-type fuzzy network with support vector learning and its application to classification problems. IEEE transactions on fuzzy systems15(5), 998-1008.
Juang, C. F., & Shiu, S. J. (2008). Using self-organizing fuzzy network with support vector learning for face detection in color images. Neurocomputing71(16-18), 3409-3420.
Dahal, K., Almejalli, K., Hossain, M. A., & Chen, W. (2015). GA-based learning for rule identification in fuzzy neural networks. Applied soft computing35, 605-617.
Nguyen, N. N., Zhou, W. J., & Quek, C. (2015). GSETSK: a generic self-evolving TSK fuzzy neural network with a novel Hebbian-based rule reduction approach. Applied soft computing35, 29-42.
Tavoosi, J., Suratgar, A. A., & Menhaj, M. B. (2016). Nonlinear system identification based on a self-organizing type-2 fuzzy RBFN. Engineering applications of artificial intelligence54, 26-38.
Han, H. G., Lin, Z. L., & Qiao, J. F. (2017). Modeling of nonlinear systems using the self-organizing fuzzy neural network with adaptive gradient algorithm. Neurocomputing266, 566-578.
Leung, J. H., Kuo, Y. L., Weng, T. W., & Chin, C. L. (2017). Hybrid-Neuro-Fuzzy System and Adaboost-Classifier for Classifying Breast Calcification. Journal of Computers28(2), 29-42.
Lin, C. M., & Le, T. L. (2017). PSO-self-organizing interval type-2 fuzzy neural network for antilock braking systems. International journal of fuzzy systems19(5), 1362-1374.
Han, H., Wu, X. L., & Qiao, J. F. (2013). Nonlinear systems modeling based on self-organizing fuzzy-neural-network with adaptive computation algorithm. IEEE transactions on cybernetics44(4), 554-564.
Han, H. G., Guo, Y. N., & Qiao, J. F. (2018). Nonlinear system modeling using a self-organizing recurrent radial basis function neural network. Applied soft computing71, 1105-1116.
Meng, X., Rozycki, P., Qiao, J. F., & Wilamowski, B. M. (2017). Nonlinear system modeling using RBF networks for industrial application. IEEE transactions on industrial informatics14(3), 931-940.
Kumar, R., Srivastava, S., & Gupta, J. R. P. (2018). Online modeling and adaptive control of robotic manipulators using Gaussian radial basis function networks. Neural computing and applications30(1), 223-239.
 Tavoosi, J., & Badamchizadeh, M. A. (2013). A class of type-2 fuzzy neural networks for nonlinear dynamical system identification. Neural computing and applications23(3-4), 707-717.
Kasabov, N., & Song, Q. (2002). DENFIS: dynamic evolving neural-fuzzy inference system and its application for time series prediction. IEEE transactions on fuzzy systems, 10(2).
Leng, G., McGinnity, T. M., & Prasad, G. (2005). An approach for on-line extraction of fuzzy rules using a self-organising fuzzy neural network. Fuzzy sets and systems150(2), 211-243.
Leng, G., McGinnity, T. M., & Prasad, G. (2005). An approach for on-line extraction of fuzzy rules using a self-organising fuzzy neural network. Fuzzy sets and systems150(2), 211-243.
 Jang, J. S. (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE transactions on systems, man, and cybernetics23(3), 665-685.