IJE TRANSACTIONS B: Applications Vol. 27, No. 8 (August 2014) 1185-1194   

downloaded Downloaded: 358   viewed Viewed: 2577

H. A. Bazoobandi and M. Eftekhari
( Received: October 27, 2013 – Accepted: April 17, 2014 )

Abstract    Abstract Many parameter-tuning algorithms have been proposed for training Fuzzy Wavelet Neural Networks (FWNNs). Absence of appropriate structure, convergence to local optima and low speed in learning algorithms are deficiencies of FWNNs in previous studies. In this paper, a Memetic Algorithm (MA) is introduced to train FWNN for addressing aforementioned learning lacks. Differential Evolution (DE) is utilized as the global search. The main contributions of this paper are: (i) Proposing a new fast and effective local search based on spatial distribution (that is named Spatial Distribution Local Search (SDLS)), SDLS can adjust the step size of parameters adaptively toward obtaining the better ones. (ii) Introducing an adaptive selection method to select appropriate individuals from current population for local refinement in MA. (iii) Improving the selection operator in standard DE by an adaptive strategy. In this strategy, worse offspring has a chance to be replaced with its parent to prevent trapping in local optima and controlling the selection pressure. The proposed MA is compared with several training algorithms of FWNNs over some benchmark problems. Experimental results obtained, confirm the effectiveness of the proposed MA for improving the convergence rate and modeling accuracy in comparison to the other training methods.


Keywords    Fuzzy Wavelet Neural Network (FWNN), Memetic Algorithm, Differential Evolution, Spatial Distribution Local Search, Adaptive Selection Strategy


چکیده    چكيده الگوريتم هاي بسياري تا کنون براي تنظيم پارامترهاي شبکه هاي عصبي فازي موجک معرفي شده اند. نبود يک ساختار مناسب، همگرايي به بهينه هاي محلي و سرعت پائين را مي توان از مهم ترين اشکالات مطالعات گذشته در مورد شبکه هاي عصبي فازي موجک دانست. در اين مقاله يک الگوريتم ممتيک براي رفع اين اشکالات پيشنهاد شده است. در روش پيشنهادي الگوريتم تکامل تفاضلي به عنوان جستجو کننده سراسري استفاده شده است. مهم ترين نوآوري هاي اين مقاله عبارتند از: 1) در اين مقاله يک الگوريتم جستجوي محلي براساس توزيع فضايي معرفي شده است. 2) يک روش انتخاب جديد براي انتخاب افراد مناسب از جمعيت براي اعمال جستجوي محلي ارائه شده است. 3) عملگر انتخاب در روش جستجوي تفاضلي تکاملي به گونه اي بهبود داده شده است که براي افراد نامناسب از جمعيت هم امکان انتخاب وجود داشته باشد. اين استراتژي باعث ميشود فشار انتخاب در مراحل الگوريتم به خوبي کنترل گردد. الگوريتم ممتيک پيشنهادي با چندين الگوريتم يادگيري ديگر روي توابع محک مقايسه شده است. نتايج عملي بدست آمده بهبود نرخ همگرايي و دقت بالاتر مدل هاي بدست آمده را در مقايسه با ساير روش هاي يادگيري نشان مي دهد.



1.        Ho, D.W., Zhang, P.-A. and Xu, J., "Fuzzy wavelet networks for function learning", Fuzzy Systems, IEEE Transactions on,  Vol. 9, No. 1, (2001), 200-211.

2.        Mousavi, S.H., Noroozi, N., Safavi, A.A. and Ebadat, A., "Modeling and control of nonlinear systems using novel fuzzy wavelet networks: The output adaptive control approach", Communications in Nonlinear Science and Numerical Simulation,  Vol. 16, No. 9, (2011), 3798-3814.

3.        Ebadat, A., Noroozi, N., Safavi, A.A. and Mousavi, S.H., "New fuzzy wavelet network for modeling and control: The modeling approach", Communications in Nonlinear Science and Numerical Simulation,  Vol. 16, No. 8, (2011), 3385-3396.

4.        Wei, H.-L., Billings, S.A., Zhao, Y. and Guo, L., "An adaptive wavelet neural network for spatio-temporal system identification", Neural Networks,  Vol. 23, No. 10, (2010), 1286-1299.

5.        Yilmaz, S. and Oysal, Y., "Fuzzy wavelet neural network models for prediction and identification of dynamical systems", Neural Networks, IEEE Transactions on,  Vol. 21, No. 10, (2010), 1599-1609.

6.        Zainuddin, Z. and Pauline, O., "Modified wavelet neural network in function approximation and its application in prediction of time-series pollution data", Applied Soft Computing,  Vol. 11, No. 8, (2011), 4866-4874.

7.        Abiyev, R.H. and Kaynak, O., "Fuzzy wavelet neural networks for identification and control of dynamic plants—a novel structure and a comparative study", Industrial Electronics, IEEE Transactions on,  Vol. 55, No. 8, (2008), 3133-3140.

8.        Tzeng, S.-T., "Design of fuzzy wavelet neural networks using the ga approach for function approximation and system identification", Fuzzy Sets and Systems,  Vol. 161, No. 19, (2010), 2585-2596.

9.        Davanipoor, M., Zekri, M. and Sheikholeslam, F., "Fuzzy wavelet neural network with an accelerated hybrid learning algorithm", Fuzzy Systems, IEEE Transactions on,  Vol. 20, No. 3, (2012), 463-470.

10.     Bodyanskiy, Y. and Vynokurova, O., "Hybrid adaptive wavelet-neuro-fuzzy system for chaotic time series identification", Information Sciences,  Vol. 220, No., (2013), 170-179.

11.     Cao, J., Lin, Z. and Huang, G.-B., "Composite function wavelet neural networks with differential evolution and extreme learning machine", Neural Processing Letters,  Vol. 33, No. 3, (2011), 251-265.

12.     Jun, W. and Hong, P., "Constructing fuzzy wavelet network modeling", International Journal of Information Technology,  Vol. 11, No. 6, (2005), 68-74.

13.     Lee, C.-H. and Teng, C.-C., "Identification and control of dynamic systems using recurrent fuzzy neural networks", Fuzzy Systems, IEEE Transactions on,  Vol. 8, No. 4, (2000), 349-366.

14.     Zekri, M., Sadri, S. and Sheikholeslam, F., "Adaptive fuzzy wavelet network control design for nonlinear systems", Fuzzy Sets and Systems,  Vol. 159, No. 20, (2008), 2668-2695.

15.     Zekri, M., Sadri, S. and Sheikholeslam, F., "Approximation of feedback linearization control input based on fuzzy wavelet neural networks", Iranian Journal of Science & Technology Transaction B, Engineering,,  Vol. 31, No. 6, (2007), 651–662.

16.     Abiyev, R.H., Kaynak, O. and Kayacan, E., "A type-2 fuzzy wavelet neural network for system identification and control", Journal of the Franklin Institute,  Vol. 350, No. 7, (2013), 1658-1685.

17.     Zhang, Z., "Learning algorithm of wavelet network based on sampling theory", Neurocomputing,  Vol. 71, No. 1, (2007), 244-269.

18.     Piotrowski, A.P. and Napiorkowski, J.J., "A comparison of methods to avoid overfitting in neural networks training in the case of catchment runoff modelling", Journal of Hydrology,  Vol. 476, No., (2013), 97-111.

19.     Khedri, P.A., Eftekhari, M. and Maazallahi, R., "Comparing evolutionary algorithms on tuning the parameters of fuzzy wavelet neural network",  Vol., No.

20.     Neri, F. and Cotta, C., "Memetic algorithms and memetic computing optimization: A literature review", Swarm and Evolutionary Computation,  Vol. 2, No., (2012), 1-14.

21.     Sudholt, D., "The impact of parametrization in memetic evolutionary algorithms", Theoretical Computer Science,  Vol. 410, No. 26, (2009), 2511-2528.

22.     Jia, D., Zheng, G. and Khurram Khan, M., "An effective memetic differential evolution algorithm based on chaotic local search", Information Sciences,  Vol. 181, No. 15, (2011), 3175-3187.

23.     Zhang, C., Chen, J. and Xin, B., "Distributed memetic differential evolution with the synergy of lamarckian and baldwinian learning", Applied Soft Computing,  Vol. 13, No. 5, (2013), 2947-2959.

24.     Bao, Y., Hu, Z. and Xiong, T., "A pso and pattern search based memetic algorithm for svms parameters optimization", Neurocomputing,  Vol. 117, No., (2013), 98-106.

25.     Neri, F. and Tirronen, V., "Recent advances in differential evolution: A survey and experimental analysis", Artificial Intelligence Review,  Vol. 33, No. 1-2, (2010), 61-106.

International Journal of Engineering
E-mail: office@ije.ir
Web Site: http://www.ije.ir