NONLINEAR SYSTEM IDENTIFICATION BASED ON FUZZY RADIAL BASIS NEURAL NETWORK WITH MULTI-CONNECTED WEIGHT CONNECTIONS

Authors

• Kabul Khudaybergenov, Department of Applied Informatics, Kimyo International University in Tashkent. Address: st. Shota Rustaveli, 156, 100121 Tashkent, Uzbekistan. E-mail: [email protected], Phone: +998-91-371-51-27.Follow

Abstract

This paper builds on our earlier radial basis function network with multiple connections (RBFMC) by placing it within a fuzzy inference framework for nonlinear system identification. The idea is inspired by the diversity of neurotransmitters found in biological neurons: instead of a single hidden-to-output weight, RBFMC gives each hidden unit a multi-dimensional connection whose components act as independent filters. Once fuzzy logic is added, each hidden neuron becomes a fuzzy rule, and its antecedent is built from several Gaussian membership functions, one per connection. The resulting Fuzzy RBFMC produces an interpretable, multi-filter description of local regions of the input space without sacrificing universal approximation capability, and it is trained through gradient descent with momentum, . regularisation, and early stopping. After describing the architecture, the cooperative and competitive interplay among multi-connection rules, and the training procedure, we evaluate the model on four benchmark nonlinear systems - Narendra-Li, the Box-Jenkins gas furnace, a continuous stirred tank reactor, and a high-dimensional NARX plant - comparing it against standard RBF, fuzzy RBF, RBFMC, and ANFIS. Across all four benchmarks, Fuzzy RBFMC achieves the best accuracy, the fastest convergence, and the most compact rule base, and it produces interpretable fuzzy rules that expose local system dynamics, a conclusion supported by ablation, noise-robustness, and statistical significance analyses.

First Page

91

Last Page

105

References

1. Marakhimov, A.R., Khudaybergenov, K., Muminov, Z.I. (2026). Radial basis neural network with multiple connections. Lecture Notes in Networks and Systems. (accepted).
2. Yin, Y., Han, Z., Jian, M., Wang, G., Chen, L., Wang, R. (2023). AMSUnet: A neural network using atrous multi-scale convolution for medical image segmentation. Computers in Biology and Medicine. 162. 107120.
3. Farrell, M., Recanatesi, S., Shea-Brown, E. (2023). From lazy to rich to exclusive task representations in neural networks and neural codes. Current Opinion in Neurobiology. 83. 102780.
4. Apicella, A., Isgro, F., Pollastro, A., Prevete, R. (2023). Adaptive filters in Graph Convolutional Neural Networks. Pattern Recognition. 144. 109867.
5. Gari, J., Romero, E., Mazzanti, F. (2024). Learning restricted Boltzmann machines with pattern induced weights. Neurocomputing. 610. 128469.
6. Fan, M., Zuo, K., Wang, J., Zhu, J. (2023). A lightweight multiscale convolutional neural network for garbage sorting. Systems and Soft Computing. 5. 200059.
7. Novoa-Paradela, D., Fontenla-Romero, O., Guijarro-Berdinas, B. (2023). Fast deep autoencoder for federated learning. Pattern Recognition. 143. 109805.
8. Revin, I., Balabanov, N., Litvintseva, A. (2023). Light-weight ensembling of deep neural models for object recognition in remote sensing data. Procedia Computer Science. 229. 18-27.
9. Herrera, C., Whittle, N., Leek, M.R., Brodbeck, C., Lee, G., Barcenas, C., Barnes, S., Holshouser, B., Yi, A., Venezia, J.H. (2023). Cortical networks for recognition of speech with simultaneous talkers. Hearing Research. 437. 108856.
10. Kolb, B., Whishaw, I.Q. (2014). An Introduction to Brain and Behavior. 4th ed. Worth Publishers.
11. Lauder, J.M. (1993). Neurotransmitters as growth regulatory signals: role of receptors and second messengers. Trends in Neurosciences. 16(6). 233-240.
12. Jang, J.-S.R. (1993). ANFIS: Adaptive-network-based fuzzy inference system. IEEE Transactions on Systems, Man, and Cybernetics. 23(3). 665-685.
13. Babuška, R., Verbruggen, H. (2003). Neuro-fuzzy methods for nonlinear system identification. Annual Reviews in Control. 27(1). 73-85.
14. Marakhimov, A.R., Khudaybergenov, K.K. (2022). A fuzzy MLP approach for nonlinear system identification. Journal of Mathematical Sciences. 265(1). 43-51.
15. Funahashi, K. (1989). On the approximate realization of continuous mappings by neural networks. Neural Networks. 2. 183-192.
16. Jin, Y., Sendhoff, B. (2003). Extracting interpretable fuzzy rules from RBF networks. Neural Processing Letters. 17(2). 149-164.
17. Marakhimov, A.R., Khudaybergenov, K.K. (2020). Approach to the synthesis of neural network structure during classification. International Journal of Computing. 19(1). 20-26.
18. Kurkova, V. (1992). Kolmogorov's theorem and multilayer neural networks. Neural Networks. 5. 501-506.
19. Maiorov, V., Pinkus, A. (1999). Lower bounds for approximation by MLP neural networks. Neurocomputing. 25. 81-91.
20. Ismailov, V.E. (2014). On the approximation by neural networks with bounded number of neurons in hidden layers. Journal of Mathematical Analysis and Applications. 417(2). 963-969.
21. Hinton, G.E. (2009). Deep belief networks. Scholarpedia. 4(5). 5947.
22. Salakhutdinov, R., Hinton, G.E. (2009). Deep Boltzmann machines. Artificial Intelligence and Statistics. 5(2). 448-455.
23. Chen, S., Cowan, C.F.N., Grant, P.M. (1991). Orthogonal least squares learning algorithm for radial basis function networks. IEEE Transactions on Neural Networks. 2(2). 302-309.
24. Chen, S., Billings, S.A. (1992). Neural networks for nonlinear dynamic system modelling and identification. International Journal of Control. 56(2). 319-346.
25. Hernandez, D., et al. (2020). Modelling and control of a CSTR: A benchmark for data-driven methods. Journal of Process Control. 95. 1-12.

Recommended Citation

Khudaybergenov, Kabul (2026) "NONLINEAR SYSTEM IDENTIFICATION BASED ON FUZZY RADIAL BASIS NEURAL NETWORK WITH MULTI-CONNECTED WEIGHT CONNECTIONS," Chemical Technology, Control and Management: Vol. 2026: Iss. 4, Article 10.
DOI: https://doi.org/10.59048/2181-1105.1790