Construction of an analytical method for limiting the complexity of neural-fuzzy models with guaranteed accuracy

Authors

DOI:

https://doi.org/10.15587/1729-4061.2019.160719

Keywords:

approximation, guaranteed accuracy, fuzzy logic, neural networks, imitation simulation

Abstract

We have proposed an analytical method for limiting the complexity of neural-fuzzy models that provide for the guaranteed accuracy of their implementation when approximating functions with two or more derivatives. The method makes it possible to determine the required minimal number of parameters for systems that employ fuzzy logic, as well as neural models.

We have estimated the required number of neurons (terms) in a model, which ensure the accuracy required for the area of a model curve to approach the system one along the sections of function approximation. The estimate for an approximation error was obtained based on the residual members of decomposition, in the Lagrangian form, of areas of the approximated system function into a Maclaurin series. The results received make it possible to determine the required number of approximation sections and the number of neurons (terms) in order to ensure the assigned relative and absolute error of approximation.

We have estimated the required number of neurons (terms) that provide for the necessary accuracy of model implementation based on the maximum deviation between the system and model curves along the section of approximation. This makes it possible to select, depending on the assigned required accuracy, the number of terms of fuzzy variables, input and output variables, linguistic rules, coordinates of modal values along the axes of input and output variables.

To verify validity of the proposed solutions, we modeled the system curves in the Matlab/Simulink environment, which confirmed the guaranteed accuracy of their implementation in accordance with the analytical calculations reported earlier.

The results obtained could be applied in modern intelligent technical systems of management, control, diagnosis, and decision-making. Using the proposed methods for selecting and applying the minimal number of terms (neurons) would help reduce the required computing power in nonlinear systems

Author Biographies

Borys Sytnik, Ukrainian State University of Railway Transport Feierbakh sq., 7, Kharkiv, Ukraine, 61050

PhD, Associate Professor

Department of Information Technologies

Volodymyr Bryksin, Ukrainian State University of Railway Transport Feierbakh sq., 7, Kharkiv, Ukraine, 61050

PhD, Associate Professor

Department of Information Technologies

Sergiy Yatsko, Ukrainian State University of Railway Transport Feierbakh sq., 7, Kharkiv, Ukraine, 61050

PhD, Associate Professor

Department of Electrical Energetics, Electrical Engineering and Electromechanics

Yaroslav Vashchenko, Ukrainian State University of Railway Transport Feierbakh sq., 7, Kharkiv, Ukraine, 61050

PhD, Senior Lecturer

Department of Electrical Energetics, Electrical Engineering and Electromechanics

References

  1. Basov, H. H., Yatsko, S. I. (2005). Rozvytok elektrychnoho motorvahonnoho rukhomoho skladu. Ch. 2. Kharkiv: «Apeks+», 248.
  2. Yatsko, S., Sytnik, B., Vashchenko, Y., Sidorenko, A., Liubarskyi, B., Veretennikov, I., Glebova, M. (2019). Comprehensive approach to modeling dynamic processes in the system of underground rail electric traction. Eastern-European Journal of Enterprise Technologies, 1 (9 (97)), 48–57. doi: https://doi.org/10.15587/1729-4061.2019.154520
  3. Chao, C.-T., Sutarna, N., Chiou, J.-S., Wang, C.-J. (2017). Equivalence between Fuzzy PID Controllers and Conventional PID Controllers. Applied Sciences, 7 (6), 513. doi: https://doi.org/10.3390/app7060513
  4. Helwa, M. K., Heins, A., Schoellig, A. (2018). Provably robust learning-based approach for high-accuracy tracking control of Lagrangian systems. 57th IEEE Conference on decision and control, 1–8.
  5. Bryksin, V. A., Mihaylenko, V. S., Sytnik, B. T., Yac'ko, S. I. (2011). Realizaciya neyronechetkih modeley i regulyatorov garantirovannoy tochnosti. Informatsiyno-keruiuchi systemy na zaliznychnomu transporti, 4, 24–29.
  6. Landowski, M. (2018). A discussion on “On the solution of a class of fuzzy system of linear equations”. Sādhanā, 43 (12). doi: https://doi.org/10.1007/s12046-018-0972-1
  7. Rakytianska, H. B. (2015). Neural-network approach to structural tuning of classification rules based on fuzzy relational equations. Eastern-European Journal of Enterprise Technologies, 4 (2 (76)), 51–57. doi: https://doi.org/10.15587/1729-4061.2015.47124
  8. Liu, Y.-J., Gao, Y., Tong, S., Li, Y. (2016). Fuzzy Approximation-Based Adaptive Backstepping Optimal Control for a Class of Nonlinear Discrete-Time Systems With Dead-Zone. IEEE Transactions on Fuzzy Systems, 24 (1), 16–28. doi: https://doi.org/10.1109/tfuzz.2015.2418000
  9. Lozynskyy, A., Demkiv, L. (2016). Synthesis of fuzzy logic controller of nonlinear dynamic system with variable parameters. Computational problems of electrical engineering, 6 (2), 91–98.
  10. Orlowska-Kowalska, T., Kaminski, M., Szabat, K. (2010). Implementation of a Sliding-Mode Controller With an Integral Function and Fuzzy Gain Value for the Electrical Drive With an Elastic Joint. IEEE Transactions on Industrial Electronics, 57 (4), 1309–1317. doi: https://doi.org/10.1109/tie.2009.2030823
  11. Chen, B., Liu, X. P., Ge, S. S., Lin, C. (2012). Adaptive Fuzzy Control of a Class of Nonlinear Systems by Fuzzy Approximation Approach. IEEE Transactions on Fuzzy Systems, 20 (6), 1012–1021. doi: https://doi.org/10.1109/tfuzz.2012.2190048
  12. Anastassiou, G. A. (2014). Higher Order Multivariate Fuzzy Approximation by basic Neural Network Operators. Cubo (Temuco), 16 (3), 21–35. doi: https://doi.org/10.4067/s0719-06462014000300003
  13. Piegat, A. (2001). Fuzzy Modeling and Control. Springer, 728. doi: https://doi.org/10.1007/978-3-7908-1824-6
  14. Ronzhin, A., Rigoll, G., Meshcheryakov, R. (Eds.) (2018). Interactive collaborative Robotics. Proceedings of the Third International Conference on Interactive Collaborative Robotics, ICR. Springer, 302. doi: https://doi.org/10.1007/978-3-319-99582-3
  15. Kimura, S., Sonoda, K., Yamane, S., Matsumura, K., Hatakeyama, M. (2007). Function approximation approach to the inference of neural network models of genetic networks. IPSJ Transactions on bioinformatics, 48, 9–19.

Published

2019-03-25

How to Cite

Sytnik, B., Bryksin, V., Yatsko, S., & Vashchenko, Y. (2019). Construction of an analytical method for limiting the complexity of neural-fuzzy models with guaranteed accuracy. Eastern-European Journal of Enterprise Technologies, 2(4 (98), 6–13. https://doi.org/10.15587/1729-4061.2019.160719

Issue

Section

Mathematics and Cybernetics - applied aspects