The neurocomputing using of the development metamodels stage in the optimal surrogate antennas synthesis process


  • V. Ya. Halchenko Cherkasy State Technological University, Ukraine
  • R. V. Trembovetska Cherkasy State Technological University, Ukraine
  • V. V. Tychkov Cherkasy State Technological University, Ukraine



antenna synthesis, surrogate optimization, metamodel, computer experiment plan, LPτ-sequence, response surface, neural network


Introduction. A computational developing metamodels technology for optimal antenna synthesis problems is proposed. This computational technology is created using methods of data mining, artificial intelligence and modern computer methods of experiment planning. To develop an approximation model, the mathematical apparatus of artificial neural networks, namely the RBF-network, is applied.
Analysis of metamodels developing research. The computer experiment plan is performed with the help of Sobol’s $LP_\tau$-sequences $(\xi_1, \xi_2)$, which in the general case uniformly fill the points with the search space in the unit hypercube. Verification of the proposed technology is performed on test functions of the two variables goal. The obtained metamodels have rather high accuracy of approximation and improved computational efficiency. The created computing metamodels developing technology of provides high modeling speed which makes a possible realization of optimum antennas synthesis procedure. This technology is effective and correct for more complex problems of approximating multidimensional hypersurfaces.
Metamodels developing. To develop the RBF-metamodel, an automatic and user-defined strategy with random sampling is used in the ratio: 70% - training, 15% - control, 15% - test. Training and control samples were used in the metamodel developing, and the test - for cross-verification. At the stage of training best neural networks selection was carried out by indicators: determination coefficient $R^2$; standard forecast error deviations ratio and learning data $S.D.ratio$; average relative model error magnitude MAPE,%; residual average squared error $MS_R$; residues histogram; scattering diagrams.
Results of numerical experiments. Obtained metamodels for test functions $f_1(x,y)$ - RBF-2-130-1 (44); $f_2(x,y)$ - RBF-2-150-1 (6); $f_3(x,y)$ - RBF-2-185-1 (10) have a high enough approximation accuracy and improved computational efficiency. For these metamodels, we checked the adequacy and informativeness of Fisher's criterion. The results of metamodels checking adequacy calculations at the stage of response surface recovery are presented. The created computing metamodels developing technology provides a high simulation speed, which makes possible the implementation of the procedure for optimal antennas synthesis. This technology is effective and correct for more complex problems of multidimensional hypersurfaces approximation.
Conclusions. The numerical experiments results analysis is evidence of the high efficiency of the proposed computing developing metamodels technology, which is created using methods of intellectual data analysis, artificial intelligence and modern computer experiment planning methods. The metamodels developing with its use are characterized by fairly high accuracy of approximation and improved computational efficiency. It is these advantages that allow their using with the optimal surrogate antennas synthesis.

Author Biographies

V. Ya. Halchenko, Cherkasy State Technological University

Halchenko V.Ya.

R. V. Trembovetska, Cherkasy State Technological University

Trembovetska R. V.

V. V. Tychkov, Cherkasy State Technological University

Tychkov V. V.


Zelkin E.G., Sokolov V.G. (1980) Metody sinteza antenn. Fazirovannye antennye reshetki i antenny s nepreryvnym raskryvom [Methods of Synthesizing Antennas. Fixed Antenna Arrays and Antennas with Continuous Opening]. Moskow, Soviet radio, 296 p.

Rahmat-Samii Y. and Christodoulou C. (2007) Guest Editorial for the Special Issue on Synthesis and Optimization Techniques in Electromagnetics and Antenna System Design. IEEE Transactions on Antennas and Propagation, Vol. 55, Iss. 3, pp. 518-522. DOI: 10.1109/tap.2007.891879

Gazizov T.T. (2013) Sintez optimal'nykh provodnykh antenn [Synthesis of Optimal Wired Antennas]. Tomsk, 120 p.

Andriichuk M.I., Voitovich N.N., Savenko P.A. and Tkachuk V.P. (1993) Sintez antenn po amplitudnoi diagramme napravlennosti. Chislennye metody i algoritmy [Synthesis of Antennas from the Amplitude Pattern. Numerical Methods and Algorithms]. Kyiv, Naukova dumka, 256 p.

Grigor'ev A.D. (2012) Metody vychislitel'noi elektrodinamiki [Methods of Computational Electrodynamics]. Moskow, Fizmathlit, 432 p.

Il'inskii A.S., Kravtsov V.V. and Sveshnikov A.G. (1991) Matematicheskie modeli elektrodinamiki [Mathematical Models of Electrodynamics]. Moskow, Vysshaya shkola, 224 p.

Gal'chenko V.Ya. and Vorob'ev M.A. (2003) Ispol'zovanie geneticheskikh algoritmov v strukturnom sinteze istochnikov magnitnykh polei s zadannymi svoistvami [The use of genetic algorithms in the structural synthesis of sources of magnetic fields with specified properties]. Informatsionnye tekhnologii}, No 7, pp. 7-12.

Gal'chenko V.Y. and Vorob'ev M.A. (2005) Structural synthesis of attachable eddy-current probes with a given distribution of the probing field in the test zone. Russian Journal of Nondestructive Testing, Vol. 41, Iss. 1, pp. 29-33. DOI: 10.1007/s11181-005-0124-7

Galchenko V.Y., Yakimov A.N. and Ostapushchenko D.L. (2011) Solution of the inverse problem of creating a uniform magnetic field in coercimeters with partially closed magnetic systems. Russian Journal of Nondestructive Testing, Vol. 47, Iss. 5, pp. 295-307. DOI: 10.1134/s1061830911050056

Gal’chenko V.Y., Yakimov A.N. and Ostapushchenko D.L. (2012) Pareto-optimal parametric synthesis of axisymmetric magnetic systems with allowance for nonlinear properties of the ferromagnet. Technical Physics, Vol. 57, Iss. 7, pp. 893-899. DOI: 10.1134/s1063784212070110

Garifullin M.R., Naumova E.A., Zhuvak O.V. and Barabash A.V. (2016) Surrogate modeling in construction. Construction of Unique Buildings and Structures, No 2 (41), pp. 118-132. (in Russian)

Burnaev E.V., Panov M., Kononenko D. and Konovalenko I. (2012) Comparative analysis of optimization procedures based on Gaussian processes. Informatsionnye tekhnologii i sistemy pp. 167-172. (in Russian)

Burnaev E.V. and Prikhod'ko P.V. (2013) Metodologiya postroeniya surrogatnykh modelei dlya approksimatsii prostranstvenno-neodnorodnykh funktsii [Methodology for constructing surrogate models for the approximation of spatially inhomogeneous functions]. Trudy MFTI. Informatika, matematika, Vol. 5, No 4, pp. 122-132.

Bedrintsev A. and Chepyzhov V. (2016) Convex approximation of the design space in the aircraft wing optimization problem. Informatsionnye protsessy. Vol. 16, No 2, pp. 91-102.

Bondarenko M. (2016) Optimization methods using response surfaces adapted to the tasks of analysis and synthesis of thin-walled machine structures design parameters. Bulletin of the National Technical University «KhPI» Series: New solutions in modern technologies, Iss. 42 (1214), pp. 22-28. DOI: 10.20998/2413-4295.2016.42.04

Bandler J., Cheng Q., Dakroury S., Mohamed A., Bakr M., Madsen K. and Sondergaard J. (2004) Space Mapping: The State of the Art. IEEE Transactions on Microwave Theory and Techniques, Vol. 52, Iss. 1, pp. 337-361. DOI: 10.1109/tmtt.2003.820904

Bandler J., Hailu D., Madsen K. and Pedersen F. (2004) A Space-Mapping Interpolating Surrogate Algorithm for Highly Optimized EM-Based Design of Microwave Devices. IEEE Transactions on Microwave Theory and Techniques, Vol. 52, Iss. 11, pp. 2593-2600. DOI: 10.1109/tmtt.2004.837197

Bakr M., Bandler J., Ismail M., Rayas-Sanchez J. and Zhang Q. () Neural space mapping EM optimization of microwave structures. 2000 IEEE MTT-S International Microwave Symposium Digest (Cat. No.00CH37017). DOI: 10.1109/mwsym.2000.863320

Queipo N.V., Haftka R.T., Shyy W., Goel T., Vaidyanathan R. and Tucker P.K. (2005) Surrogate-based analysis and optimization. Progress in Aerospace Sciences, Vol. 41, Iss. 1, pp. 1-28. DOI: 10.1016/j.paerosci.2005.02.001

Coleman C., Rothwell E. and Ross J. (2004) Investigation of Simulated Annealing, Ant-Colony Optimization, and Genetic Algorithms for Self-Structuring Antennas. IEEE Transactions on Antennas and Propagation, Vol. 52, Iss. 4, pp. 1007-1014. DOI: 10.1109/tap.2004.825658

Kabir H., Zhang L., Yu M., Aaen P., Wood J. and Zhang Q. (2010) Smart Modeling of Microwave Devices. IEEE Microwave Magazine, Vol. 11, Iss. 3, pp. 105-118. DOI: 10.1109/mmm.2010.936079

Liu B., Koziel S. and Ali N. (2017) SADEA-II: A generalized method for efficient global optimization of antenna design. Journal of Computational Design and Engineering, Vol. 4, Iss. 2, pp. 86-97. DOI: 10.1016/j.jcde.2016.11.002

Dubrovka, F. F., Vasylenko, D. O. (2008) Neural-genetic method for synthesis of antennas and microwave devices. Visn. NTUU KPI, Ser. Radioteh. radioaparatobuduv., no. 36, pp. 60-66. (in Ukrainian) DOI: 10.20535/RADAP.2018.72.42-46

Dubrovka, F. F., Vasylenko, D. O. (2008) Synthesis of ultrawideband planar dipole bow-tie antenna by neural-genetic method. Visn. NTUU KPI, Ser. Radioteh. radioaparatobuduv., no. 37, pp. 53-60. (in Ukrainian) DOI: 10.20535/RADAP.2008.37.53-60

Dubrovka F.F. and Vasylenko D.O. (2009) Synthesis of UWB planar antennas by means of natural optimization algorithms. Radioelectronics and Communications Systems, Vol. 52, Iss. 4, pp. 167-178. DOI: 10.3103/s0735272709040013

Vasylenko D., Edenhofer P. and Dubrovka F. (2008) Genetic algorithm based inversion of neural networks applied to optimised design of UWB planar antennas. Electronics Letters, Vol. 44, Iss. 3, pp. 177. DOI: 10.1049/el:20083395

Guodong H., Wei W. and Biao D. (2012) Perturbation alternating projections method for pattern synthesis of phased array antenna. Proceedings of 2012 5th Global Symposium on Millimeter-Waves. DOI: 10.1109/gsmm.2012.6314080

Tselykh V.R. (2012) Multivariate adaptive regression splines. Mashinnoe obuchenie i analiz dannykh, Vol. 1, No 3, pp. 272-278. (in Russian)

Afonin P.V. (2016) Optimizatsiya modelei slozhnykh sistem na osnove metaevristicheskikh algoritmov i neironnykh setei [Optimization of models of complex systems based on meta-heuristic algorithms and neural networks]. Inzhenernyi vestnik, No 11, pp. 508-516.

Haykin S. (1998) Neural networks. A comprehensive foundation (2nd Edition), Prentice Hall, 864 p.

Belyaev M.G. (2013) Approksimatsiya mnogomernykh zavisimostei po strukturirovannym vyborkam [Approximation of multivariate dependencies on structured samples]. Iskusstvennyi intellekt i prinyatie reshenii, No 3, pp. 24–39. Approximation problem for factorized data. No 3, pp. 24–39.

Belyaev M.G. and Lyubin A.D. (2011) Osobennosti optimizatsionnoi zadachi, voznikayushchei pri postroenii approksimatsii mnogomernoi zavisimosti [Features of the optimization problem arising in the construction of the approximation of a multidimensional dependence]. Informatsionnye Tekhnologii i Sistemy, pp. 415–422.

Sobol' I.M. and Statnikov R.B. (2006) Vybor optimal'nykh parametrov v zadachakh so mnogimi kriteriyami [The choice of optimal parameters in problems with many criteria]. Moskow, Drofa, 175 p.

Radchenko S.G. Metodologiya regressionnogo analiza [The methodology of regression analysis: monograph]. Kyiv, Kornіichuk, 376 p.

Trembovetska R.V., Halchenko V.Ya. and Tychkov V.V. (2018) Zastosuvannia MLP-metamodelei v zadachakh surohatnoi optymizatsii [Application of MLP-metamodels in surrogate optimization tasks]. Molodyi vchenyi, No 2 (54), pp. 32-39.





Computing methods in radio electronics