Research Article
BibTex RIS Cite

COMPARISON OF PERFORMANCE OF DIFFERENT SAMPLING METHODS IN ARTIFICIAL NEURAL NETWORKS

Year 2022, Volume: 10 Issue: 2, 412 - 428, 30.06.2022
https://doi.org/10.21923/jesd.1012106

Abstract

In optimization modeling problems, the selection of the data to be used is of great importance. In this study, the effect of different sampling methods on the performance of Artificial Neural Networks is compared on the microstrip patch antenna by using the Multilayer Perceptron, which is the model that is frequently used in antenna designs. The selected neural network black box model consists of 5 input and 1 output parameters. The application of Latin Hypercube and Monte Carlo sampling methods was examined in the selection of data to be used in microstrip patch antenna modeling. First of all, Latin Hypercube and Monte Carlo samples were obtained according to their unique creation method in the specified value ranges of the sample number (training and validation data sets) input parameters. Afterwards, the problems with 2 different sample numbers were randomly separated at a rate of 50% and 33%, and training and validation data were created. In the performance comparison, a total of 12 different networks with 3 algorithms and 4 different architectural structures, as well as 4 different training and validation data sets are used. When the results are compared with each other, it is seen that Monte Carlo sampling method gives more successful results in terms of performance in modeling with low or high educational sample number. On the other hand, it is seen that the Latin Hypercube sampling method, on the other hand, increased the number of training samples and caused a partial improvement. However, it still lags behind the other Monte Carlo sampling method, which has less sample size, in terms of performance. Therefore, it was concluded that the Monte Carlo sampling method is more applicable for this and similar problems.

References

  • Chávez-Hurtado, J. L., Rayas-Sánchez, J. E., Brito-Brito, Z., 2016. Multiphysics polynomial-based surrogate modeling of microwave structures in frequency domain. IEEE MTT-S Latin America Microwave Conference (LAMC).
  • Rayas-Sánchez, J. E., Chávez-Hurtado, J. L., Brito-Brito, Z., 2015. Enhanced formulation for polynomial-based surrogate modeling of microwave structures in frequency domain. IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization (NEMO).
  • Rayas-Sánchez, J. E., Vargas-Chávez, N., 2010. Design optimization of microstrip lines with via fences through surrogate modeling based on polynomial functional interpolants. 19th Topical Meeting on Electrical Performance of Electronic Packaging and Systems, 125-128.
  • Rayas-Sánchez, J. E., Aguilar-Torrentera, J., Jasso-Urzúa, J. A., 2010. Surrogate modeling of microwave circuits using polynomial functional interpolants. IEEE MTT-S International Microwave Symposium, 197-200.
  • Khusro, A., Hashmi, M. S., Ansari, A. Q., 2018. Exploring Support Vector Regression for Modeling of GaN HEMT. IEEE MTT-S International Microwave and RF Conference (IMaRC).
  • Güneş F., Demirel S., Mahouti P., 2014. Design of a Front–End Amplifier for the Maximum Power Delivery and Required Noise by HBMO with Support Vector Microstrip Model. Radioengineering, 23(1).
  • Geng, L., Gao, X., 2014. Support vector machine dynamic modeling method and its application in the fermentation process. The 26th Chinese Control and Decision Conference (2014 CCDC), 4478-4482.
  • Jacobs, J. P., Koziel, S., Ogurtsov, S., 2013. Computationally Efficient Multi-Fidelity Bayesian Support Vector Regression Modeling of Planar Antenna Input Characteristics. IEEE Transactions on Antennas and Propagation, 61(2), 980-984.
  • Tokan, N. T., Güneş, F., 2008. Analysis and Synthesis of the Microstrip Lines Based on Support Vector Regression. 38th European Microwave Conference, 1473-1476.
  • Yuan, L., Yang, X-S., Wang, C., Wang, B-Z., 2020. Multibranch Artificial Neural Network Modeling for Inverse Estimation of Antenna Array Directivity. IEEE Transactions on Antennas and Propagation, 68(6), 4417-4427.
  • Xiao, L-Y., Shao, W., Jin, F-L., Wang, B-Z., 2018. Multiparameter Modeling with ANN for Antenna Design. IEEE Transactions on Antennas and Propagation, 66(7), 3718-3723.
  • Kapetanakis, T. N., Vardiambasis, I. O., Ioannidou, M. P., Maras, A., 2018. Neural Network Modeling for the Solution of the Inverse Loop Antenna Radiation Problem. IEEE Transactions on Antennas and Propagation, 66(11), 6283-6290.
  • Xiao, L-Y., Shao, W., Jin, F-L., Wang, B-Z., Liu, Q. H., 2021. Inverse Artificial Neural Network for Multiobjective Antenna Design. IEEE Transactions on Antennas and Propagation, 69(10), 6651-6659.
  • Mahouti, P., Güneş, F., Belen, M. A., Demirel, S., 2017. Symbolic Regression for Derivation of an Accurate Analytical Formulation using "Big Data" An Application Example. Applied Computational Electromagnetics Society Journal, 32(5), 372-380.
  • Uluslu, A., 2021. Rekabetçi Evrimsel Algoritmalar ile Yuvarlak Papyon Anten Tasarımı. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, 12(4), 551-564, DOI: 10.24012/dumf.1001870.
  • Uluslu, A., 2021. Triangular Bowtie Antenna Design and Modelling - Research & Reviews in Engineering. Gece Kitaplığı Yayınevi, 93-106.
  • Güneş, F., Mahouti, P., Demirel, S., Belen, M. A., Uluslu, A., 2017. Cost-effective GRNN-based modeling of microwave transistors with a reduced number of measurements. International journal of numerical modelling: electronic networks, devices and fields, 30(3-4).
  • Mahouti, P., Güneş, F., Demirel, S., Uluslu, A., Belen, M. A., 2014. Efficient scattering parameter modeling of a microwave transistor using Generalized Regression Neural Network. 20th International Conference on Microwaves, Radar and Wireless Communications (MIKON).
  • Liu, B., Yang, H., Lancaster, M. J., 2018. Synthesis of coupling matrix for diplexers based on a self-adaptive differential evolution algorithm. IEEE Transactions on Microwave Theory and Techniques, 66(2), 813–821.
  • Lei, G.Y., 2003. Study on Monte Carlo and quasi-Monte Carlo methods, Zhejiang Province.
  • McKay, M. D., Beckman, R. J., Conover, W. J., 1979. Comparison of three methods for selecting values of input variables in the analysis of output from a computer code, Technometrics, 21(2), 239-245.
  • Uluslu, A., 2021. Microstrip Low Pass Filter Analysis and Design. Research & Reviews in Engineering- Gece Kitaplığı Yayınevi, 35-52. Mahouti, T., Yıldırım, T., Kuşkonmaz, N., 2021. Artificial intelligence–based design optimization of nonuniform microstrip line band pass filter. International Journal of Numerical Modelling.
  • Mahouti, T., Kuşkonmaz, N., Yıldırım, T., 2019. Modelling of Non-Planar Microstrip Lines via Artificial Neural Networks. Innovations in Intelligent Systems and Applications Conference (ASYU).
  • Uluslu, A., 2021. Kültürel Algoritma Tabanlı 24 GHz ISM Bant Uygulamaları için H Şeklinde Yama Anten Tasarımı. 10th Union Radio-Scientifique Internationale/International Union of Radio Science (URSI'2021).
  • Uluslu, A., 2021. Triangular Patch Antenna Optimization for 77 GHz Automotive Radar Based on Genetic Algorithm. II. International Conference on Innovative Engineering Applications (CIEA’2021).
  • Balanis, C.A., 1982. Antenna Theory Analysis and Design, John Wiley and Sons, Arizona State University, 4-6.
  • Safran, M.I., Aydin, E., 2006. Pek Geniş Bant Anten Tasarımı ve İmalatı, Atılım Üniversitesi.
  • Hossain, F., Anagnostou, E. N., Bagtzoglou, A. C., 2005. On Latin Hypercube sampling for efficient uncertainty estimation of satellite rainfall observations in flood prediction, 32, 776-792.
  • Wan, Y., Lv, Z. Z., Yuan, X. K., 2008. Latin Hypercube sampling and updated Latin Hypercube sampling method for reliability sensitivity and its variance analysis, 30(6), 927-934.
  • Wei, Q., Lu, W. X., Tian, Z. J., 2004. Application of Monte-Carlo to annual precipitation forecast. Journal of Arid Land Resources and Environment, 4(18), 144-146.
  • Jin, C., 2005. Study on random number generator and random sampling in Monte Carlo method. Dalian University of Technology.
  • Zhou, Y., 1997. Study of Sampling Methods in Monte Carlo Simulation. Structure Environment Engineering, (3), 14-18.
  • Şen, Z., 2004. Yapay Sinir Ağları İlkeleri, Su Vakfı Yayınları, İstanbul.
  • Mahouti, P., 2019. Design optimization of a pattern reconfigurable microstrip antenna using differential evolution and 3D EM simulation-based neural network model. International Journal of RF and Microwave Computer-Aided Engineering, 29(8).

FARKLI ÖRNEKLEME YÖNTEMLERİNİN YAPAY SİNİR AĞLARINDA PERFORMANSININ KARŞILAŞTIRILMASI

Year 2022, Volume: 10 Issue: 2, 412 - 428, 30.06.2022
https://doi.org/10.21923/jesd.1012106

Abstract

Optimizasyon modelleme problemlerinde, kullanılacak verinin seçimi büyük önem arz etmektedir. Bu çalışmada farklı örnekleme yöntemlerinin Yapay Sinir Ağlarının başarısındaki etkisini, anten tasarımlarında sıkça kullanılan model olan Çok Katmanlı Algılayıcı kullanılarak mikroşerit yama anteni üzerinde performans karşılaştırılması yapılmaktadır. Seçilen Yapay Sinir Ağı kara kutu modeli 5 giriş ve 1 çıkış parametrelerinden oluşmaktadır. Mikroşerit yama anten modellenmesinde kullanılacak veri seçiminde Latin Hiperküp ve Monte Carlo örnekleme yönteminin uygulanması incelenmiştir. Öncelikle örnek sayısı (eğitim ve doğrulama veri setleri) giriş parametrelerinin belirlenen değer aralıklarında Latin Hiperküp ve Monte Carlo örnekleri kendilerine özgün oluşturulma yöntemine göre temin edilmiştir. Akabinde oluşturulan 2 farklı örnek sayısına sahip problemler kendi içlerinde %50 ve %33 oranında rastgele ayrılarak eğitim ve doğrulama verileri oluşturulmuştur. Performans karşılaştırmasında, 3 algoritma ile 4 farklı mimari yapıya sahip toplamda 12 farklı ağ ile birlikte 4 farklı sayıda eğitim ve doğrulama veri setleri kullanılmaktadır. Çıkan sonuçlar birbirleri ile mukayese edildiğinde, düşük veya yüksek eğitim örnek sayısına sahip modellemede de Monte Carlo örnekleme yönteminin performans olarak daha başarılı sonuçlar verdiği görülmektedir. Buna karşın kendi içinde Latin Hiperküp örnekleme yönteminin ise eğitim örnek sayısında artışa gidilmesi kısmi iyileşmeye neden olduğu görülmektedir. Fakat yine de daha az örnek sayısına sahip olan diğer Monte Carlo örnekleme yönteminin performans olarak gerisinde kalmıştır. Dolayısı ile Monte Carlo örnekleme yönteminin bu ve benzer problemler için daha uygulanabilir olduğuna kanaat getirilmiştir.

References

  • Chávez-Hurtado, J. L., Rayas-Sánchez, J. E., Brito-Brito, Z., 2016. Multiphysics polynomial-based surrogate modeling of microwave structures in frequency domain. IEEE MTT-S Latin America Microwave Conference (LAMC).
  • Rayas-Sánchez, J. E., Chávez-Hurtado, J. L., Brito-Brito, Z., 2015. Enhanced formulation for polynomial-based surrogate modeling of microwave structures in frequency domain. IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization (NEMO).
  • Rayas-Sánchez, J. E., Vargas-Chávez, N., 2010. Design optimization of microstrip lines with via fences through surrogate modeling based on polynomial functional interpolants. 19th Topical Meeting on Electrical Performance of Electronic Packaging and Systems, 125-128.
  • Rayas-Sánchez, J. E., Aguilar-Torrentera, J., Jasso-Urzúa, J. A., 2010. Surrogate modeling of microwave circuits using polynomial functional interpolants. IEEE MTT-S International Microwave Symposium, 197-200.
  • Khusro, A., Hashmi, M. S., Ansari, A. Q., 2018. Exploring Support Vector Regression for Modeling of GaN HEMT. IEEE MTT-S International Microwave and RF Conference (IMaRC).
  • Güneş F., Demirel S., Mahouti P., 2014. Design of a Front–End Amplifier for the Maximum Power Delivery and Required Noise by HBMO with Support Vector Microstrip Model. Radioengineering, 23(1).
  • Geng, L., Gao, X., 2014. Support vector machine dynamic modeling method and its application in the fermentation process. The 26th Chinese Control and Decision Conference (2014 CCDC), 4478-4482.
  • Jacobs, J. P., Koziel, S., Ogurtsov, S., 2013. Computationally Efficient Multi-Fidelity Bayesian Support Vector Regression Modeling of Planar Antenna Input Characteristics. IEEE Transactions on Antennas and Propagation, 61(2), 980-984.
  • Tokan, N. T., Güneş, F., 2008. Analysis and Synthesis of the Microstrip Lines Based on Support Vector Regression. 38th European Microwave Conference, 1473-1476.
  • Yuan, L., Yang, X-S., Wang, C., Wang, B-Z., 2020. Multibranch Artificial Neural Network Modeling for Inverse Estimation of Antenna Array Directivity. IEEE Transactions on Antennas and Propagation, 68(6), 4417-4427.
  • Xiao, L-Y., Shao, W., Jin, F-L., Wang, B-Z., 2018. Multiparameter Modeling with ANN for Antenna Design. IEEE Transactions on Antennas and Propagation, 66(7), 3718-3723.
  • Kapetanakis, T. N., Vardiambasis, I. O., Ioannidou, M. P., Maras, A., 2018. Neural Network Modeling for the Solution of the Inverse Loop Antenna Radiation Problem. IEEE Transactions on Antennas and Propagation, 66(11), 6283-6290.
  • Xiao, L-Y., Shao, W., Jin, F-L., Wang, B-Z., Liu, Q. H., 2021. Inverse Artificial Neural Network for Multiobjective Antenna Design. IEEE Transactions on Antennas and Propagation, 69(10), 6651-6659.
  • Mahouti, P., Güneş, F., Belen, M. A., Demirel, S., 2017. Symbolic Regression for Derivation of an Accurate Analytical Formulation using "Big Data" An Application Example. Applied Computational Electromagnetics Society Journal, 32(5), 372-380.
  • Uluslu, A., 2021. Rekabetçi Evrimsel Algoritmalar ile Yuvarlak Papyon Anten Tasarımı. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, 12(4), 551-564, DOI: 10.24012/dumf.1001870.
  • Uluslu, A., 2021. Triangular Bowtie Antenna Design and Modelling - Research & Reviews in Engineering. Gece Kitaplığı Yayınevi, 93-106.
  • Güneş, F., Mahouti, P., Demirel, S., Belen, M. A., Uluslu, A., 2017. Cost-effective GRNN-based modeling of microwave transistors with a reduced number of measurements. International journal of numerical modelling: electronic networks, devices and fields, 30(3-4).
  • Mahouti, P., Güneş, F., Demirel, S., Uluslu, A., Belen, M. A., 2014. Efficient scattering parameter modeling of a microwave transistor using Generalized Regression Neural Network. 20th International Conference on Microwaves, Radar and Wireless Communications (MIKON).
  • Liu, B., Yang, H., Lancaster, M. J., 2018. Synthesis of coupling matrix for diplexers based on a self-adaptive differential evolution algorithm. IEEE Transactions on Microwave Theory and Techniques, 66(2), 813–821.
  • Lei, G.Y., 2003. Study on Monte Carlo and quasi-Monte Carlo methods, Zhejiang Province.
  • McKay, M. D., Beckman, R. J., Conover, W. J., 1979. Comparison of three methods for selecting values of input variables in the analysis of output from a computer code, Technometrics, 21(2), 239-245.
  • Uluslu, A., 2021. Microstrip Low Pass Filter Analysis and Design. Research & Reviews in Engineering- Gece Kitaplığı Yayınevi, 35-52. Mahouti, T., Yıldırım, T., Kuşkonmaz, N., 2021. Artificial intelligence–based design optimization of nonuniform microstrip line band pass filter. International Journal of Numerical Modelling.
  • Mahouti, T., Kuşkonmaz, N., Yıldırım, T., 2019. Modelling of Non-Planar Microstrip Lines via Artificial Neural Networks. Innovations in Intelligent Systems and Applications Conference (ASYU).
  • Uluslu, A., 2021. Kültürel Algoritma Tabanlı 24 GHz ISM Bant Uygulamaları için H Şeklinde Yama Anten Tasarımı. 10th Union Radio-Scientifique Internationale/International Union of Radio Science (URSI'2021).
  • Uluslu, A., 2021. Triangular Patch Antenna Optimization for 77 GHz Automotive Radar Based on Genetic Algorithm. II. International Conference on Innovative Engineering Applications (CIEA’2021).
  • Balanis, C.A., 1982. Antenna Theory Analysis and Design, John Wiley and Sons, Arizona State University, 4-6.
  • Safran, M.I., Aydin, E., 2006. Pek Geniş Bant Anten Tasarımı ve İmalatı, Atılım Üniversitesi.
  • Hossain, F., Anagnostou, E. N., Bagtzoglou, A. C., 2005. On Latin Hypercube sampling for efficient uncertainty estimation of satellite rainfall observations in flood prediction, 32, 776-792.
  • Wan, Y., Lv, Z. Z., Yuan, X. K., 2008. Latin Hypercube sampling and updated Latin Hypercube sampling method for reliability sensitivity and its variance analysis, 30(6), 927-934.
  • Wei, Q., Lu, W. X., Tian, Z. J., 2004. Application of Monte-Carlo to annual precipitation forecast. Journal of Arid Land Resources and Environment, 4(18), 144-146.
  • Jin, C., 2005. Study on random number generator and random sampling in Monte Carlo method. Dalian University of Technology.
  • Zhou, Y., 1997. Study of Sampling Methods in Monte Carlo Simulation. Structure Environment Engineering, (3), 14-18.
  • Şen, Z., 2004. Yapay Sinir Ağları İlkeleri, Su Vakfı Yayınları, İstanbul.
  • Mahouti, P., 2019. Design optimization of a pattern reconfigurable microstrip antenna using differential evolution and 3D EM simulation-based neural network model. International Journal of RF and Microwave Computer-Aided Engineering, 29(8).
There are 34 citations in total.

Details

Primary Language Turkish
Subjects Electrical Engineering
Journal Section Research Articles
Authors

Ahmet Uluslu 0000-0002-5580-1687

Publication Date June 30, 2022
Submission Date October 19, 2021
Acceptance Date February 19, 2022
Published in Issue Year 2022 Volume: 10 Issue: 2

Cite

APA Uluslu, A. (2022). FARKLI ÖRNEKLEME YÖNTEMLERİNİN YAPAY SİNİR AĞLARINDA PERFORMANSININ KARŞILAŞTIRILMASI. Mühendislik Bilimleri Ve Tasarım Dergisi, 10(2), 412-428. https://doi.org/10.21923/jesd.1012106