Koncept i primena mašinskog učenja u predviđanju induktivnosti u višeslojnim pravougaonim spiralnim mikrozavojnicama

Ključne reči: pravougaona planarna mikrozavojnica, višeslojna planarna zavojnica, induktivnost, mašinsko učenje, sintetički skup podataka

Sažetak


Uvod/cilj: Ovim istraživanjem uvodi se nov pristup projektovanju skupa podataka pravougaonih planarnih zavojnica pomoću komplementarnih softverskih alata. MATLAB funkcioniše kao okruženje za projektovanje visokog nivoa, a FastHenry deluje kao računarski okvir za rešavanje Maksvelovih jednačina i dobijanje vrednosti induktivnosti. Generišu se dva različita sintetička skupa podataka pomoću naprednih tehnika uzorkovanja za različite konfiguracije, uključujući metodu uzorkovanja latinske hiperkocke. Ovi skupovi podataka se zatim obrađuju i obučavaju pomoću algoritama mašinskog učenja za predviđanje vrednosti induktivnosti na osnovu dobijenih geometrijskih parametara.

Metode: Za generisanje ekstenzivnih sintetičkih skupova podataka koji sadrže 20 000 redova za dvoslojne konfiguracije zavojnica i 15 000 redova za troslojne konfiguracije prvo se koristi MATLAB. Nakon procesa generisanja, proverava se da li su skupovi podataka spremni za obuku. Šest modela mašinskog učenja: Gaussian Process Regressor (GPR), KNeighborsRegressor (KNN), BayesianRidge, ElasticNetCV, GammaRegressor, kao i Bagging Regressor obučeno je i procenjeno pomoću metrika kao što su R² i RMSE. Modeli se zatim ispituju na nepoznatim podacima za ispitivanje i ocenjuju pomoću tehnike unakrsne validacije kako bi se utvrdilo koliko mogu da generalizuju.

Rezultati: Skupovi podataka su uspešno generisani, a modeli KNeighborsRegressor, Gaussian Process Regressor (GPR) i Bagging Regressor ostvarili su najbolje rezultate, iskazali su veliku tačnost i malu grešku.

Zaključak: Rezultati pokazuju da je mašinsko učenje praktičan i efikasan metod za predviđanje induktivnosti u višeslojnim pravougaonim planarnim zavojnicama na osnovu geometrije.

Biografije autora

Benazzouz younes, University of Oran 2 Mohamed Ben Ahmed,Industrial Maintenance and Safety Institute (IMSI)

Benazzouz Younes

Second-Year PhD Candidate Laboratory of Production Engineering and Industrial Maintenance (LGPMI) Industrial Maintenance and Safety Institute (IMSI) Mohamed Ben Ahmed University of Oran 2
Guendouz djilalia, University of Oran 2 Mohamed Ben Ahmed,Industrial Maintenance and Safety Institute (IMSI)

Laboratory of Production Engineering and Industrial Maintenance (LGPMI),

Reference

Alghrairi, M., Sulaiman, N., Mutashar, S., Wan Hasan, W. Z., Jaafar, H., & Algriree, W. (2022). Designing and analyzing multi-coil multi-layers for wireless power transmission in stent restenosis coronary artery. AIP Advances, 12(12). Available at: https://doi.org/10.1063/5.0121532

Altman, N. S. (1992). An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression. The American Statistician, 46(3), 175–185. Available at: https://doi.org/10.1080/00031305.1992.10475879

Ansari, M.A., & Agarwal, P. (2024). Inductance Modelling of Planar Meander Structure Using RBM and kNN. SN Computational Sciences, 5, 1169. Available at: https://doi.org/10.1007/s42979-024-03516-7

Asadi, M., Rezaei, A., & Abazari, A. M. (2023, December). Calculation of Mutual Inductance between Two Planar Coils with Custom Specifications and Positions Using a Machine Learning Approach. In Proceedings of the International Conference on New Trends in Applied Sciences (Vol. 1, pp. 20-30). Available at: https://doi.org/10.58190/icontas.2023.50

Bayazıt, G. H., Caarls, E. I., & Lomonova, E. A. (2023). Inductance Map Regression of Doubly Excited Electrical Machines Considering Cross-Saturation. IEEE Transactions on Magnetics, 59(11), 1-5. Available at: https://doi.org/10.1109/TMAG.2023.3292500

Chen, X., Zheng, C., & Golestanirad, L. (2023). Application of Machine learning to predict RF heating of cardiac leads during magnetic resonance imaging at 1.5 T and 3 T: A simulation study. Journal of Magnetic Resonance, 349, 107384. Available at: https://doi.org/10.1016/j.jmr.2023.107384

Ding, W., Wang, Y., Chen, T., Lu, Z., You, Y., Wang, J., & Huang, Z. (2025). A stacking machine‐learning based method for accelerating magnetic coupler design with ferrite cores in inductive power transfer applications. International Journal of Circuit Theory and Applications, 53(1), 67-80. Available at: https://doi.org/10.1002/cta.4096

Farooq, M., Amin, B., Elahi, A., Wijns, W., & Shahzad, A. (2023). Planar Elliptical Inductor Design for Wireless Implantable Medical Devices. Bioengineering, 10(2), 151. Available at: https://doi.org/10.3390/bioengineering10020151

Faria, A. R. S., Marques, L. S., Gaspar, J., Alves, F. S., & Cabral, J. M. N. (2021, March). High precision, geometry independent analytical method for self-inductance calculation in planar coils. In 2021 22nd IEEE International Conference on Industrial Technology (ICIT) (Vol. 1, pp. 1234-1239). IEEE. Available at: http://doi.org/10.1109/ICIT46573.2021.9453559

Gastineau, L. A., Lewis, D. D., & Ionel, D. M. (2024, November). Combined 3D FEA and Machine Learning Design of Inductive Polyphase Coils for Wireless EV Charging. In 2024 13th International Conference on Renewable Energy Research and Applications (ICRERA) (pp. 1811-1816). IEEE. Available at: https://doi.org/10.1109/ICRERA62673.2024.10815533

MacKay, D. J. (1992). Bayesian interpolation. Neural Computation, 4(3), 415-447. Available at: https://doi.org/10.1162/neco.1992.4.3.415

Noroozi, B., & Morshed, B. I. (2021). Design and optimization of printed spiral coils for wireless passive sensors. IET Wireless Sensor Systems, 11(4), 169-178. Available at: https://doi.org/10.1049/wss2.12019

SIDUN, Aleksandr, SRIVASTAVA, Manish, O’DONOGHUE, Kilian, et al. Planar On-Silicon Inductor Design for Electromagnetic Tracking. IEEE Sensors Journal, 2023. Available at: https://doi.org/10.1109/JSEN.2023.3296471

Stillig, J., Parspour, N., Ewert, D., & Jung, T. J. (2022, July). Feasibility Study on Machine Learning-based Method for Determining Self-and Mutual Inductance. In 2022 International Seminar on Intelligent Technology and Its Applications (ISITIA) (pp. 193-198). IEEE. Available at: https://doi.org/10.1109/ISITIA56226.2022.9855321

Tolpygo, S. K., Golden, E. B., Weir, T. J., & Bolkhovsky, V. (2022). Mutual and self-inductance in planarized multilayered superconductor integrated circuits: Microstrips, striplines, bends, meanders, ground plane perforations. IEEE Transactions on Applied Superconductivity, 32(5), 1-31. Available at: https://doi.org/10.1109/TASC.2022.3162758

Wang, F., Zhang, X., Shokoueinejad, M., Iskandar, B. J., Webster, J. G., & Medow, J. E. (2018). Spiral planar coil design for the intracranial pressure sensor. Medical Devices & Sensors, 1(3), e10012. Available at: https://doi.org/10.1002/mds3.10012

Wheeler, H. A. (1928). Simple inductance formulas for radio coils. Proceedings of the Institute of Radio Engineers, 16(10), 1398-1400. Available at: https://doi.org/10.1109/JRPROC.1928.221309

Williams, C. K., & Rasmussen, C. E. (2006). Gaussian processes for machine learning (Vol. 2, No. 3, p. 4). Cambridge, MA: MIT press.

Wu, D., Lin, S., Chen, H., & Wang, X. (2023). Mutual inductance model of double printed circuit board‐based coplanar rectangular spiral coils for eddy‐current testing. International Journal of Numerical Modelling: Electronic Networks, Devices and Fields, 36(1), e3038. Available at: https://doi.org/10.1002/jnm.3038

Zhang, J., Hou, X., Qian, S., Bi, X., Hu, D., Liu, J., ... & Chou, X. (2024). Flexible multilayer MEMS coils and their application in energy harvesters. Science China Technological Sciences, 67(4), 1282-1293. Available at: https://doi.org/10.1007/s11431-023-2474-9

Zhao, J. (2010). A new calculation for designing multilayer planar spiral inductors. EDN (Electrical Design News, 55(14), 37.

Objavljeno
2025/12/17
Rubrika
Originalni naučni radovi