DEVELOPMENT OF A HYBRID INVERSE ANALYSIS MODEL FOR EVALUATING SPECTRAL CHARACTERISTICS OF MULTILAYERED STRUCTURES
DOI:
https://doi.org/10.35546/kntu2078-4481.2025.1.2.3Keywords:
multilayer structures, spectral analysis, mathematical modeling, hybrid model, iterative methods, deep learning, neural networksAbstract
Multilayer structures are key elements in modern optics, nanotechnology and photonics, where their spectral characteristics determine the efficiency and performance of devices. However, modern analysis methods have a number of limitations, such as low accuracy and insufficient noise immunity, which complicates work with complex systems.The aim of this study is to develop a new hybrid inverse analysis model that combines classical iterative methods and deep neural networks. The proposed model uses the advantages of pre-training of neural networks for fast initialization of the parameters of multilayer structures and iterative methods for their optimization.In the course of the work, an algorithm and a corresponding software product were created, which were implemented in Python and tested on synthetic data with noise. The model was implemented using the NumPy, SciPy, Matplotlib libraries, as well as TensorFlow and Keras for building and training deep neural networks. This approach ensured efficient data processing, high accuracy of results, and the ability to adapt to various experimental conditions. The results showed that the model provides high accuracy in restoring spectral parameters even under conditions of significant noise levels. This is confirmed by low values of the mean square error and high coefficient of determination, which exceed the results of traditional approaches. In addition, the model turned out to be adaptive to changes in the geometry of the layers and optical properties, and its use allowed to reduce the number of iterations due to pre-training of deep neural networks.The application of the developed model is promising for spectroscopy, development of optical coatings, sensors and photonic devices. Its flexibility allows working with small training samples, and the ability to adapt to noise expands the analysis capabilities. Further improvement of the algorithm, including optimization of neural networks and expansion of the training base, can significantly expand the scope of its application and provide even higher accuracy.
References
Xuyang, L., Hongle, A., Wensheng, C., Xueguang, S. (2024). Deep learning in spectral analysis: Modeling and imaging. Trends in Analytical Chemistry, 172, Article 117612. https://doi.org/10.1016/j.trac.2024.117612
Mishra, P., Passos, D., Marini, F., Xu, J., Amigo, M. J., Gowen, et al. (2022). Deep learning for near-infrared spectral data modelling: Hypes and benefits. Trends in Analytical Chemistry, 157(2), 10.1016/s0731-7085(01), Article 00690-2. https://doi.org/10.1016/j.trac.2022.116804
Schuetzke, J., Szymanski, N. J., Reischl, M. (2023). Validating neural networks for spectroscopic classification on a universal synthetic dataset. npj Comput Mater, 9(100). https://doi.org/10.1038/s41524-023-01055-y
Primrose, M. S., Giblin, J., Smith, C., Anguita, M. R., Weedon, G. H. (2022). One dimensional convolutional neural networks for spectral analysis. Algorithms, Technologies, and Applications for Multispectral and Hyperspectral Imaging XXVIII, 120940C. https://doi.org/10.1117/12.2618487
Liu, J., Osadchy, M., Ashton, L., Foster, M., Solomone, C. J., Gibson, S. J. (2017). Deep convolutional neural networks for Raman spectrum recognition: a unified solution. The Analyst, 21. https://doi.org/10.1039/C7AN01371J.
Madden, M. G., Ryder, A. G. (2002). Machine learning methods for quantitative analysis of Raman spectroscopy data. Proc. SPIE 4876, Opto-Ireland 2002: Optics and Photonics Technologies and Applications. https://doi.org/ 10.1117/12.464039
Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O., Walsh, A. (2018). Machine learning for molecular and materials science. Nature, 559, 547–555. https://doi.org/10.1038/s41586-018-0337-2
Mahani, M. K., Chaloosi, M., Maragheh, M. G., Khanchi, A. R., Afzali, D. (2007). Comparison of Artificial Neural Networks with Partial Least Squares Regression for Simultaneous Determinations by ICP-AES. Chinese Journal of Chemistry, 25(11), 1658–1662. https://doi.org/10.1002/cjoc.200790306
Marini, F., Bucci, R., Magrì, A. L., Magrì, A. D. (2008). Artificial neural networks in chemometrics: History, examples and perspectives. Microchemical Journal, 88(2), 178–185. https://doi.org/10.1016/j.microc.2007.11.008
Born, M., & Wolf, E. (1999). Principles of Optics: Electromagnetic Theory of Propagation, Interference, and Diffraction of Light. Cambridge University Press. https://doi.org/10.1017/CBO9781139644181
Vassallo, E., Laguardia, L., Catellani, M., Cremona, A., Dellera, F., Ghezzi, F. (2007). Characterization of Poly(3‐ Methylthiophene)‐like Films Produced by Plasma Polymerization. Plasma Processes and Polymers, 4, S801–S805. https://doi.org/10.1002/ppap.200731909
Han, K., Chang, C. H. (2014). Numerical Modeling of Sub-Wavelength Anti-Reflective Structures for Solar Module Applications. Nanomaterials, 4(1), 87–128. https://doi.org/10.3390/nano4010087
Zhou, W., Cao, Y., Zhao, H., Li, Z., Feng, P., & Feng, F. (2022). Fractal Analysis on Surface Topography of Thin Films: A Review. Fractal and Fractional, 6(3), 135. https://doi.org/10.3390/fractalfract6030135
Zhang, S., Deng, Q., Ding, Z. (2022). Multilayer graph spectral analysis for hyperspectral images. EURASIP Journal on Advances in Signal Processing. https://doi.org/10.1186/s13634-022-00926-8
Palacz, M. (2018). Spectral Methods for Modelling of Wave Propagation in Structures in Terms of Damage Detection – A Review. Applied Sciences, 8(7), 1124. https://doi.org/10.3390/app8071124
Xuyang, L., Hongle, A., Wensheng, C., Xueguang, S. (2024). Deep learning in spectral analysis: Modeling and imaging. TrAC Trends in Analytical Chemistry, 172, Article 117612. https://doi.org/10.1016/j.trac.2024.117612
Parrein, P., Moussy, N., Poupinet, L., Gidon, P. (2009). Multilayer structure for a spectral imaging sensor. Applied Optics, 48, 653–657. https://doi.org/10.1364/AO.48.000653
Koleva, M. N., Vulkov, L. G. (2024). Numerical Solution of External Boundary Conditions Inverse Multilayer Diffusion Problems. Symmetry, 16(10), 1396. https://doi.org/10.3390/sym16101396
Nishawala, V. V., Ostoja-Starzewski, M., Porcu, E., Shen, L. (2020). Random Fields with Fractal and Hurst Effects in Mechanics. In: Altenbach, H., Öchsner, A. (eds) Encyclopedia of Continuum Mechanics. Springer. https://doi.org/ 10.1007/978-3-662-55771-6_74
Terven, J., Cordova-Esparza, D. M., Ramirez-Pedraza, A., Chávez Urbiola, E. (2023). Loss Functions and Metrics in Deep Learning. A Review. https://doi.org/10.48550/arXiv.2307.02694
Paschotta, R. (2025). Fresnel Equations. RP Photonics Encyclopedia. Retrieved January 19, 2025. https://doi.org/ 10.61835/tql
Zhang, J. (2019). Gradient Descent Based Optimization Algorithms for Deep Learning Models Training. arXiv. Retrieved January 19, 2025. https://doi.org/10.48550/arXiv.1903.03614
Downey, A. B. (2017). Modeling and Simulation in Python. Green Tea Press. Needham, Massachusetts. 228 p. https://greenteapress.com/wp/modsimpy/ https://allendowney.github.io/ModSimPy/
Shuaibov, A., Minya, A., Malinina, A., Gritsak, R., Malinin, A., Bilak, Yu., Vatrala, M. (2022). Characteristics and Plasma Parameters of the Overstressed Nanosecond Discharge in Air between an Aluminum Electrode and a Chalcopyrite Electrode (CuInSe2). Surface Engineering and Applied Electrochemistry, 58(4), 369–385. https://doi.org/10.3103/ S1068375522040123
Bondar, I., Suran, V., Minya, O., Shuaibov, O., Bilak, Yu., Shevera, I., Malinina, A., Krasilinets, V. (2023). Synthesis of surface structures during laser-stimulated evaporation of a copper sulfate solution in distilled water. Ukrainian Journal of Physics, 68(2), 138. https://doi.org/10.15407/ujpe68.2.138
Shuaibov, O., Hrytsak, R., Minya, O., Malinina, A., Shevera, I., Bilak, Yu., Homoki, Z. (2024). Conditions for pulsed gas-discharge synthesis of thin tungsten oxide films from a plasma mixture of air with tungsten vapors. Physics and Chemistry of Solid State, 25(4), 684–688. https://doi.org/10.15330/pcss.25.4.684-688
Shuaibov, O. K., Hrytsak, R. V., Minya, O. I., Malinina, A. A., Bilak, Yu. Yu., & Homoki, Z. T. (2022). Spectroscopic diagnostics of overstressed nanosecond discharge plasma between zinc electrodes in air and nitrogen. Journal of Physical Studies, 26(2), 2501(8 p.). https://doi.org/10.30970/jps.26.2501
Shuaibov, O., Minya, O., Hrytsak, R., Malinina, A., Malinin, O., Bilak, Y., & Homoki, Z. (2024). Conditions for the deposition of selenium thin films from the plasma of an overvoltaged nanosecond discharge. J. Clin Bio Med Adv, 3(1), 01-09.
Kozubovsky, V. R., & Bilak, Yu. Yu. (2022). Express analysis of gas mixtures using a spectral correlator based on the Fabry–Perot interferometer. Journal of Applied Spectroscopy, 89(3), 495–499. https://doi.org/10.1007/ s10812-022-01385-7
