Abstract
The inverse design of nanophotonic devices has been widely concerned by researchers, and on-demand design is the difficulty of inverse design. In inverse design, researchers usually define a target spectrum based on the performance indicators and experiences and then inverse design the structural parameters from the target spectrum. Due to the uncertainty of inverse design and “one-to-many” problem, it is not usually possible to guarantee that the target spectrum is sure to correspond to a real nanostructure. In order to solve these problems, an inverse design method combining generative model and genetic algorithm is proposed in this paper. Before the inverse design, the real spectrum is compressed into a latent space by the generation model, and then, the target spectrum is decoded from the latent space according to the performance index. Finally, the hybrid optimization algorithm combining genetic algorithm and forward prediction network is used to optimize the generated spectrum. The design method follows the process from performance indicators to target spectrum to structural parameter, and we successfully realized the inverse design of multilayer nanofilms on demand by using this method in the experimental part. The inverse design method proposed in this paper provides a possible solution for the inverse design of nanophotonic devices on demand.
Similar content being viewed by others
Data Availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
Hell SW, Wichmann J (1994) Breaking the diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy. Opt Lett 19(11):780–782
Gan Z, Cao Y, Evans RA, Gu M (2013) Three-dimensional deep sub-diffraction optical beam lithography with 9 nm feature size. Nat Commun 4(1):2061
Lang C, LaNasa JA, Utomo N, Xu Y, Nelson MJ, Song W, Hickner MA, Colby RH, Kumar M, Hickey RJ (2019) Solvent-non-solvent rapid-injection for preparing nanostructured materials from micelles to hydrogels. Nat Commun 10(1):3855
Jahani S, Jacob Z (2016) All-dielectric metamaterials. Nat Nanotechnol 11(1):23–36
Mokkapati S, Beck FJ, Waele RD, Polman A, Catchpole KR (2011) Resonant nano-antennas for light trapping in plasmonic solar cells. J Phys D Appl Phys 44(18):185101
Pelton M, Aizpurua J, Bryant G (2008) Metal-nanoparticle plasmonics. Laser Photonics Rev 2(3):136–159
Khorasaninejad M, Shi Z, Zhu AY, Chen WT, Sanjeev V, Zaidi A, Capasso F (2017) Achromatic metalens over 60 nm bandwidth in the visible and metalens with reverse chromatic dispersion. Nano Lett 17(3):1819–1824
Arbabi A, Horie Y, Bagheri M, Faraon A (2015) Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission. Nat Nanotechnol 10(11):937–943
Peurifoy J, Shen Y, Jing L, Yang Y, Cano-Renteria F, Delacy BG, Joannopoulos JD, Tegmark M, Soljačić M (2018) Nanophotonic particle simulation and inverse design using artificial neural networks. Sci Adv 4(6):1–7
Baxter J, Calà A, Lesina JMG, Weck A, Berini P (2019) Plasmonic colours predicted by deep learning. Sci Rep 9(1):8074
Unni R, Yao K, Zheng Y (2020) Deep convolutional mixture density network for inverse design of layered photonic structures. ACS Photonics 7(10):2703–2712
Liu DJ, Tan YX, Khoram E, Yu ZF (2018) Training deep neural networks for the inverse design of nanophotonic structures. ACS Photonics 5(4):1365–1369
So S, Yang Y, Lee T, Rho J (2021) On-demand design of spectrally sensitive multiband absorbers using an artificial neural network. Photonics Res 9(4):B153–B158
Deng Y, Ren S, Fan K, Malof JM, Padilla WJ (2021) Neural-adjoint method for the inverse design of all-dielectric metasurfaces. Opt Express 29(5):7526–7534
Ma W, Cheng F, Xu Y, Wen Q, Liu Y (2019) Probabilistic representation and inverse design of metamaterials based on a deep generative model with semi-supervised learning strategy. Adv Mater 31(35):1901111
Kong W, Chen J, Huang Z, Kuang D (2021) Bidirectional cascaded deep neural networks with a pretrained autoencoder for dielectric metasurfaces. Photonics Res 9(8):1607–1615
Fu J, Yang Z, Liu M, Zhang H, Zhang Y (2023) Highly-efficient design method for coding metasurfaces based on deep learning. Opt Commun 529:129043
Qian Y, Ni B, Feng Z, Ni H, Zhou X, Yang L, Chang J (2023) Deep learning for the design of random coding Metasurfaces. Plasmonics. https://doi.org/10.1007/s11468-023-01919-5
Zhan T, Liu QS, Sun YJ, Qiu L, Wen T, Zhang R (2022) A general machine learning-based approach for inverse design of one-dimensional photonic crystals toward targeted visible light reflection spectrum. Opt Commun 510:127920
Ghosh A, Pal A, Das NR (2020) An approach to design photonic crystal gas sensor using machine learning. Optik 208:163997
Peralta-Ángeles JA, Reyes-Esqueda JA (2022) Analytically supported hybrid photonic–plasmonic crystal design using artificial neural networks. Plasmonics 17:1501–1525
Li X, Ning S, Liu Z, Yan Z, Luo C, Zhuang Z (2020) Designing phononic crystal with anticipated band gap through a deep learning based data-driven method. Comput Methods Appl Mech Eng 361:112737
Yuan M, Yang G, Song S, Zhou L, Minasian R, Yi X (2022) Inverse design of a nano-photonic wavelength demultiplexer with a deep neural network approach. Opt Express 30(15):26201–26211
Zhang T, Wang J, Liu Q, Zhou J, Dai J, Han X, Zhou Y, Xu K (2019) Efficient spectrum prediction and inverse design for plasmonic waveguide systems based on artificial neural networks. Photonics Res 7(3):368–380
Tranter AD, Slatyer HJ, Hush MR, Leung AC, Everett JL, Paul KV, Vernaz-Gris P, Lam PK, Buchler BC, Campbell GT (2018) Multiparameter optimisation of a magneto-optical trap using deep learning. Nat Commun 9(1):4360
Chugh S, Gulistan A, Ghosh S, Rahman BMA (2019) Machine learning approach for computing optical properties of a photonic crystal fiber. Opt Express 27(25):36414–36425
Zibar D, Brusin AMR, de Moura UC, Da Ros F, Curri V, Carena A (2019) Inverse system design using machine learning: the Raman amplifier case. J Lightwave Technol 38(4):736–753
Kim N, Lee D, Hong Y (2023) Data-efficient deep generative model with discrete latent representation for high-fidelity digital materials. ACS Mater Lett 5(3):730–737
Ma W, Cheng F, Liu Y (2018) Deep-learning-enabled on-demand design of chiral metamaterials. ACS Nano 12(6):6326–6334
An S, Zheng B, Tang H, Shalaginov MY, Zhou L, Li H, Kang M, Richardson KA, Gu T, Hu J (2021) Multifunctional metasurface design with a generative adversarial network. Adv Opt Mater 9(5):2001433
Hong Y, Nicholls DP (2022) Data-driven design of thin-film optical systems using deep active learning. Opt Express 30(13):22901–22910
Naseri P, Hum S (2021) A generative machine learning-based approach for inverse design of multilayer metasurfaces. IEEE Trans Antennas Propag 69(9):5725–5739
Liu Z, Zhu D, Rodrigues SP, Lee KT, Cai W (2018) Generative model for the inverse design of metasurfaces. Nano Lett 18(10):6570–6576
Yan R, Wang T, Jiang X, Zhong Q, Huang X, Wang L, Yue X (2020) Design of high-performance plasmonic nanosensors by particle swarm optimization algorithm combined with machine learning. Nanotechnology 31(37):375202
Qiu C, Wu X, Luo Z, Yang H, He G, Huang B (2021) Nanophotonic inverse design with deep neural networks based on knowledge transfer using imbalanced datasets. Opt Express 29(18):28406–28415
Chen J, Ding W, Li XM, Xi X, Ye KP, Wu HB, Wu RX (2021) Absorption and diffusion enabled ultrathin broadband metamaterial absorber designed by deep neural network and PSO. IEEE Antennas Wirel Propag Lett 20(10):1993–1997
Yu R, Liu Y, Zhu L (2022) Inverse design of high degree of freedom meta-atoms based on machine learning and genetic algorithm methods. Opt Express 30(20):35776–35791
Han C, Zhang B, Wang H, Xu J, Ding J (2022) Predicting the eigenstructures of metamaterials with QR-code meta-atoms by deep learning. Opt Lett 47(7):1863–1866
Funding
This work is supported by the National Natural Science Foundation of China (61967007, 61963016, 62366015) and the Key Research and Development Program of Jiangxi Province (20201BBF61012).
Author information
Authors and Affiliations
Contributions
All authors contributed to the conception and design of the study. Data set collection and simulation analysis were performed by (Yue Li) and (Danlong Zong), the first draft of the paper was written by (Yue Li) and (Lu Zhu), the experimental part was performed by (Zhikang Yang), and the revision and integration of the paper was performed by (Yuanyuan Liu). All authors have commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethics Approval and Consent to Participate
Not applicable.
Consent for Publication
Not applicable.
Competing Interests
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhu, L., Li, Y., Yang, Z. et al. An On-Demand Inverse Design Method for Nanophotonic Devices Based on Generative Model and Hybrid Optimization Algorithm. Plasmonics (2023). https://doi.org/10.1007/s11468-023-02075-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11468-023-02075-6