-
本文中研究一种基底-高折射率光栅-金属光栅的透射式滤光片结构,如图 1所示。金属光栅层材料为Al,折射率为nm,色散曲线依照Drude模型[9-11]。高折射率介质光栅层折射率nd=2.4,基底层折射率ns=1.5,Λ代表光栅的周期,光栅的宽度W=Λf,f为光栅的占空比,金属光栅层的厚度为dm,介质光栅的厚度为dd。当TM光从光栅表面垂直入射,金属表面发生等离子体共振。在金属和基底之间增加高折射率介质光栅层,满足波导共振激发条件,发生导模共振。由于这两种效应的共同作用,在特定波长发生透射增强。
2013年,ZHOU[2]提出当结构中的金属光栅厚度dm=0.06μm,介质光栅厚度dd=0.08μm,占空比f=0.75,周期Λ=0.4μm时,光谱峰值位置波长为0.65μm,透射效率为67%。为了提高光栅结构滤光片对波长为0.65μm红光的透射效率,缩小带宽,本文中提出利用PSO算法,逆优化结构参量。
在PSO中,群体的每个粒子被认为是N维搜索空间中的一个无质量的点,粒子仅具有两个属性:位置X和速度v,位置代表粒子移动的方向,速度代表粒子移动的快慢。每个粒子在搜索空间中寻找最优解,将其记为当前个体极值Pm,并将个体极值与整个粒子群里的其他粒子共享,找到最优的那个个体极值作为整个粒子群的当前全局最优解G,粒子群中的所有粒子根据自己找到的当前个体极值Pm和整个粒子群共享的当前全局最优解G来调整自己的速度和位置,直到输出全局最优解[12-14]。速度和位置的迭代公式为:
$\begin{array}{l} {\mathit{\boldsymbol{v}}_{m, k + 1}} = \mathit{\boldsymbol{\omega }}{\mathit{\boldsymbol{v}}_{m, k}} + {c_1}{\rm{ran}}{{\rm{d}}_1}\left( {} \right)\left( {{\mathit{\boldsymbol{P}}_\mathit{m}} - {\mathit{\boldsymbol{X}}_{m, k}}} \right) + \\ \;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;{c_2}{\rm{ran}}{{\rm{d}}_2}\left( {\mathit{\boldsymbol{G - }}{\mathit{\boldsymbol{X}}_{m, k}}} \right) \end{array} $
(1) ${\mathit{\boldsymbol{X}}_{m, k + 1}} = {\mathit{\boldsymbol{X}}_{m, k}} + {\mathit{\boldsymbol{v}}_{m, k + 1}}\Delta \mathit{t} $
(2) 式中,ω是惯性权重,设定为从0.9~0.4递减,c1和c2设为1.49,Δt为时间步长,设为1,参量设置参见参考文献[15],rand( )为MATLAB自带函数,返回一个0~1之间的随机数,当前个体极值Pm={p1m, p2m, …, pnm},n=20, G=Pm,Pm和G迭代规则为:如果F(Xm, k+1)比F(Pm)小,则Pm由Xm, k+1取代。如果F(Pm)比F(G)小,则G被Pm取代。直到迭代结束。
对于通过PSO设计的金属光栅滤光片,本文中选择透射波长0.65μm的红光,选择4个参量作为优化参量:光栅厚度dm,介质层厚度dd,光栅周期Λ和占空比f。因此,粒子群数组为X={Λ,dm,dd,f}。设定目标函数,求适应值F公式如下[16]:
$F = {\left\{ {\frac{1}{M}\sum\limits_{{\mathit{\lambda }_\mathit{i}}} {{{\left[ {{\mathit{R}_{\rm{t}}}\left( \mathit{\lambda } \right) - {R_{\rm{d}}}\left( \mathit{\lambda } \right)} \right]}^2}} } \right\}^{1/2}} $
(3) 式中,Rt是不同波长入射光期望达到的透射效率,为了实现红光透射效率高、旁带低的要求,这里设定波长为0.65μm的透射效率达到最大值,此时最大值为1,带宽控制在小于50nm,设定波长为0.625μm和0.675μm时透射效率为0.3,其它波长对应透射效率设为0, 以减小旁带。Rd是把PSO中的优化参量代入RCWA程序计算所得到的透射效率,M是所取波长点的数量。粒子的F为两者的方差,反映了优化的程度,F越小,Rt越接近Rd,就越符合优化的结果。
基于粒子群优化算法的透射滤光片设计
Design of transmittance filters based on particle swarm optimization algorithm
-
摘要: 为了优化器件的结构参量,把光栅结构的周期、高度、占空比作为优化的粒子,采用粒子群优化算法和严格耦合波分析算法,对粒子群中粒子的适应值进行了比较,对金属-介质光栅滤光片结构进行了理论分析和仿真优化,在一定范围内找到最优参量。结果表明,根据MATLAB仿真结果进行优化,得到周期为0.300μm,金属光栅厚度为0.035μm,介质光栅厚度为0.400μm,光栅占空比为0.77;在TM光垂直入射时,该结构对波长为0.65μm的红光透射率达到80.27%,旁带透射效率不超过15%;该结构实现了特定波长光的高效透射,从而实现了滤光。该结构为亚波长光栅的设计、制备研究和实际应用提供了参考。Abstract: In order to optimize the structure parameters of the device, the period, height and duty ratio of grating structure were taken as the optimized particles. By using particle swarm optimization algorithm and rigorous coupled wave analysis algorithm, fitness values of the particles in particle swarm were compared. After theoretical analysis and simulative optimization of the structure of metal-medium grating filter, the optimal parameters were found in a certain range. The results show that, according to MATLAB simulation results, structure parameters are optimized:period is 0.300μm, the thickness of metal grating is 0.035μm, the thickness of dielectric grating is 0.400μm and grating duty ratio is 0.77. When TM light is vertically incident, the transmission rate of structure to red light in 0.65μm wavelength is 80.27% and the transmission rate of the side bands is less than 15%. This structure achieves high transmission to specific wavelength light and then achieves light filtering. This structure provides the reference for the design, fabrication and practical application of sub-wavelength gratings.
-
Key words:
- gratings /
- filtering /
- particle swarm optimization algorithm /
- transmission efficiency
-
Figure 4. Relationship between transmission spectra of grating structure[2], the optimal structure and the target and wavelength
-
[1] YE Y, ZHOU Y, ZHANG H, et al. Polarizing color filter based on micron metal grating[J].Acta Optica Sinica, 2011, 31(4):0405003(in Chinese). doi: 10.3788/AOS [2] ZHOU Y. Design and Fabricate the color filter and the polarizer based on the subwavelength grating[D]. Suzhou: Soochow University, 2013: 10-28(in Chinese). [3] ZHOU W Ch, WU Y H, HAO P, et al. Transmission bandpass filters based on two-dimensional subwavelength metallic gratings[J]. Acta Optica Sinica, 2013, 33(11):1105001(in Chinese). doi: 10.3788/AOS [4] KONG W J, ZHENG B B, YUN M J, et al. Guided-mode resonance filter with narrow waveband or three primary colors[J]. Acta Optica Sinica, 2011, 31(10):1005006(in Chinese). doi: 10.3788/AOS [5] WANG S S, MAGUSSON R. Theory and applications of guided-mode resonance filter[J]. Applied Optics, 1993, 32(14):2606-2613. doi: 10.1364/AO.32.002606 [6] ZHENG H Y, HU F R, Design of narrowband guided-mode resonance filters in visible wavelength region[J]. Laser Technology, 2016, 40(1):118-121(in Chinese). [7] SHOKOOH-SAREMI M, MAGNUSSON R. Particle swarm optimization and its application to the design of diffraction grating filters[J]. Optics Letters, 2007, 32(8):893-894. [8] SHOKOOH-SAREMI M, MAGNUSSON R. Properties of two-dimensional resonant reflectors with zero-contrast gratings[J]. Optics Letters, 2014, 39(24):6958-6961. doi: 10.1364/OL.39.006958 [9] ORDAL M A, LONG L L, BELL R J, et al. Optical properties of the metals, Al, Co, Cu, Au, Fe, Pb, Ni, Pd, Pt, Ag, Ti and W in the infrared and far infrared[J]. Appied Optics, 1983, 22(7):1099-1120. doi: 10.1364/AO.22.001099 [10] ZHANG G P, YE J X, LI Z G. Coupled-wave analysis of anti-reflective properties of binary phase grating[J]. Journal of Optoelectronics·Laser, 1996, 7(2):98-102(in Chinese). [11] SONG D J, XIE K, XIAO J. Mode field and dispersion analysis of photonic crystal fiber based on finite element method[J]. Laser Technology, 2012, 36(1):111-114(in Chinese). [12] TANG G J, WANG X L. Parameter optimized variational mode decomposition method with application to incipient fault diagnosis of rolloing bearing[J]. Journal of Xi'an Jiaotong University, 2015, 49(5):73-80(in Chinese). [13] DAI C, WANG Y Q, XU F. 3-D lidar echo decomposition based on particle swarm optimization[J]. Laser Technology, 2016, 40(2):284-287(in Chinese). [14] YANG L Y. Research on vehicle routing problem based on modified particle swarm optimization algorithm[D]. Kaifeng: Henan University, 2011: 10-28(in Chinese). [15] ROBINSON J, RAHMAT-SAMⅡ Y. Particle swarm optimization in electromagnetics[J].IEEE Transactions on Antennas and Propagation, 2004, 52(2):397-407. doi: 10.1109/TAP.2004.823969 [16] ZHAO J R, YU Zh H, DU L. Particle swarm optimization algorithms in synthesis of fiber gratings for the non-dispersion filters in the optical communication[J]. Journal of Southern Yangtze University (Natural Science Edition), 2007, 6(1):69-71(in Chinese). [17] DING Q L, ZHOU Y, YE Y, et al. Reflection characteristics of metal-dielectric-metal reflective-type color filter[J]. Journal of Applied Optics, 2012, 33(4):694-696(in Chinese). [18] MA W T, ZHOU J, HUANG Sh P, et al. Characteristic of subwavelength dielectric grating with metal layer and its sensing applications[J]. Chinese Journal of Lasers, 2011, 38(9):0905008(in Chinese). doi: 10.3788/CJL [19] BAI L F, HAN J, ZHANG Y, et al. Registration algorithm of infrared and visible images based on improved gradient normalized mutual information and particle swarm optimization[J]. Infrared and Laser Engineering, 2012, 41(1):250-252(in Chinese).