CN113268911A - Structural parameter optimization method for chiral super-structure surface and micro-nano device - Google Patents
Structural parameter optimization method for chiral super-structure surface and micro-nano device Download PDFInfo
- Publication number
- CN113268911A CN113268911A CN202110688282.9A CN202110688282A CN113268911A CN 113268911 A CN113268911 A CN 113268911A CN 202110688282 A CN202110688282 A CN 202110688282A CN 113268911 A CN113268911 A CN 113268911A
- Authority
- CN
- China
- Prior art keywords
- chiral
- individual
- polarized light
- circularly polarized
- population
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Genetics & Genomics (AREA)
- Computing Systems (AREA)
- Physiology (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a structural parameter optimization method of a chiral super-structure surface and a micro-nano device, wherein the method comprises the following steps: determining a parameter space of the structural parameters based on the optimization range of the structural parameters, and determining an initial population according to the parameter space of the structural parameters; acquiring the transmissivity of left-handed circularly polarized light and right-handed circularly polarized light of chiral superstructure surfaces corresponding to each individual in a population under a target wavelength by using a finite difference time domain algorithm, and calculating the difference value; determining the fitness of each individual based on the acquired transmissivity difference corresponding to each individual; selecting, crossing and mutating the individuals in the initial population based on the fitness of each individual, and generating an optimized population; and selecting the corresponding individual as the optimal structure parameter of the chiral superstructure surface when the transmittance difference of the left-handed circularly polarized light and the right-handed circularly polarized light under the target wavelength is the minimum value from the optimal population. By adopting the method, the optimal structure parameters of the chiral superstructure surface can be rapidly and accurately obtained in a huge parameter space.
Description
Technical Field
The invention relates to the technical field of electronic devices, in particular to a structural parameter optimization method of a chiral super-structure surface and a micro-nano device.
Background
A substance that cannot coincide with its mirror image, either by rotation or translation, is considered chiral, and its mirror images are referred to as two opposite chiral enantiomers. Chirality is widely found in nature and is used in life sciences, sensing, imaging and biomedicine; for example, a chiral biomolecule may exhibit good drug therapeutic effects, while another enantiomer may be highly toxic. In addition, the chirality of the geometric structure can be reflected in the optical chiral response at the same time, i.e. the absorption of the left-handed and right-handed circularly polarized light by the chiral substance is different. The chiral optical response of naturally chiral molecules is often weak and difficult to detect, and therefore artificial chiral structures need to be designed to enhance the optical response of naturally chiral molecules.
The nanostructured surface is widely noticed because of its advantages of arbitrarily adjusting optical response, enhancing response, etc., and can be used to enhance the response intensity of chiral molecules. In the process of designing a metamaterial surface, structural parameters of the metamaterial surface are generally required to be optimized, the traditional modes comprise a manual searching mode and a parameter scanning mode, the optimal solutions can be searched in a very small parameter space, the required time exponentially increases with the increase of the parameter space, the searching accuracy rate also decreases, and therefore the method cannot be used under a complex optimization target. Therefore, how to efficiently and accurately optimize the structural parameters of the chiral superstructure surface to obtain the desired chiral response is an urgent technical problem to be solved.
Disclosure of Invention
In view of the above, the invention provides a structural parameter optimization method for a chiral nanostructure surface and a micro-nano device, so as to solve one or more problems in the prior art.
According to one aspect of the invention, a method for optimizing structural parameters of a chiral nanostructured surface is disclosed, the method comprising:
determining a parameter space of the structural parameters based on the optimization range of the structural parameters, and determining an initial population according to the parameter space of the structural parameters; the structural parameter is at least one of a distance parameter between two adjacent layers of metal structural bodies, a length and width parameter of each metal body, an interval parameter of adjacent stacked structures and a period number parameter of the metal structural bodies in the chiral superstructure surface;
acquiring the transmissivity of left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population under the target wavelength by using a time domain finite difference algorithm, and calculating the transmissivity difference of the left-handed circularly polarized light and the right-handed circularly polarized light corresponding to each individual;
determining the fitness of each individual based on the calculated transmissivity difference corresponding to each individual;
carrying out selection operation, cross operation and variation operation on the individuals in the initial population based on the fitness of each individual, and generating an optimized population;
and selecting the corresponding individual with the minimum fitness as the optimal structure parameter of the chiral superstructure surface from the optimized population.
In some embodiments of the present invention, the,
obtaining the transmissivity of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population under the target wavelength by using a time domain finite difference algorithm, and calculating the transmissivity difference of the left circularly polarized light and the right circularly polarized light corresponding to each individual, wherein the transmissivity difference comprises the following steps:
calculating the transmissivity of left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual in a target waveband range by using a time domain finite difference algorithm, and calculating the average value of the transmissivity difference value of the left-handed circularly polarized light and the right-handed circularly polarized light corresponding to each individual; wherein the target wavelength is a center wavelength of the target waveband;
determining the fitness of each individual based on the acquired transmittance difference corresponding to each individual, wherein the fitness comprises the following steps:
and determining the fitness of each individual based on the transmittance difference value of the chiral superstructure surface corresponding to each individual at the central wavelength of the target waveband and the average value of the transmittance difference values in the target waveband.
In some embodiments of the present invention, the target wavelength is 1035nm, and the target wavelength band ranges from 1030nm to 1040 nm.
In some embodiments of the present invention, selecting, from the optimized population, an individual corresponding to the minimum fitness as an optimal structural parameter of the chiral superstructure surface includes:
and selecting the corresponding individual as the optimal structure parameter of the chiral superstructure surface when the transmittance difference value of the left-handed circularly polarized light and the right-handed circularly polarized light under the target wavelength is the minimum value and the average value of the transmittance difference values in the target wave band is smaller than a preset value from the optimized population.
In some embodiments of the present invention, performing a selection operation, a crossover operation, and a mutation operation on the individuals in the initial population based on the fitness of each individual, and generating an optimized population, includes:
acquiring cross probability and variation probability;
selecting individuals for crossing from the initial population based on the fitness of each individual;
performing cross operation on the two selected cross individuals according to the cross probability;
and carrying out mutation operation on individuals in the population according to the mutation probability.
In some embodiments of the invention, the selection method used to select individuals for crossover from the initial population is a roulette method.
In some embodiments of the present invention, the,
the cross probability is self-adaptive cross probability, and the calculation formula is as follows:
wherein, PCTo cross probability, k1And k3Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fcFor a greater fitness of the two individuals selected for crossover,
the average fitness of individuals in the population.
In some embodiments of the present invention, the mutation probability is an adaptive mutation probability, and the calculation formula is:
wherein, PmAs the mutation probability, k2And k4Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fiThe fitness of the variant individual is shown as the fitness,
the average fitness of individuals in the population.
In some embodiments of the present invention, obtaining transmittance of left circularly polarized light and right circularly polarized light of a chiral superstructure surface corresponding to each individual in the initial population at a target wavelength by using a finite difference time domain algorithm, and calculating a transmittance difference between the left circularly polarized light and the right circularly polarized light corresponding to each individual, includes:
acquiring transmission spectrums with the wavelengths between 700nm and 1400nm corresponding to all individuals in the initial population by using finite difference time domain analysis software, and acquiring circular dichroism spectrums according to the transmission spectrums;
and acquiring the transmittance difference value of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual under the target wavelength based on each circular dichroic spectrum.
According to another aspect of the invention, the invention further discloses a super-structured surface micro-nano device with stronger chiral optical response, which comprises a glass substrate and a double-layer chiral metal structure body positioned on the surface of the glass substrate, wherein the structural parameters of the double-layer chiral metal structure body adopt the optimal structural parameters selected by the structural parameter optimization method based on any one of the embodiments.
Through the embodiment, the structural parameter optimization method of the chiral superstructure surface combines the adaptive genetic algorithm and the time domain finite difference algorithm, and can quickly and accurately select the optimal structural parameters in a huge parameter space based on the determined parameter space and the optimization target so as to optimize the spectral response.
In addition, the method adaptively adjusts the cross probability and the mutation probability, further improves the convergence speed and the optimization effect, and reduces the time required by the optimization process.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
It will be appreciated by those skilled in the art that the objects and advantages that can be achieved with the present invention are not limited to the specific details set forth above, and that these and other objects that can be achieved with the present invention will be more clearly understood from the detailed description that follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. For purposes of illustrating and describing some portions of the present invention, corresponding parts of the drawings may be exaggerated, i.e., may be larger, relative to other components in an exemplary apparatus actually manufactured according to the present invention. In the drawings:
fig. 1 is a schematic structural diagram of a chiral nanostructured surface according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a chiral metal structure according to an embodiment of the invention.
Fig. 3 is a schematic structural diagram of a left-handed enantiomer and a right-handed enantiomer in accordance with an embodiment of the present invention.
Fig. 4 is a schematic flow chart of a method for optimizing structural parameters of a chiral nanostructured surface according to an embodiment of the present invention.
Fig. 5 is a schematic flow chart of a method for optimizing structural parameters of a chiral nanostructured surface according to another embodiment of the present invention.
Fig. 6 is a circular dichroic spectrum simulation diagram of a left-handed structure and a right-handed structure obtained by optimization of a general genetic algorithm according to an embodiment of the present invention.
Fig. 7 is a circular dichroic spectrum simulation diagram of a left-handed structure and a right-handed structure obtained by optimization of an adaptive genetic algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
It should be noted that, in order to avoid obscuring the present invention with unnecessary details, only the structures and/or processing steps closely related to the scheme according to the present invention are shown in the drawings, and other details not closely related to the present invention are omitted.
It should be emphasized that the term "comprises/comprising/comprises/having" when used herein, is taken to specify the presence of stated features, elements, steps or components, but does not preclude the presence or addition of one or more other features, elements, steps or components.
For optimization problems of structural parameters of a nanostructured surface, numerical simulations are usually required. The numerical simulation is to simulate the response of the structural parameters by using numerical simulation software based on a time domain finite difference algorithm or a finite element method to verify a guess, and the guess, the adjustment and the verification of possible structures need to be artificially performed according to a physical principle. The parameter scanning is based on numerical simulation, the structural parameters are not adjusted manually, but the structural parameters are automatically adjusted in a certain parameter interval by using a computer, the consumption of time is not counted, all possible solutions are calculated once by repeated iteration according to the specified optimized parameters and precision, and then the optimal solution is selected. Therefore, how to guarantee that a better solution is obtained in a shorter time in a complex problem and a huge parameter space is a problem to be solved by the scheme.
In order to solve the problems, the invention provides a method for efficiently optimizing the parameters of the ultrastructural surface structure, which combines a genetic algorithm and a time domain finite difference algorithm, can search an optimal solution in a parameter space according to a set parameter range and an optimization target, adopts a search mode to simulate the Darwinian evolutionary theory, and reserves the optimal solution in a population through the processes of selection, intersection, variation and the like, thereby obtaining a reliable result in a short time. Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same reference numerals denote the same or similar parts, or the same or similar steps.
Fig. 4 is a schematic flow chart of a method for optimizing structural parameters of a chiral nanostructured surface according to an embodiment of the present invention, and as shown in fig. 4, the method for optimizing structural parameters of a chiral nanostructured surface according to the embodiment includes the following steps S10 to S50.
Step S10: determining a parameter space of the structural parameters based on the optimization range of the structural parameters, and determining an initial population according to the parameter space of the structural parameters; the structural parameter is at least one of a distance parameter between two adjacent layers of metal structural bodies, a length and width parameter of each metal body, an interval parameter of adjacent stacked structures and a period number parameter of the metal structural bodies in the chiral superstructure surface.
The structural parameters refer to corresponding parameters of the chiral structure on the chiral superstructure surface, and specifically refer to size parameters of the chiral structure. The optimization range of the structural parameters refers to the value range corresponding to each preset parameter; for example, when the structural parameter includes a width, the preset width value range is 30nm to 90nm, and the parameter space refers to all solutions between 30nm and 90 nm; when the selected precision is 1nm during optimization, the parameter space corresponding to the width includes solutions of 30nm, 31nm, 33nm, … 89nm and 90nm, for example. For another example, when the structural parameter includes a length parameter in addition to a width parameter, if the preset length value range is 100nm to 230nm, the possible solutions corresponding to the length parameter are 100nm, 101nm … 229nm, and 230 nm; in this case, solutions corresponding to the width parameter and the length parameter may be arranged and combined into 7800(60 × 130) solution vectors, and the size of the parameter space at this time is also 7800. During specific optimization, the initial population can be determined through a parameter space of the structural parameters, during specific optimization, partial solutions or solution vectors in the parameter space can be directly selected to form the initial population, and all solutions in the parameter space can be selected to generate the initial population.
Step S20: and acquiring the transmissivity of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population under the target wavelength by using a time domain finite difference algorithm, and calculating the transmissivity difference of the left circularly polarized light and the right circularly polarized light corresponding to each individual.
In the step, the finite difference time domain algorithm can be specifically an FDTD algorithm, namely, the modeling simulation of the super-structure surface is carried out through FDTD software; the transmittance difference, that is, the Circular Dichroism (CD) value, is a difference between the transmittance when left-handed circularly polarized light is incident and the transmittance when right-handed circularly polarized light is incident. Because the initial population has a plurality of individuals, each individual is used for reflecting one structural parameter of the super-structural surface, the transmission spectrum and the CD spectrum corresponding to each different individual obtained through FDTD software simulation are different.
In one embodiment of the invention, the surface of the nanostructure to be optimized is a periodic double-layer structure based on a Born-Kuhn model, and the gold nanorods on the upper layer and the lower layer are in an angle stacking form and have C4Rotational symmetry, linear birefringence can be avoided, thereby avoiding unnecessary polarization conversion. In this embodiment, a periodic structure of the super-structured surface is shown in fig. 1, and it can be seen from fig. 1 that, after left-handed circularly polarized light and right-handed circularly polarized light are incident into the periodic structure, the periodic structure absorbs the two circularly polarized lights differently, resulting in different transmittances of the two emitted circularly polarized lights, which is called Circular Dichroism (CD), and is defined as follows: CD ═ TLCP-TRCP. Wherein, TLCPRepresents the transmittance of incident left-handed circularly polarized light, and TRCPAnd the transmittance of the right-handed circularly polarized light when the right-handed circularly polarized light enters is shown, and the CD is the transmittance difference value of the left-handed circularly polarized light and the right-handed circularly polarized light.
FIG. 2 is a schematic structural diagram of the metal structure of the chiral nanostructured surface shown in FIG. 1, and FIG. 3 shows structural diagrams of the left-handed enantiomer (LH) and the right-handed enantiomer (RH) of the chiral structure; as can be seen from fig. 2 and 3, the structural parameters of the super-structured surface having the double-layer structure include: the distance D between the upper layer of gold nanorods and the lower layer of gold nanorods, the length L of the nanorods, the width W of the nanorods, and the gap G between the adjacent stacked structures. In addition to the above, the structural parameters of the optimized metamaterial surface may also include the period P. It should be understood that the structural parameters of the nanostructured surface can be defined by specific application scenarios, and in specific optimization, the structural parameters can be one or more of a distance parameter between two adjacent layers of metal structures, a length and a width parameter of each metal body, an adjacent stacked structure spacing parameter, and a period number parameter of the metal structures in the chiral nanostructured surface.
Further, the optimization range of the structure parameter of this embodiment is shown in table 1, the precision is selected to be 1nm during optimization, in order to prevent the conflict between the period and the structure, that is, the size of the unit period structure does not exceed the period set during simulation, the scheme introduces a parameter p related to the period to calculate the period, which is exemplary: the value range of p is [20,80], and then p belongs to [20,80 ]. The algorithm replaces the optimization period by optimizing a P parameter, wherein the P parameter is the total length of the substrate on the surface of the chiral structure, and the calculation formula is as follows: p ═ 2(L + P) + G; wherein L is the length of the gold nanorods, p is the period number, and G is the distance between adjacent stacked structures.
TABLE 1 optimized Range of structural parameters of a nanostructured surface
| D(nm) | L(nm) | W(nm) | G(nm) | P(nm) | |
| Minimum value | 20 | 100 | 30 | 20 | 260 |
| Maximum value | 140 | 230 | 90 | 70 | 690 |
In addition, since the parameter space size refers to all possible solutions at a given precision, the formula can be expressed as:
wherein S represents the size of the parameter space, M represents the number of structural parameters to be optimized, precision represents precision, and precision is selected to be 1nm in this embodiment, so the size of the parameter space in the structural parameter optimization range corresponding to table 1 is: s120 × 130 × 60 × 50 × 60 ═ 2.808 × 109。
Step S30: and determining the fitness of each individual based on the acquired transmittance difference corresponding to each individual.
In the step, a transmission spectrum corresponding to incidence of left-handed circularly polarized light and a transmission spectrum corresponding to incidence of right-handed circularly polarized light are obtained through FDTD software modeling simulation, the wavelength range of the transmission spectrum is 700nm to 1400nm, and then the transmittance difference value of the left-handed circularly polarized light and the right-handed circularly polarized light is calculated based on the obtained transmission spectrum. Wherein, the target wavelength may be a wavelength corresponding to 1035nm, and the fitness of each individual in the initial population is a transmittance difference between left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual at the wavelength 1035 nm.
In this embodiment, the optimization objective is: the CD value of the chiral super-structured surface corresponding to the optimal structure parameter under the wavelength of 1035nm is minimum, and is expressed by the following formula: opt. minF ═ CD1035(ii) a Wherein, CD1035Refers to the CD value at a wavelength of 1035nm, and opt. minF represents the value at which the CD value at a wavelength of 1035nm is at a minimumThe corresponding individual. It should be understood that under this optimized condition, the fitness of each individual is related to the transmittance difference.
In another embodiment, in addition to the above optimization condition, there is a second constraint condition that an average value of transmittance differences of left-handed circularly polarized light and right-handed circularly polarized light of the chiral meta-structure surface corresponding to each individual in the target wavelength band range is smaller than a predetermined value; in this embodiment, the fitness of each individual is related to the transmittance difference corresponding to the target wavelength and also related to the average value of the transmittance difference in the target wavelength band, and the fitness of each individual is determined based on the obtained transmittance difference corresponding to each individual, that is, the fitness of each individual is determined based on the transmittance difference of the chiral superstructure surface corresponding to each individual at the center wavelength of the target wavelength band and the average value of each transmittance difference in the target wavelength band. Accordingly, the method comprises the steps of: and calculating the transmissivity of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual in the target waveband range by using a time domain finite difference algorithm, and calculating the average value of the transmissivity difference values of the left circularly polarized light and the right circularly polarized light corresponding to each individual. Similarly, the finite difference time domain algorithm is specifically an FDTD algorithm.
In this embodiment, the target wavelength is the center wavelength of the target band, which may range from 1030nm to 1040nm when the target wavelength is specifically 1035nm wavelength. In addition, the average value of the transmittance difference values of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual within the target waveband range may be specifically 0.4, where the condition is expressed by a formula: s.t.average { | CDi|i∈[1030,1040)}>0.4; wherein, CDiRepresents the CD value at the i wavelength.
And step S40, selecting, crossing and mutating the individuals in the initial population based on the fitness of each individual, and generating an optimized population.
In this step, the selection operation is to select a plurality of pairs of preferred individuals from the initial population, and the selection method adopted can be a roulette method. Before selection, individuals in the population can be ranked according to the fitness, and obviously, the probability of selecting the individuals with higher fitness is higher. In addition, the roulette selection method may discard better individuals during the selection process, so an elite mechanism, i.e., a mechanism in which better individuals in the initial population are directly selected into the next generation population, may also be used during the selection process.
In addition, in the selection process of the parent individuals, the chance of each individual being selected should not be equal, because the good individuals have a greater chance of being selected to ensure that the children are better, and the bad individuals should not be completely excluded, which may also result in good children. Thus, the roulette selection method may determine the chance of selecting an individual based on the fitness of the individual, with the probability of each individual being selected being:
wherein M is the population size, f (x)i) Indicates the fitness of the ith individual,
representing the sum of fitness of M individuals within the population. The cumulative probability of each individual in the population being selected can be further calculated based on the probability of each individual in the population being selected:
when selecting specifically, it can be first [0,1 ]]A random number n is generated if n<q1Then select individual 1, otherwise select one wherein q is satisfiedi-1<n<qiI, the process is repeated M times.
Furthermore, after a plurality of pairs of better individuals are selected by adopting a roulette method, further performing cross operation on each pair of better individuals, wherein the cross process can adopt a single-point cross method or other types of cross methods; and then carrying out mutation operation on individuals needing mutation. After the selection, the crossing and the variation, a new optimized population is generated.
In an embodiment of the present invention, the selecting, crossing, and mutating the individuals in the initial population based on the fitness of each individual, and generating the optimized population may specifically include the following steps: acquiring cross probability and variation probability; selecting individuals for crossing from the initial population based on the fitness of each individual; performing cross operation on the two selected cross individuals according to the cross probability; and carrying out mutation operation on individuals in the population according to the mutation probability.
Step S50: and selecting the corresponding individual with the minimum fitness as the optimal structure parameter of the chiral superstructure surface from the optimized population.
From this step, it can be seen that the optimization conditions of the structural parameters of the chiral nanostructured surface are: under the optimal structure parameter, the fitness corresponding to the optimal structure parameter is minimum, in other words, the difference value of incidence rates of the left-handed circularly polarized light and the right-handed circularly polarized light of the chiral superstructure surface corresponding to the optimal structure parameter under the target wavelength is the minimum value (the absolute value of the corresponding difference value is maximum). In addition, after the optimal structure parameters of the chiral superstructure surface are obtained by optimization based on the method of the embodiment of the invention, the desired chiral response can be further obtained based on the chiral structure surface corresponding to the optimal structure parameters.
For the structural parameter optimization method of the chiral superstructure surface of the above embodiment, in the optimization process, if fixed cross probability and variation probability are adopted, both good individuals and poor individuals need to undergo cross and variation operations with the same probability, which may cause two problems: (1) it is unfair to use the same cross probability and variation probability for high-quality and poor-quality individuals; because for good individuals, the probability of cross mutation should be reduced so that the good individuals can be stored as much as possible; for the inferior individuals, the cross mutation probability should be increased to change the inferior conditions as much as possible, and the constant cross mutation probability affects the efficiency of the algorithm. (2) The same probability can not well meet the requirement in the population evolution process, the population needs higher crossover and variation probability at the initial stage of iteration so as to achieve the purpose of quickly searching the optimal solution, and the population needs smaller crossover and variation probability at the later stage of convergence so as to help the population to quickly converge after the optimal solution is searched.
Therefore, to further solve the above problem, in another embodiment of the present invention, the cross probability and/or mutation probability is adaptively adjusted based on the algebra of the population. Specifically, the cross probability is calculated by the following formula:
wherein, PCTo cross probability, k1And k3Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fcFor a greater fitness of the two individuals selected for crossover,
the average fitness of individuals in the population.
The calculation formula of the variation probability is as follows:
wherein, PmAs the mutation probability, k2And k4Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fiThe fitness of the variant individual is shown as the fitness,
the average fitness of individuals in the population.
For the embodiment, in the optimization process, the cross probability and the mutation probability are adaptively adjusted, so that the optimization effect can be further effectively improved, and the time spent in the optimization process is further reduced. FIG. 6 is a circular dichroism spectrum simulation diagram of a left-handed structure and a right-handed structure obtained by a general Genetic Algorithm (GA) optimization according to an embodiment of the present invention; FIG. 7 is a circular dichroism spectrum simulation of a left-handed structure and a right-handed structure obtained by Adaptive Genetic Algorithm (AGA) optimization according to an embodiment of the present invention; for the simulation result, on the premise that the precision is 10nm, the optimal solution obtained by adopting a common scanning algorithm is that the CD value at the wavelength of 1035nm is more than-0.4, and the optimization process can be completed within about 15 days; on the premise that the precision is 1nm, the minimum CD value of an optimization result obtained by adopting an AGA combined FDTD optimization algorithm at a wavelength of 1035nm can reach-0.52, and the process can be completed within about three days. Compared with an optimization algorithm combining GA and FDTD, the adoption of AGA can accelerate the convergence speed, further save the optimization time (GA needs about twice time), and further improve the optimization effect.
The chiral response of the nanostructured surface shown in the above embodiment is strong, and the interaction between light and a substance in the chiral nanostructured surface can be analyzed through a harmonic mode, which is of great significance for effective structural parameter design and deep understanding of the interaction between the chiral nanostructured surface and a chiral molecule in practical applications such as biosensing. The structural parameter optimization method for the chiral superstructure surface in the above embodiment can be specifically used for solving the structural parameter design problem of the chiral superstructure surface, that is, the problem of optimizing the spectral response by optimizing the structural parameters. By combining the genetic algorithm and the FDTD algorithm, the search parameter space can be enlarged, and the time required for obtaining an optimized result is saved; to further increase the speed of convergence and the optimization effect, the crossover probability P is adaptively adjustedcAnd the mutation probability PmThe number of the population is changed, so that the number of the individuals in the previous generations is large, and the number of the population in the later generations is small. Simulation results show that the algorithm for optimizing the chiral response of the ultrastructural surface provided by the invention can improve the optimization effect, reduce the time required by optimization and finally obtain a satisfactory optimization result.
According to another aspect of the invention, the invention further provides a super-structure surface micro-nano device with stronger chiral optical response. The micro-nano device comprises a glass substrate and a double-layer chiral metal structure body positioned on the surface of the glass substrate, wherein the double-layer chiral metal structure body can be realized by stacking gold nanorods at an angle through an upper layer and a lower layer, the gold nanorods are covered by a layer of dielectric material, the refractive index of the dielectric material is 1.3, the dielectric constant of the gold nanorods is obtained according to Johnson and Christy measurement, and the double-layer chiral metal structure body is positioned on the surface of the glass substrate. For the micro-nano device, the structural parameters of the double-layer chiral metal structure body are the optimal structural parameters selected by adopting the structural parameter optimization method disclosed by any one of the embodiments.
In order that those skilled in the art will better understand the present invention, embodiments of the present invention will be described below with reference to specific examples.
In this embodiment, the development kit lumopt is a python-based inverse design kit, and the simulation software uses FDTD software. In order to reduce the time for obtaining the optimization result and improve the reliability of the optimization result, the cross probability and the variation probability are adaptively adjusted in the optimization process; specifically, fig. 5 is a schematic flow chart of a method for optimizing structural parameters of a chiral nanostructured surface according to another embodiment of the present invention, as shown in fig. 5, the specific operation steps of this embodiment are as follows:
(1) and initializing parameters. Randomly initializing a population, wherein each individual in the population comprises a value of a structural parameter to be optimized, the value of each structural parameter is in a given range, and the given range is also the preset optimization range of the structural parameter; initializing a population size K and initializing a cross probability PcAnd the mutation probability Pm。
(2) Simulating the structure parameters of each individual in the initialized population by adopting FDTD software, obtaining the CD spectrum between 700nm and 1400nm corresponding to each individual, and obtaining the CD according to the formula opt1035And s.t.average { | CDi|i∈[1030，1040)}>And 0.4, evaluating the fitness of each individual and obtaining the fitness of each individual. And finally, carrying out binary Gray coding on the individuals in the population so as to facilitate the crossover and mutation operations.
(3) Selecting individuals for crossover from the parent by using a roulette selection method; wherein, the parent refers to the initial population.
(4) And performing cross operation on the selected parent to obtain a new child. Each pair of parents has PcCross over the probability of (c); in this step, the cross probability PcIs initially set in step (1).
(5) Performing mutation operations on the generated offspring. Each offspring generated has PmI.e. one or more bits on the binary code are changed. Wherein the mutation probability PmAlso initially set in step (1).
(6) Judging whether the current algebra is a preset termination algebra; if yes, switching to the step (7); if not, the step (8) is switched to.
(7) And finishing the optimization and outputting an optimization result.
(8) Adjusting the size of the population and adaptively adjusting the cross probability PcAnd the mutation probability Pm. Because the simulation of each individual by FDTD requires 3 to 5 minutes, in order to obtain reliable results in a short time, requirements on the evolution generation number and the population number are required, the embodiment adopts a larger population number in the first generation of the algorithm to ensure the diversity of the population, and adopts a smaller population number in the later generation, so that the total time cost can be saved. Meanwhile, in order to improve the convergence precision of the genetic algorithm and accelerate the convergence speed, the cross probability P is adaptively adjustedcAnd the mutation probability Pm. Cross probability PcThe self-adaptive adjustment formula is as follows:
in the above formula, k1、k2、k3And k4Are all constants of 0 to 1.
(9) The generated offspring is mixed with part of the parent, and the binary gray code is decoded into decimal to generate a new population. And taking the new population as a final output result or a parent of the next circulation until the set algebra is met and outputting an optimization result.
Through the embodiment, the method for optimizing the structural parameters of the chiral superstructure surface combines a genetic algorithm and a finite difference time domain algorithm, and can automatically acquire the optimal solution of the structural parameters in a parameter space according to the preset structural parameter preset range and the optimization target, so as to obtain the optimal structural parameters; the optimization method is adopted to simulate the Darwin evolutionary theory, and the optimal solution in the population is reserved through the processes of selection, crossing, variation and the like; in addition, the cross probability and the variation probability are adaptively changed according to the similarity of individuals in the population, so that the convergence precision is improved, the convergence speed is accelerated, the number of the population is also changed along with the change of the genetic algebra, so that the balance between the population diversity and the simulation time is ensured, and the method can obtain a reliable result in a short time.
Those of ordinary skill in the art will appreciate that the various illustrative components, systems, and methods described in connection with the embodiments disclosed herein may be implemented as hardware, software, or combinations of both. Whether this is done in hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Features that are described and/or illustrated with respect to one embodiment may be used in the same way or in a similar way in one or more other embodiments and/or in combination with or instead of the features of the other embodiments in the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for optimizing structural parameters of a chiral nanostructured surface, the method comprising:
determining a parameter space of the structural parameters based on the optimization range of the structural parameters, and determining an initial population according to the parameter space of the structural parameters; the structural parameter is at least one of a distance parameter between two adjacent layers of metal structural bodies, a length and width parameter of each metal body, an interval parameter of adjacent stacked structures and a period number parameter of the metal structural bodies in the chiral superstructure surface;
acquiring the transmissivity of left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population under the target wavelength by using a time domain finite difference algorithm, and calculating the transmissivity difference of the left-handed circularly polarized light and the right-handed circularly polarized light corresponding to each individual;
determining the fitness of each individual based on the calculated transmissivity difference corresponding to each individual;
carrying out selection operation, cross operation and variation operation on the individuals in the initial population based on the fitness of each individual, and generating an optimized population;
and selecting the corresponding individual with the minimum fitness as the optimal structure parameter of the chiral superstructure surface from the optimized population.
2. The method for optimizing structural parameters of a chiral nanostructured surface according to claim 1,
obtaining the transmissivity of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population under the target wavelength by using a time domain finite difference algorithm, and calculating the transmissivity difference of the left circularly polarized light and the right circularly polarized light corresponding to each individual, wherein the transmissivity difference comprises the following steps:
calculating the transmissivity of left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual in a target waveband range by using a time domain finite difference algorithm, and calculating the average value of the transmissivity difference value of the left-handed circularly polarized light and the right-handed circularly polarized light corresponding to each individual; wherein the target wavelength is a center wavelength of the target waveband;
determining the fitness of each individual based on the calculated transmittance difference value corresponding to each individual, wherein the fitness comprises the following steps:
and determining the fitness of each individual based on the transmittance difference value of the chiral superstructure surface corresponding to each individual at the central wavelength of the target waveband and the average value of the transmittance difference values in the target waveband.
3. The method for optimizing structural parameters of a chiral nanostructured surface according to claim 2, wherein said target wavelength is 1035nm and said target wavelength band is 1030nm to 1040 nm.
4. The method for optimizing the structural parameters of the chiral superstructure surface according to claim 2, wherein the selecting the individual corresponding to the minimum fitness from the optimized population as the optimal structural parameters of the chiral superstructure surface comprises:
and selecting the corresponding individual as the optimal structure parameter of the chiral superstructure surface when the transmittance difference value of the left-handed circularly polarized light and the right-handed circularly polarized light under the target wavelength is the minimum value and the average value of the transmittance difference values in the target wave band is smaller than a preset value from the optimized population.
5. The method for optimizing structural parameters of a chiral nanostructured surface according to claim 1, wherein the steps of performing selection, crossover and mutation operations on individuals in the initial population based on the fitness of each individual, and generating an optimized population comprise:
acquiring cross probability and variation probability;
selecting individuals for crossing from the initial population based on the fitness of each individual;
performing cross operation on the two selected cross individuals according to the cross probability;
and carrying out mutation operation on individuals in the population according to the mutation probability.
6. The method for optimizing structural parameters of a chiral nanostructured surface according to claim 5, characterized in that the selection method used for selecting individuals for crossing from the initial population is the roulette method.
7. The method for optimizing structural parameters of a chiral nanostructured surface according to claim 5,
the cross probability is self-adaptive cross probability, and the calculation formula is as follows:
wherein, PCTo cross probability, k1And k3Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fcFor selecting two for crossingIs determined by the individual's fitness,
the average fitness of individuals in the population.
8. The method for optimizing the structural parameters of the chiral nanostructured surface according to claim 5, wherein the variation probability is an adaptive variation probability, and the calculation formula is as follows:
wherein, PmAs the mutation probability, k2And k4Is a constant of 0 to 1, fmaxIs the maximum fitness of individuals in the population, fiThe fitness of the variant individual is shown as the fitness,
the average fitness of individuals in the population.
9. The method for optimizing the structural parameters of the chiral superstructure surface of claim 1, wherein the method for obtaining the transmittance of left-handed circularly polarized light and right-handed circularly polarized light of the chiral superstructure surface corresponding to each individual in the initial population at a target wavelength by using a finite difference time domain algorithm and calculating the transmittance difference of the left-handed circularly polarized light and the right-handed circularly polarized light corresponding to each individual comprises:
acquiring transmission spectrums with the wavelengths between 700nm and 1400nm corresponding to all individuals in the initial population by using finite difference time domain analysis software, and acquiring circular dichroism spectrums according to the transmission spectrums;
and acquiring the transmittance difference value of the left circularly polarized light and the right circularly polarized light of the chiral superstructure surface corresponding to each individual under the target wavelength based on each circular dichroic spectrum.
10. A micro-nano device based on a chiral super-structured surface comprises a glass substrate and a double-layer chiral metal structure body positioned on the surface of the glass substrate, and is characterized in that the structural parameters of the double-layer chiral metal structure body adopt the optimal structural parameters selected by the structural parameter optimization method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110688282.9A CN113268911B (en) | 2021-06-21 | 2021-06-21 | Structural parameter optimization method for chiral super-structure surface and micro-nano device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110688282.9A CN113268911B (en) | 2021-06-21 | 2021-06-21 | Structural parameter optimization method for chiral super-structure surface and micro-nano device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113268911A true CN113268911A (en) | 2021-08-17 |
| CN113268911B CN113268911B (en) | 2022-06-17 |
Family
ID=77235678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110688282.9A Active CN113268911B (en) | 2021-06-21 | 2021-06-21 | Structural parameter optimization method for chiral super-structure surface and micro-nano device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113268911B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008088830A2 (en) * | 2007-01-16 | 2008-07-24 | Evolved Nanomaterial Sciences, Inc. | Chiral separating agents with active support |
| CN112904469A (en) * | 2021-01-28 | 2021-06-04 | 暨南大学 | Random polarization state polarizing device based on dielectric nano brick super-structured surface |
-
2021
- 2021-06-21 CN CN202110688282.9A patent/CN113268911B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008088830A2 (en) * | 2007-01-16 | 2008-07-24 | Evolved Nanomaterial Sciences, Inc. | Chiral separating agents with active support |
| US20080314835A1 (en) * | 2007-01-16 | 2008-12-25 | Regina Valluzzi | Chiral separating agents with active support |
| CN112904469A (en) * | 2021-01-28 | 2021-06-04 | 暨南大学 | Random polarization state polarizing device based on dielectric nano brick super-structured surface |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113268911B (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Tackling photonic inverse design with machine learning | |
| Khatib et al. | Deep learning the electromagnetic properties of metamaterials—a comprehensive review | |
| WO2018146683A1 (en) | Method and system for characterizing a nanostructure by machine learning | |
| CN114815496B (en) | Pixelated optical proximity effect correction method and system applied to super-resolution lithography | |
| Prabhat et al. | Deep neural networks-based denoising models for CT imaging and their efficacy | |
| On et al. | Optimal design of microwave absorber using novel variational autoencoder from a latent space search strategy | |
| CN113268911B (en) | Structural parameter optimization method for chiral super-structure surface and micro-nano device | |
| WO2023098053A1 (en) | Neural network-based apparatus parameter acquisition method and system, and related component | |
| Malkiel et al. | Machine learning for nanophotonics | |
| Liu et al. | A hybrid strategy for the discovery and design of photonic nanostructures | |
| CN111461303B (en) | Digital core reconstruction method and system based on generation of antagonistic neural network | |
| CN114611387A (en) | Method and device for improving measurement precision of electromagnetic characteristics of equipment | |
| Zhang et al. | Compression of deep neural networks: bridging the gap between conventional-based pruning and evolutionary approach | |
| Fontbonne et al. | Comparison of methods for end-to-end co-optimization of optical systems and image processing with commercial lens design software | |
| JP2007115256A (en) | Method for designing electro-optic image processing system, storage medium containing instruction causing computer to execute this method, and optical subsystem | |
| CN114943866B (en) | Image classification method based on evolutionary neural network structure search | |
| CN116595479A (en) | Community discovery method, system, equipment and medium based on graph double self-encoder | |
| CN116503320A (en) | Hyperspectral image anomaly detection method, hyperspectral image anomaly detection device, hyperspectral image anomaly detection equipment and readable storage medium | |
| CN116009246A (en) | Polarization optical system automatic optimization design method based on deep learning | |
| CN115616762A (en) | Super-surface design method and device and electronic equipment | |
| Xu et al. | Dual-branch deep image prior for image denoising | |
| CN112419169A (en) | CNN medical CT image denoising method based on noise prior | |
| Çarpınlıoğlu et al. | Genetically programmable optical random neural networks | |
| US20240176228A1 (en) | Method and device for optimizing mask parameters | |
| CN112836816B (en) | Training method suitable for crosstalk of photoelectric storage and calculation integrated processing unit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |