A Parametric Co-optimization Design Method for Optical Waveguide Thermal Insulation Structures

Abstract

This invention belongs to the field of integrated optoelectronics technology, specifically relating to a parameter co-optimization design method for optical waveguide thermal insulation structures. The method includes the following steps: Step S1: Structure discretization and parameter definition; Step S2: Constructing a performance model under multi-dimensional parameters; Step S3: Determining the effective design solution space; Step S4: Co-optimizing parameters based on secondary optimization objectives; Step S5: Iteration and structural reconstruction. This invention provides a systematic and automated optimization framework. Designers only need to define the parameters to be optimized, performance objectives, and secondary optimization criteria, eliminating the need to guess parameter change paths based on experience, greatly reducing design difficulty and improving design efficiency.

Description

Technical Field

[0001] This invention belongs to the field of integrated optoelectronics technology, specifically relating to a parameter co-optimization design method for optical waveguide thermal insulation structures. Background Technology

[0002] Thermally adiabatic waveguide structures are key fundamental components in modern integrated photonic chips. Their core design goal is to achieve efficient and low-loss optical mode evolution or power distribution within the shortest possible device length. For example, thermally adiabatic tapered waveguides are used to change mode dimensions, and thermally adiabatic directional couplers are used to achieve broadband, fault-tolerant power splitting.

[0003] The core of thermal insulation design is to meet the "thermal insulation condition," which requires that when a light wave propagates in the structure, its local mode can smoothly and slowly evolve into the local mode of the next location, thereby suppressing crosstalk and energy loss between modes. In numerical design, a common method (hereinafter referred to as the "traditional method") is to discretize the continuously changing structure into a series of short "segments" with approximately constant cross-sections. Then, by calculating the minimum length required for each segment, the mode conversion loss within that segment is guaranteed to be below a preset threshold. Finally, all segments are connected to form the entire device.

[0004] However, such methods in the prior art have significant drawbacks:

[0005] (1) Parameter preset constraints: Before optimization begins, the variation function of the structural cross-sectional parameters (such as waveguide width and gap) along the length of the device must be preset. For example, when designing an adiabatic coupler, the designer must first assume that the gap between the two waveguides changes linearly, exponentially, or in some other functional form.

[0006] (2) Non-cooperative optimization: The target variable of the algorithm optimization is usually only one - the "length" of each segment. It cannot optimize the segment length and other key geometric parameters (such as gap size, waveguide width, etc.) at the same time during the optimization process. This leads to the optimization process searching for local optimal solutions on a limited, pre-set "design path" instead of exploring the entire multi-dimensional design space, and often fails to obtain the global optimal design.

[0007] (3) Low design freedom: Due to the above restrictions, the performance of the designed device (such as total length and loss) depends heavily on the designer's initial experience and guesswork, making it impossible to achieve true automation and optimal design, thus limiting the performance potential of the device.

[0008] Therefore, there is an urgent need for a new method that can overcome the limitations of preset parameters and achieve collaborative optimization of multiple geometric parameters in order to design thermally insulated waveguide structures with better performance and more compact size. Summary of the Invention

[0009] This invention aims to address the problems in existing thermally adiabatic waveguide structure design methods, such as the need to pre-set geometric parameters and the inability to perform multi-parameter collaborative optimization. Specifically, this invention provides a new design paradigm that allows multiple variables, such as the device's segment length and cross-sectional geometric parameters (e.g., gaps and width), to be optimized simultaneously. This enables the finding of a globally superior device structure based on specified secondary optimization objectives (e.g., minimum total length) while meeting thermal insulation performance requirements.

[0010] The technical solution adopted in this invention is as follows: A parameter co-optimization design method for optical waveguide insulation structures, specifically including the following steps: Step S1: Structure discretization and parameter definition; Step S2: Constructing a performance model under multi-dimensional parameters; Step S3: Determining the effective design solution space; Step S4: Co-optimization of parameters based on secondary optimization objectives; Step S5: Iteration and structure reconstruction.

[0011] Further, as a preferred embodiment of the present invention, step S1 specifically involves: discretizing the continuous optical waveguide structure to be designed, evolving from the initial state to the target state, into N interconnected micro-segments, i = 1, 2, ..., N. For each micro-segment i, at least two variables to be optimized are defined:

[0012] (1) Variable 1 (length): The length L of this segment i .

[0013] (2) Variable 2 (geometric parameter): The change in one or more cross-sectional geometric parameters of this segment, denoted as vector P. i For example, P i It can be the gap change ΔG between the output and input terminals of this segment. i Width change ΔW i or a combination of both (ΔG) i , ΔW i )wait.

[0014] Furthermore, as a preferred embodiment of the present invention, step S2 specifically involves: for each micro-segment i, establishing its thermal insulation performance index η. i With the variable to be optimized L i and P i The functional relationship between them. This performance index η i This is used to quantify the non-adiabatic degree of this segment, such as mode conversion loss, mode crosstalk coefficient, or adiabatic factor based on coupled-mode theory. The functional relationship is expressed as:

[0015] η i = f(L i , P i )

[0016] This function can be obtained through numerical calculation methods (such as the finite difference method, the finite element method, the beam propagation method, etc.).

[0017] Furthermore, as a preferred embodiment of the present invention, step S3 specifically involves: setting a globally unified and acceptable thermal insulation performance threshold η. target For each tiny segment i, solve the following equation:

[0018] f(L i , P i ) = η target

[0019] The solution to this equation is no longer a single point, but rather lies within (L). i , P i A contour or isoline is a surface or contour line in a multidimensional parameter space. All points on this contour / line represent an effective combination of design schemes that meet the insulation performance requirements.

[0020] Furthermore, as a preferred embodiment of the present invention, step S4 specifically involves defining a secondary optimization objective function O(L). i , P i This function is used to evaluate the "goodness" of an effective design. The objective function can be set according to actual engineering needs, for example:

[0021] (1) Objective 1 (Minimum length): O = L i .

[0022] (2) Objective 2 (Minimum manufacturing complexity): O = w1L i + w2∣P i |, where w1 and w2 are weighting factors, representing a balance between drastic changes in length and geometric parameters.

[0023] (3) Objective 3 (Other performance indicators): such as minimum bending loss, etc.

[0024] On the contour surface / line obtained in step S3, find the secondary optimization objective function O(L) i , P i The point that reaches the optimal value (e.g., minimum or maximum). The coordinates of this point (L... i * , P i * The optimal collaborative design parameters are those for the small segment i.

[0025] Further, as a preferred embodiment of the present invention, step S5 specifically involves: repeating steps S2 to S4 for all small segments i = 1, 2, ..., N to obtain the optimal collaborative design parameter set {(L1)} for each segment. * , P 1 * ), (L 2 * , P 2 * ), ..., (L N * , P N * )}.

[0026] All the optimized micro-segments are connected sequentially to form the final, optimally performing thermally adiabatic waveguide structure. Each segment of this structure has an optimal length and optimal geometric parameter variations.

[0027] The parameter collaborative optimization design method for optical waveguide thermal insulation structures described in this invention has the following technical advantages compared with existing technologies:

[0028] (1) Achieving multi-parameter collaborative optimization: This invention breaks through the limitation of traditional methods that can only optimize a single length parameter, and can simultaneously optimize the segment length and multiple key geometric parameters, thus achieving true collaborative design.

[0029] (2) Improve device performance and compactness: By searching in a wider design space, better design solutions that cannot be found by traditional methods can be found, thereby significantly shortening the total length of the device while ensuring the same low loss; or achieving lower loss with the same length.

[0030] (3) Improve design freedom and automation: This invention provides a systematic and automated optimization framework. Designers only need to define the parameters to be optimized, performance targets and secondary optimization criteria, without having to guess the parameter change path based on experience, which greatly reduces the design difficulty and improves the design efficiency. Attached Figure Description

[0031] Figure 1 This is a flowchart of the parameter collaborative optimization design method according to an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram illustrating collaborative optimization for a single micro-segment according to an embodiment of the present invention. Detailed Implementation

[0033] The present invention will be further explained in detail below with reference to the accompanying drawings, so that those skilled in the art can better understand and implement the present invention. However, the following examples are only used to explain the present invention and are not intended to limit the present invention.

[0034] like Figure 1 As shown, a parameter co-optimization design method for optical waveguide insulation structures includes the following steps: Step S1: Structure discretization and parameter definition; Step S2: Constructing a performance model under multi-dimensional parameters; Step S3: Determining the effective design solution space; Step S4: Co-optimizing parameters based on secondary optimization objectives; Step S5: Iteration and structure reconstruction.

[0035] Step S1 specifically involves discretizing the continuous optical waveguide structure to be designed, evolving from the initial state to the target state, into N interconnected segments, i = 1, 2, ..., N. For each segment i, at least two variables to be optimized are defined:

[0036] (1) Variable 1 (length): The length L of this segment i .

[0037] (2) Variable 2 (geometric parameter): The change in one or more cross-sectional geometric parameters of this segment, denoted as vector P. i For example, P i It can be the gap change ΔG between the output and input terminals of this segment. i Width change ΔW i or a combination of both (ΔG) i , ΔW i )wait.

[0038] Step S2 specifically involves: for each small segment i, establishing its thermal insulation performance index η. i With the variable to be optimized L i and P i The functional relationship between them. This performance index η i This is used to quantify the non-adiabatic degree of this segment, such as mode conversion loss, mode crosstalk coefficient, or adiabatic factor based on coupled-mode theory. The functional relationship is expressed as:

[0039] η i = f(L i , P i )

[0040] This function can be obtained through numerical calculation methods (such as the finite difference method, the finite element method, the beam propagation method, etc.).

[0041] Step S3 specifically involves setting a globally uniform and acceptable thermal performance threshold η.target For each tiny segment i, solve the following equation:

[0042] f(L i , P i ) = η target

[0043] The solution to this equation is no longer a single point, but rather lies within (L). i , P i A contour or isoline is a surface or contour line in a multidimensional parameter space. All points on this contour / line represent an effective combination of design schemes that meet the insulation performance requirements.

[0044] Step S4 specifically involves defining a secondary optimization objective function O(L). i , P i This function is used to evaluate the "goodness" of an effective design. The objective function can be set according to actual engineering needs, for example:

[0045] (1) Objective 1 (Minimum length): O = L i .

[0046] (2) Objective 2 (Minimum manufacturing complexity): O = w1L i + w2∣P i |, where w1 and w2 are weighting factors, representing a balance between drastic changes in length and geometric parameters.

[0047] (3) Objective 3 (Other performance indicators): such as minimum bending loss, etc.

[0048] On the contour surface / line obtained in step S3, find the secondary optimization objective function O(L) i , P i The point that reaches the optimal value (e.g., minimum or maximum). The coordinates of this point (L... i * , P i * The optimal collaborative design parameters are those for the small segment i.

[0049] Step S5 specifically involves repeating steps S2 to S4 for all small segments i = 1, 2, ..., N to obtain the optimal collaborative design parameter set {(L1)} for each segment. * P1 * ), (L2 * P2 * ), ..., (L N * , P N * )}.

[0050] Figure 2 This diagram illustrates the collaborative optimization of individual micro-segments in the method of this invention, showing contour lines in a two-dimensional parameter space (segment length, gap variation) and the optimal solution found based on the secondary optimization objective. All optimized micro-segments are connected sequentially to form the final, overall optimal thermally adiabatic waveguide structure. Each segment of this structure has optimal length and optimal geometric parameter variation.

[0051] Specific implementation: Collaborative optimization design of thermally adiabatic directional couplers:

[0052] This embodiment aims to design an asymmetric thermally adiabatic directional coupler with the shortest total length, which functions to efficiently couple input light from a wider waveguide 1 to a narrower waveguide 2.

[0053] Step 1: Structure Discretization and Parameter Definition (corresponding to Step S1)

[0054] The coupler structure is discretized into N = 50 small segments along the propagation direction. For the i-th segment, its input width is (W 1, i-1 W 2, i-1 ), gap is G i-1 Its output width is (W) 1, i W 2, i ), gap is G i .

[0055] In this embodiment, the width change ΔW 1, i = W 1, i - W 1,i-1 and ΔW 2,i = W 2,i - W 2,i-1 It is evenly distributed to each segment according to the overall width variation requirement, and it is a known quantity.

[0056] The variable to be co-optimized is the length L of each segment. i and gap change ΔG i = G i- G i-1 .

[0057] Step 2: Construct a performance model (corresponding to step S2)

[0058] For the i-th segment, its thermal insulation performance index η i Defined as mode conversion loss, it is the power retention rate of light from the input fundamental mode (or supermode) to the output fundamental mode (or supermode). The higher the loss, the lower the efficiency of η. iThe larger the value, the greater the loss. The mode characteristics of this segment are calculated using a finite difference eigenmode solver (FDE Solver), and the loss is calculated using coupled-mode theory or the beam propagation method (BPM). This establishes the functional relationship:

[0059] η i = Loss(L i , ΔG i )

[0060] Step 3: Determine the effective design solution space (corresponding to step S3)

[0061] Set the maximum acceptable loss threshold for each segment to η. target = 0.001 dB. Then, for each segment i, solve the equation:

[0062] Loss(L i , ΔG i ) = 0.001 dB

[0063] Through two-dimensional parameter space (L i , ΔG i By performing a scanning calculation, a series of points satisfying the equation can be obtained, and these points form a contour line. Any point (L, ΔG) on this line is a valid alternative.

[0064] Step 4: Cooperative parameter optimization (corresponding to step S4)

[0065] The secondary optimization objective of this embodiment is to minimize the total length of the device. Therefore, for each segment, our objective is to minimize its segment length L. i Shortest. The secondary optimization objective function is O = L i .

[0066] Find L on the contour line i The point with the smallest coordinate value. The coordinates of this point (L... i * ,ΔG i * This represents the optimal collaborative design parameter for the i-th segment. Compared to choosing a larger ΔG... a This results in a longer L a Or a smaller ΔG b This results in a longer L b There exists an optimal ΔG. i * Make the segment length L i * It reaches the minimum value.

[0067] Step 5: Iteration and Structure Reconstruction (corresponding to step S5)

[0068] Repeat the above steps for all segments i = 1, ..., 50 to obtain the optimal length L for each segment. i * and the optimal gap change ΔG i * .

[0069] Finally, these 50 optimized segments are connected together. The geometric parameters of the i-th segment are: length L. i * Input gap G i-1 = G0 +∑ j=1 i-1 ΔG j * Output gap G i = G i-1 +ΔG i * .

[0070] The resulting coupler structure not only has the shortest total length, but its gap G(z) variation curve along the length z is automatically optimized rather than preset by humans, thus achieving a truly optimal design.

[0071] The specific implementation schemes described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific implementation schemes of the present invention and are not intended to limit the scope of the present invention. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A parameter co-optimization design method for optical waveguide insulation structures, characterized in that, Specifically, the following steps are included: Step S1: Structure discretization and parameter definition; Step S2: Construct a performance model under multi-dimensional parameters; Step S3: Determine the effective design solution space; Step S4: Optimize collaborative parameters based on secondary optimization objectives; Step S5: Iteration and structural reconstruction.

2. The parameter collaborative optimization design method for optical waveguide thermal insulation structures according to claim 1, characterized in that, The specific steps of step S1 are as follows: Discretize the continuous optical waveguide structure to be designed, which evolves from the initial state to the target state, into N interconnected micro-segments, i = 1, 2, ..., N; For each micro-segment i, define at least two variables to be optimized: Variable 1 is the length: the length L of the segment. i Variable 2 is a geometric parameter: the change in the geometric parameter of one or more cross-sections of this segment, denoted as vector P. i P i It is the change in gap ΔG between the output terminal and the input terminal of this segment. i Width change ΔW i or a combination of both (ΔG) i , ΔW i ).

3. The parameter collaborative optimization design method for optical waveguide thermal insulation structures according to claim 2, characterized in that, The specific steps of step S2 are as follows: For each small segment i, establish its thermal insulation performance index η. i With the variable to be optimized L i and P i The functional relationship between them; the performance index η i The functional relationship used to quantify the non-adiabatic degree of this segment is expressed as follows: or i = f(L i , P i ) This function is obtained through numerical calculation methods.

4. The parameter collaborative optimization design method for optical waveguide thermal insulation structures according to claim 3, characterized in that, The specific steps of step S3 are as follows: setting a globally unified and acceptable thermal performance threshold η. target For each tiny segment i, solve the following equation: f(L i , P i ) = the target The solution to this equation is no longer a single point, but rather lies within (L). i , P i A contour surface or contour line in a multidimensional parameter space, where all points on the contour surface / line represent an effective combination of design schemes that meet the thermal insulation performance requirements.

5. The parameter collaborative optimization design method for optical waveguide thermal insulation structures according to claim 4, characterized in that, The specific steps of step S4 are as follows: Define a secondary optimization objective function O(L i , P i This function is used to evaluate the quality of effective design schemes; The objective function is set according to the actual engineering requirements. Objective 1 is to minimize the length: O = L i Objective 2 is to minimize manufacturing complexity: O = w1L i + w2∣P i |, where w1 and w2 are weighting factors, representing a balance between drastic changes in length and geometric parameters; objective three is other performance indicators: minimum bending loss; On the contour surface / line obtained in step S3, find the secondary optimization objective function O(L) i , P i The point that reaches the optimal value; the coordinates of that point (L) i * , P i * The optimal collaborative design parameters are those for the small segment i.

6. The parameter collaborative optimization design method for optical waveguide thermal insulation structures according to claim 5, characterized in that, The specific steps of step S5 are as follows: Repeat steps S2 to S4 for all small segments i = 1, 2, ..., N to obtain the optimal collaborative design parameter set {(L1)} for each segment. * P1 * ), (L2 * P2 * ), ..., (L N * , P N * )}; All the optimized micro segments are connected sequentially to form the final, overall optimal thermally insulated optical waveguide structure; each segment of this structure has the optimal length and the optimal geometric parameter variation.

Images