Uniform light distribution lens system based on free-form surface and design method thereof

Abstract

The invention provides a uniform light distribution lens system based on a free-form surface and a design method thereof, and relates to the technical field of lens design, and the method comprises the steps: obtaining a target surface light intensity distribution diagram under a lens sample, and recognizing a to-be-improved region in the light intensity distribution diagram based on a uniform light distribution standard; dividing the control points on the free-form surface into optimization control points and reference control points according to types based on the to-be-improved area; and after random disturbance is carried out on the thickness value of the control point based on the type to which the control point belongs, the optimal three-dimensional mathematical model is determined by adopting a simulated annealing algorithm by taking the maximization of the light intensity distribution uniformity degree of the target surface and the smoothness of the free-form surface as an optimization target. The optimization control point and the reference control point are pertinently perturbed and updated to improve the quality of new solution generation, so that the efficiency of optimizing the free-form surface lens based on the simulated annealing algorithm is improved.

Description

Technical Field

[0001] This invention relates to the field of lens design technology, specifically to a uniform light distribution lens system based on a freeform surface and its design method. Background Technology

[0002] Beam shaping is the manipulation of the spatial intensity distribution of a light beam, and it has wide applications in uniform illumination and laser processing. The design methods for optical shaping elements are mainly divided into two categories: one is to use Fourier optics algorithms to design diffraction elements for optical shaping; the other is to calculate the surface profile of the transmission or reflection optical element required for beam shaping based on the energy distribution of the incident beam and the energy distribution of the target beam, according to Snell's law and the law of conservation of energy.

[0003] Freeform optical surfaces are optical surfaces that do not possess rotational or translational symmetry. Compared to traditional optical surfaces (spherical, aspherical, parabolic, etc.), they offer greater freedom in optical design and more flexible spatial layout. When applied to illumination optical systems, they can simplify the structure of the optical system while providing more precise beam control capabilities. Therefore, the design methods of freeform lens lenses are receiving increasing attention from relevant technical personnel.

[0004] In the prior art, a design method and system for a freeform surface light distribution lens is disclosed in "CN116381930A". The method first sets a virtual plane at a position perpendicular to the optical axis, and transforms the required illuminance distribution on the three-dimensional target surface to the virtual plane through a self-built mathematical model to obtain a virtual illuminance distribution. Then, a beam control mathematical model is constructed through the law of conservation of energy, boundary constraints and integrability conditions, and a freeform surface light distribution lens is designed to realize the illuminance distribution on the virtual plane.

[0005] However, existing technologies still have significant problems. For example, when optimizing the design of freeform surface lenses, existing technologies uniformly and blindly update and change the positions of each point on the freeform surface. This leads to problems such as updating reasonable point positions to unreasonable point positions and blindly updating unreasonable point positions during actual iterative updates. This seriously affects the stability of the new solution quality and results in low optimization efficiency.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a uniform light distribution lens system based on a freeform surface and its design method, so as to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A design method for a freeform surface lens includes the following steps: S1. Model the lens sample containing the freeform surface to generate its three-dimensional mathematical model, and input the three-dimensional mathematical model into the optical simulation software to obtain the light intensity distribution map of the target surface, and identify the areas to be improved in the light intensity distribution map that are uneven in light intensity distribution based on the uniform light distribution standard. S2, based on the ray tracing algorithm, analyzes the area to be improved, classifies the control points set on the freeform surface into optimized control points and reference control points according to their type, and performs reduction processing on the reference control points; S3, after randomly perturbing the thickness value based on the type of the control point, the three-dimensional mathematical model is reconstructed by combining the double interpolation algorithm and the second simulation is performed based on optical simulation software. The optimization objectives are to maximize the uniformity of light intensity distribution on the target surface and maximize the smoothness of the freeform surface. The optimal three-dimensional mathematical model is determined by the simulated annealing algorithm.

[0009] Furthermore, the logic for identifying the area to be improved is as follows: 1) Determine the size of the evaluation grid based on the height and width of the light intensity distribution map, and then generate an evaluation network covering the light intensity distribution map, which consists of several evaluation grids. The area of ​​the light intensity distribution map that falls within each evaluation grid is treated as a separate evaluation sub-region. 2) A uniform light distribution standard consisting of minimum light intensity constraint, light intensity extreme ratio constraint, and light intensity variation constraint is preset. Each evaluation sub-region is judged to determine whether it meets the above three constraints at the same time. If it does, it is judged as a qualified region; otherwise, it is judged as a region to be improved.

[0010] Furthermore, the logic for dividing the light intensity distribution map into several evaluation sub-regions is as follows: 1.1) Preset a scaling factor that is a positive decimal, calculate the product of the light intensity distribution map height value and the scaling factor as the evaluation grid height value, and calculate the product of the light intensity distribution map width value and the scaling factor as the evaluation grid width value, thereby determining the size of the evaluation grid; 1.2) Lay the first evaluation grid on the light intensity distribution map, and make sure that the geometric center of the evaluation grid coincides with the geometric center of the light intensity distribution map, so as to determine the initial laying position of the evaluation grid; 1.3) Using the newly laid evaluation grid as the center grid, the evaluation grid laying is performed based on the following rules: Rule 1: Determine whether there is a light intensity distribution area directly above the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly above the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid upward by one evaluation grid height value. Rule 2: Determine whether there is a light intensity distribution area directly below the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly below the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid down by one evaluation grid height value. Rule 3: Determine whether there is a light intensity distribution area to the left of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the left of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the left by one evaluation grid width value. Rule 4: Determine whether there is a light intensity distribution area to the right of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the right of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the right by one evaluation grid width value. 1.4) Repeat step 1.3) until no new evaluation grid can be laid, thereby generating an evaluation network that covers the light intensity distribution map. Then, the area of ​​the light intensity distribution map that falls within each evaluation grid is treated as a separate evaluation sub-region.

[0011] Furthermore, the minimum light intensity constraint condition is: the light intensity value at each location in the evaluation sub-region is not less than the minimum light intensity threshold. The constraint condition for the extreme value ratio of light intensity is: the ratio of the maximum light intensity to the minimum light intensity in the evaluation sub-region is not higher than the extreme value ratio threshold, and the extreme value ratio threshold is greater than one; The constraint condition for the degree of light intensity variation is: the ratio of the standard deviation of light intensity to the average value of light intensity within the evaluation sub-region is not higher than the light intensity variation threshold, and the light intensity variation threshold is a positive decimal less than one.

[0012] Furthermore, the logic for setting control points on the freeform surface is as follows: uniform sampling is performed on the plane of the lens sample at a fixed sampling interval to generate several sampling points. For each sampling point, its projection position on the freeform surface along the direction perpendicular to the plane is used as a control point set on the freeform surface.

[0013] Furthermore, the logic for classifying control points is as follows: for any control point, the ray tracing function in the optical simulation software is used to simulate the position of the light rays after passing through the control point and incident on the light intensity distribution map. If this position is located in the area to be improved, then the control point is used as the optimization control point; otherwise, the control point is used as the reference control point. The logic for reducing reference control points is as follows: Define the sampling point corresponding to the reference control point as the reference sampling point. Randomly select a reference sampling point from all reference sampling points on the plane, and remove all reference sampling points whose distance from the reference sampling point is less than the reduction radius. Iterate in this way until the distance between all remaining reference sampling points on the plane is not less than the reduction radius. Retain the reference control points that correspond one-to-one with the remaining reference sampling points on the plane, and the reduction radius is greater than the sampling interval.

[0014] Furthermore, the logic for determining the optimal three-dimensional mathematical model is as follows: 1) Set the initial temperature, final temperature, and cooling rate; 2) Using the thickness value of each control point in the lens sample as the initial solution, and combining it with the light intensity distribution map of the target surface under the initial solution, the energy value of the initial solution is determined. It is generated by weighted fusion of the uniform light distribution index, which characterizes the degree of adaptation of the light intensity distribution map of the target surface to the uniform light distribution standard, and the smoothness index, which characterizes the smoothness of the freeform surface. 3) The initial solution is used as the current solution for the first round of annealing optimization. The thickness values ​​of each control point in the current solution are randomly perturbed based on the control point type to generate a new solution. The energy values ​​of the current solution and the new solution are compared to determine the current solution for the next round of annealing optimization. The iteration is repeated until the annealing temperature is no greater than the termination temperature. The current solution under the last round of annealing optimization is taken as the optimal solution, and the optimal three-dimensional mathematical model is determined based on the optimal solution.

[0015] Furthermore, for the initial solution, the calculation logic for its corresponding uniform light distribution index is as follows: In the light intensity distribution map of the target surface corresponding to the initial solution, each evaluation sub-region is subjected to a fit analysis with the three constraints in the uniform light distribution standard to obtain the fit value of each evaluation sub-region relative to each constraint. The fit values ​​of the same evaluation sub-region relative to each constraint are weighted and fused to obtain the fit index of the evaluation sub-region relative to the uniform light distribution standard. The average value of the fit index of all evaluation sub-regions is taken as the uniform light distribution index corresponding to the initial solution. For the initial solution, the corresponding smoothness index is calculated as follows: calculate the standard deviation and average value of the thickness values ​​of each control point in the initial solution, and use the ratio of the standard deviation and average value of the thickness values ​​as the smoothness index.

[0016] Furthermore, for any given round of annealing optimization, the logic for generating a new solution based on the current solution is as follows: 1) Calculate the ratio of the annealing temperature to the initial temperature under this round of annealing optimization, and use the product of this ratio and the initial value of the perturbation amplitude as the perturbation amplitude under this round of annealing optimization, and the initial value of the perturbation amplitude is a positive decimal less than one. 2) Multiply the perturbation amplitude under this round of annealing optimization by a scaling factor with a positive decimal value to generate the reference perturbation amplitude under this round of annealing optimization. Use the reference perturbation amplitude as the upper limit value and the product of the reference perturbation amplitude and -1 as the lower limit value to construct the reference perturbation interval. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, multiply its thickness value by a random value in the reference perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, sum its thickness value and thickness perturbation value to obtain the thickness value of the control point in the new solution under this round of annealing optimization. 3) The perturbation amplitude under this round of annealing optimization is used as the upper limit, and the product of the perturbation amplitude under this round of annealing optimization and -1 is used as the lower limit to construct the optimization perturbation interval. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, if there exists a reference control point whose distance between its corresponding sampling point and its corresponding sampling point is less than the search radius, then this reference control point is used as its baseline control point, and the search radius is greater than the reduction radius. Based on the distance between the corresponding sampling points of each baseline control point and its corresponding sampling point, the attenuation weight of each baseline control point relative to it is determined based on the Gaussian function. The thickness values ​​of each baseline control point are weighted and fused based on the attenuation weight to obtain the baseline thickness value of the control point. The baseline thickness value is multiplied by a random value in the optimization perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, its thickness value and thickness perturbation value are summed to obtain the thickness value of this control point in the new solution under this round of annealing optimization.

[0017] A uniform light distribution lens system based on a freeform surface includes a freeform surface lens manufactured using the above-described design method, a light source for providing illumination, a bracket for fixing the light source, and a light intensity adjustment device for adjusting the brightness of the light source.

[0018] Compared with the prior art, the beneficial effects of the present invention are: The present invention relates to a freeform surface-based uniform light distribution lens system and its design method. When optimizing the design of a freeform surface lens, based on the region to be improved in the light intensity distribution map, the control points on the freeform surface are divided into reasonable reference control points and unreasonable optimization control points. Thus, when subsequently updating and optimizing the freeform surface lens, targeted perturbation updates are performed on the optimization control points and reference control points based on the type of control points, thereby improving the quality of the generated new solution and improving the efficiency of optimizing the freeform surface lens based on the simulated annealing algorithm. Attached Figure Description

[0019] Figure 1This is a flowchart illustrating the design method of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0021] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly. Example 1

[0022] Please see Figure 1 This invention provides a design method for a freeform surface lens, comprising the following steps: S1. Model the lens sample containing the freeform surface to generate its three-dimensional mathematical model, and input the three-dimensional mathematical model into the optical simulation software to obtain the light intensity distribution map of the target surface, and identify the areas to be improved in the light intensity distribution map that are uneven in light intensity distribution based on the uniform light distribution standard. The lens sample is an existing freeform lens, which includes a freeform surface facing the light source and a plane facing away from the light source. Specifically, it can be a freeform lens from an existing uniform light distribution lens system. The design method of this technical solution is to further design a freeform lens with better uniform light distribution performance than this sample based on the existing lens sample. The specific scheme for modeling the lens sample is as follows: 1) Use a 3D scanner to scan the lens sample to obtain its point cloud data. Specifically, you can use existing laser 3D scanners such as FARO and Leica Geosystems or structured light 3D scanners such as 3D Systems Sense and David 3DScanner to achieve high-precision scanning of the lens sample and obtain high-resolution point cloud data of the lens sample, which is convenient for subsequent modeling to generate a high-precision 3D mathematical model. 2) Preprocess the point cloud data, including noise reduction, downsampling and alignment, to improve the accuracy of subsequent modeling. Preprocessing is a conventional technique for those skilled in the art and can be implemented using existing point cloud processing software such as MeshLab or CloudCompare. The specific implementation algorithms and processes will not be elaborated here. 3) Uniform sampling is performed on the preprocessed point cloud data to obtain several feature points. Then, the feature points are analyzed based on the surface fitting algorithm to generate the mathematical surface of the lens sample. This transforms the discrete control points into a continuous mathematical surface expression. Specifically, non-uniform rational B-splines or Bezier surfaces can be used for surface fitting algorithms, which will not be elaborated here. In addition, when performing uniform sampling, the sampling interval between adjacent control points can be set between 1mm and 5mm to reduce the computational complexity while ensuring the accuracy of subsequent modeling. 4) Analyze the mathematical surfaces of the lens sample based on mathematical modeling software to construct a three-dimensional mathematical model of the lens sample. Specifically, the three-dimensional mathematical model of the lens sample can be constructed using conventional three-dimensional modeling software such as MATLAB and Blender. This is existing technology and will not be elaborated here. The logic for obtaining the light intensity distribution map of the target surface is as follows: 1) Import the three-dimensional mathematical model of the lens sample into the optical simulation software and configure the actual application scenario of the uniform light distribution lens system. This includes the spatial arrangement of the point light source, the freeform lens, and the target surface, as well as the parameters of the point light source and the freeform lens. The target surface is the display surface in the uniform light distribution lens system. The point light source parameters include the light power and wavelength of the point light source. The freeform lens parameters include the material properties and refractive index of the freeform lens. Generally, the material properties can be set to optical glass and the refractive index can be set to 1.5. Specifically, the Zemax optical software can be used to configure the actual application scenario of the light distribution lens system. Import the three-dimensional mathematical model in STL format into the optical simulation software. The specific configuration process is a conventional technique for those skilled in the art and will not be elaborated here. 2) Set the number of rays, which can generally be set to 10,000 rays, and execute the ray tracing function in the optical simulation software to simulate the ray propagation process of light emitted from a point light source after passing through the lens sample and incident on the target surface. Finally, output the light intensity distribution map of the target surface through the visualization tool of the optical simulation software. Specifically, output the light intensity distribution map of the target surface in CSV format. Executing ray tracing and outputting the light intensity distribution map based on the optical simulation software are conventional techniques for those skilled in the art and will not be elaborated here. The logic for identifying the area to be improved is as follows: 1) The size of the evaluation grid is determined based on the height and width of the light intensity distribution map, thereby generating an evaluation network covering the light intensity distribution map. This network consists of several evaluation grids, and the region of the light intensity distribution map that falls within each evaluation grid is treated as a separate evaluation sub-region. The specific logic is as follows: 1.1) Preset a scaling factor that is a positive decimal, calculate the product of the light intensity distribution map height value and the scaling factor as the evaluation grid height value, and calculate the product of the light intensity distribution map width value and the scaling factor as the evaluation grid width value, thereby determining the size of the evaluation grid; It should be noted that the scaling factor is used to find the appropriate size of the evaluation grid, which facilitates the subsequent division of the light intensity distribution map into an appropriate number of evaluation sub-regions and the analysis of light intensity distribution uniformity in each sub-region. The scaling factor can generally be set between 0.1 and 0.15, so that the light intensity distribution map can be divided into about 49-100 evaluation sub-regions, so as to achieve fine partitioning and evaluation of the light intensity distribution map. The specific value is set by the staff according to the actual situation, but it should not be lower than 0.05 and not higher than 0.3, to avoid setting the evaluation grid size too large, which is not conducive to subsequent accurate optimization, or setting it too small, which makes it difficult to filter out small areas with poor light intensity distribution uniformity. 1.2) Lay the first evaluation grid on the light intensity distribution map, and make sure that the geometric center of the evaluation grid coincides with the geometric center of the light intensity distribution map. This will determine the initial placement of the evaluation grid and lay the foundation for laying the evaluation grid in the future. 1.3) Using the newly laid evaluation grid as the center grid, the evaluation grid laying is performed based on the following rules: Rule 1: Determine whether there is a light intensity distribution area directly above the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly above the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid upward by one evaluation grid height value. Rule 2: Determine whether there is a light intensity distribution area directly below the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly below the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid down by one evaluation grid height value. Rule 3: Determine whether there is a light intensity distribution area to the left of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the left of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the left by one evaluation grid width value. Rule 4: Determine whether there is a light intensity distribution area to the right of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the right of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the right by one evaluation grid width value. 1.4) Repeat step 1.3) until no new evaluation grid can be laid out, thereby generating an evaluation network covering the light intensity distribution map. Then, the area of ​​the light intensity distribution map that falls within each evaluation grid is set up as a separate evaluation sub-region. This is done to divide the light intensity distribution map into several evaluation sub-regions. 2) A uniform light distribution standard is preset, consisting of a minimum light intensity constraint, a light intensity extreme ratio constraint, and a light intensity variation degree constraint. Each evaluation sub-region is judged to see if it meets the above three constraints at the same time. If it does, it is considered to meet the uniform light distribution standard and is judged as a qualified region. Otherwise, it is considered to not meet the uniform light distribution standard and is judged as a region to be improved. The minimum light intensity constraint is that the light intensity value at each location in the evaluation sub-region is not less than the minimum light intensity threshold. The minimum light intensity threshold is set according to the application scenario of the uniform light distribution lens system and is not restricted here. The constraint condition for the extreme value ratio of light intensity is: the ratio of the maximum light intensity to the minimum light intensity in the evaluation sub-region is not higher than the extreme value ratio threshold, and the extreme value ratio threshold is greater than one. For example, it can be set between 1.2 and 1.7 to evaluate the degree of extreme non-uniformity of light intensity in the evaluation sub-region. The specific value is set by the staff according to the actual situation, such as 1.5. The constraint on the degree of light intensity variation is as follows: the ratio of the standard deviation of light intensity to the average light intensity in the evaluation sub-region is not higher than the light intensity variation threshold, and the light intensity variation threshold is a positive decimal less than one, such as between 0.1 and 0.3, to evaluate the overall unevenness of light intensity in the evaluation sub-region. The specific value is set by the staff according to the actual situation, such as 0.2. It should be noted that by setting the above three constraints, the light intensity distribution in each evaluation sub-region is evaluated from three aspects: the magnitude of light intensity value, the degree of extreme difference in light intensity, and the degree of overall difference in light intensity. In this way, evaluation sub-regions that do not meet the requirements of uniform light distribution are selected as areas to be improved, providing a benchmark for subsequent control point classification.

[0023] S2, based on the ray tracing algorithm, analyzes the area to be improved, classifies the control points set on the freeform surface into optimized control points and reference control points according to their type, and performs reduction processing on the reference control points; The logic for setting control points on the freeform surface is as follows: uniform sampling is performed on the plane of the lens sample at a fixed sampling interval to generate several sampling points. For each sampling point, its projection position on the freeform surface along the direction perpendicular to the plane is used as a control point set on the freeform surface. In this way, the sampling points and control points are matched one by one to achieve the purpose of setting control points on the freeform surface. It should be noted that the sampling interval can generally be set between 3mm and 7mm to avoid excessive sampling and significantly increase the workload while taking into account the number of samples. By sampling on the plane with a fixed sampling interval, the interval between adjacent sampling points is consistent, thereby ensuring uniform sampling on the plane. The logic for classifying control points is as follows: For any control point, the ray tracing function in the optical simulation software is used to simulate the position of the light rays incident on the light intensity distribution map after passing through the control point. If this position is located in the area to be improved, it means that the freeform surface design at the control point does not meet the requirements, and the control point is used as an optimization control point. Otherwise, it means that the freeform surface design at the control point meets the requirements, and the control point is used as a reference control point. The logic for reducing reference control points is as follows: define the sampling point corresponding to the reference control point as the reference sampling point, randomly select a reference sampling point from all reference sampling points on the plane, remove all reference sampling points whose distance from the reference sampling point is less than the reduction radius, and iterate in this way until the distance between all remaining reference sampling points on the plane is not less than the reduction radius, and retain the reference control points that correspond one-to-one with the remaining reference sampling points on the plane, thus completing the reduction of reference control points, and the reduction radius is greater than the sampling interval; It should be noted that there are no issues with the freeform surface design at the reference control points, so there is no need to set a high density of reference control points for subsequent optimization. This step greatly reduces the computational load of subsequent optimization by reducing the reference control points, and significantly improves the optimization efficiency of the freeform surface lens. The value of the reduction radius can generally be set between 1.5 times the sampling interval and 3 times the sampling interval. This ensures the reduction of reference control points while avoiding the problem of insufficient reference control points causing adverse effects on the subsequent reconstruction of the 3D mathematical model.

[0024] S3. After randomly perturbing the thickness value based on the type of the control point, the three-dimensional mathematical model is reconstructed by combining the double interpolation algorithm and a second simulation is performed based on optical simulation software. The optimization objectives are to maximize the uniformity of light intensity distribution on the target surface and maximize the smoothness of the freeform surface. The optimal three-dimensional mathematical model is determined by the simulated annealing algorithm. For any given control point, the thickness value is the shortest distance from that control point to the plane. The logic for determining the optimal three-dimensional mathematical model is as follows: 1) Set the initial temperature, the final temperature, and the cooling rate. The specific values ​​of these three values ​​shall be set by the staff according to the actual situation. For example, the initial temperature is set between 100-1000, the final temperature is set between 0.1-1, and the cooling rate is set between 0.8-0.99. This is the conventional choice for those skilled in the art and will not be elaborated here. 2) Using the thickness values ​​of each control point in the lens sample as the initial solution, and combining them with the light intensity distribution map of the target surface under the initial solution, the energy value of the initial solution is determined. It is generated by weighted fusion of the uniform light distribution index, which characterizes the degree of fit between the light intensity distribution map of the target surface and the smoothness index, which characterizes the smoothness of the freeform surface. The specific mathematical expression is as follows: ; In the formula, This represents the initial solution, which is an eigenvector composed of the thickness values ​​of each control point in the lens sample. The uniform light distribution index corresponding to the initial solution is used to evaluate the degree of fit of the target surface light intensity distribution map to the uniform light distribution standard under the lens sample corresponding to the initial solution. The larger the value, the worse the fit of the target surface light intensity distribution map to the uniform light distribution standard under the lens sample corresponding to the initial solution, which means the quality of the initial solution is worse. In the formula, This is a smoothness index corresponding to the initial solution, used to evaluate the smoothness of the freeform surface under the lens sample corresponding to the initial solution. The larger the value, the worse the smoothness of the freeform surface, which means that it is more difficult to manufacture the lens sample based on the initial solution, and the worse the quality of the initial solution. In the formula, The energy value corresponding to the initial solution is generated by weighted fusion of the uniform light distribution index and the smoothness index. It comprehensively considers the degree of adaptation of the light intensity distribution map of the target surface to the uniform light distribution standard under the lens sample corresponding to the initial solution, as well as the difficulty of manufacturing the lens sample based on the initial solution. This characterizes the overall quality of the initial solution. The larger the value, the worse the quality of the initial solution, which means that the manufacturing of freeform surface lenses based on the initial solution is less ideal. In the formula, , These are preset proportional coefficients, used to characterize the weighting of uniform light distribution and smoothness in energy value calculations. Because uniform light distribution lens systems require high quality in application, the degree of adaptation to the uniform light distribution standard is more important than manufacturing difficulty. On the constraints, let , , The specific value is set by the staff according to actual needs, such as... and order No restrictions are imposed here; Specifically, for the initial solution, the calculation logic for its corresponding uniform light distribution index is as follows: In the light intensity distribution map of the target surface corresponding to the initial solution, each evaluation sub-region is subjected to a fit analysis with each of the three constraints in the uniform light distribution standard to obtain the fit value of each evaluation sub-region relative to each constraint. A larger fit value indicates a worse fit for the corresponding constraint. The fit values ​​of the same evaluation sub-region relative to each constraint are weighted and fused to obtain the fit index of that evaluation sub-region relative to the uniform light distribution standard. A higher fit index indicates a worse fit for the uniform light distribution standard. The average of the fit indices of all evaluation sub-regions is taken as the uniform light distribution index corresponding to the initial solution. The specific mathematical expression is as follows: ; In the formula, This represents the first element in the light intensity distribution diagram of the target surface under the lens sample corresponding to the initial solution. Each evaluation sub-region This serves as the index for the evaluation sub-region in the target surface light intensity distribution map, and , The number of sub-regions to evaluate in the light intensity distribution map of the target surface; In the formula, This represents the minimum light intensity constraint condition. express The minimum light intensity constraint condition is met. express The fit value for the minimum light intensity constraint, in Seasonal adaptation value when the minimum light intensity constraint is met and in When the minimum light intensity constraint is not met, the seasonal adaptation value is not satisfied. Quantify in this way Adaptability to minimum light intensity constraints; In the formula, This represents the constraint condition for the extreme ratio of light intensity. express The constraint condition of extreme light intensity ratio is satisfied. express The fit value for the extreme ratio constraint of light intensity, in Seasonal adaptation value when satisfying the light intensity extreme ratio constraint condition ,exist The constraint condition of extreme light intensity ratio is not satisfied, so the fitting value is... , express The ratio of the maximum to the minimum internal light intensity. This indicates the ratio of extreme values ​​to the threshold. The larger, the more it means The greater the difference between the maximum and minimum internal light intensity, the more it indicates... The more uneven the light intensity distribution within, the better. and Quantify by relative difference The degree of fit to the constraint condition of extreme light intensity ratio. The larger it is, the more it indicates The greater the deviation from the extreme value of light intensity compared to the constraint conditions; In the formula, This represents the constraint condition for the degree of light intensity variation. express The constraint condition of light intensity variation is satisfied. express The fit value for the constraint of light intensity variation, in Satisfying the minimum light intensity constraint condition for the fit value ,exist The fitting value does not meet the constraint of light intensity variation. , express The ratio of the standard deviation of internal light intensity to the average light intensity. This represents the threshold for light intensity variation. The larger, the more it means The greater the fluctuation in light intensity at different locations within the space, the more it indicates... The more uneven the light intensity distribution within, the better. and Quantify by relative difference The degree of fit to the constraint of light intensity variation. The larger it is, the more it indicates The greater the deviation from the constraint on the degree of light intensity variation; In the formula, express The compatibility index with respect to uniform light distribution standards is determined by... , and The three adaptation values ​​are weighted and fused together to generate a comprehensive evaluation. The higher the overall fit of the three constraints, the better. The worse the compatibility with uniform light distribution standards; In the formula, , , All of these are preset proportional coefficients, used to characterize the proportion of the adaptation values ​​of the evaluation sub-region to the minimum light intensity constraint, the light intensity extreme value ratio constraint, and the light intensity variation degree constraint in the adaptation index calculation. The specific values ​​of the three are determined based on the analytic hierarchy process, which is existing technology and will not be elaborated here. Specifically, for the initial solution, the corresponding smoothness index is calculated as follows: the standard deviation and average value of the thickness values ​​at each control point in the initial solution are calculated, and the ratio of the standard deviation to the average value is used as the smoothness index. This quantifies the volatility of the thickness values ​​at each control point to measure the difficulty of manufacturing a freeform lens based on the initial solution. The specific mathematical expression is as follows:

[0025] In the formula, The standard deviation of the thickness values ​​at each control point in the initial solution. The average thickness value of each control point in the initial solution is the average thickness value of each control point in the initial solution. The ratio of the average thickness value to the average thickness value of each control point in the initial solution is the coefficient of variation of the thickness value of each control point in the initial solution. The larger the value of the coefficient of variation, the greater the fluctuation of the thickness value of each control point in the initial solution, and the more difficult it is to manufacture a freeform lens based on the initial solution. 3) The initial solution is used as the current solution for the first round of annealing optimization. The thickness values ​​of each control point in the current solution are randomly perturbed based on the control point type to generate a new solution. The energy values ​​of the current solution and the new solution are compared to determine the current solution for the next round of annealing optimization. This process is repeated until the annealing temperature is no greater than the termination temperature. The current solution under the last round of annealing optimization is taken as the optimal solution. The optimal three-dimensional mathematical model is determined based on the optimal solution to complete the design of the freeform lens. Subsequently, the freeform lens is manufactured using the optimal three-dimensional mathematical model. The logic for generating a new solution based on the current solution under any round of annealing optimization is as follows: 1) Calculate the ratio of the annealing temperature to the initial temperature under this round of annealing optimization, and use the product of this ratio and the initial value of the perturbation amplitude as the perturbation amplitude under this round of annealing optimization, and the initial value of the perturbation amplitude is a positive decimal less than one. It should be noted that for the first round of annealing optimization, the annealing temperature is the initial temperature. For other rounds of annealing optimization, the annealing temperature is the product of the annealing temperature in the previous round and the cooling rate. The perturbation amplitude is used to control the degree of change of the thickness value of each control point in the new solution compared to the current solution. The specific value of the initial perturbation amplitude can generally be set between 5% and 10%. By using the above method, the perturbation amplitude is gradually reduced during the iterative annealing optimization process, so that a larger change space is provided in the early stage of annealing optimization to perform global exploration and avoid getting trapped in local optima. In the later stage of annealing optimization, the change range is narrowed to perform fine search and facilitate finding the optimal solution. 2) Multiply the perturbation amplitude under this round of annealing optimization by a scaling factor with a positive decimal value to generate the reference perturbation amplitude under this round of annealing optimization. Use the reference perturbation amplitude as the upper limit value and the product of the reference perturbation amplitude and -1 as the lower limit value to construct the reference perturbation interval. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, multiply its thickness value by a random value in the reference perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, sum its thickness value and thickness perturbation value to obtain the thickness value of the control point in the new solution under this round of annealing optimization. It should be noted that since the freeform surface design at the reference control point meets the requirements, the thickness value at the reference control point is within a reasonable range and does not need to be adjusted excessively. Therefore, the perturbation amplitude is scaled by a scaling factor to reduce the value of the reference optimization amplitude, thereby narrowing the reference perturbation range and generating a thickness perturbation value with a smaller absolute value. This allows the thickness value of the reference control point to change slightly during the annealing optimization process to find the optimal value, avoiding excessive changes in the thickness value of the reference control point in the new solution that would deviate from the reasonable range and cause invalid search. The scaling factor can generally be set between 0.1 and 0.3, and the specific value is set by the staff according to the actual situation, which will not be elaborated here. 3) The perturbation amplitude under this round of annealing optimization is used as the upper limit value, and the product of the perturbation amplitude under this round of annealing optimization and -1 is used as the lower limit value to construct the optimization perturbation interval. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, if there is a reference control point whose corresponding sampling point is less than the search radius, then this reference control point is used as its baseline control point, and the search radius is greater than the reduction radius. Based on the distance between the sampling points corresponding to each baseline control point and their corresponding sampling points, the attenuation weight of each baseline control point relative to it is determined based on the Gaussian function. The thickness value of each baseline control point is weighted and fused based on the attenuation weight to obtain the baseline thickness value of the control point. The baseline thickness value is multiplied by a random value in the optimization perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, its thickness value and thickness perturbation value are summed to obtain the thickness value of the control point in the new solution under this round of annealing optimization. It should be noted that setting the search radius to be greater than the reduction radius provides a basis for finding a sufficient number of benchmark control points based on the optimized control points. The specific value of the search radius can be set to 2-5 times the reduction radius, and the specific setting should be made by the staff according to the actual situation. It will not be elaborated here. In practical applications, the search radius can be adjusted in real time according to the location of the optimized control point to ensure that a sufficient number of benchmark control points are found. Each optimized control point should correspond to no less than 5 benchmark control points, so that the benchmark thickness value calculated later has a sufficient number of samples to support it. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, the mathematical expression for calculating its reference thickness value is as follows: ; In the formula, For the first Among all the baseline control points corresponding to the nth optimized control point, the nth The distance between the sampling points corresponding to each baseline control point and the sampling points corresponding to the optimized control point. For all reference control points corresponding to the same optimized control point, the index of each reference control point is given, and , The number of baseline control points corresponding to the same optimized control point. To optimize the indexing of control points, For the first The standard deviation of the distance between the sampling points corresponding to each benchmark control point and the sampling points corresponding to the optimal control point, among all the benchmark control points corresponding to each optimal control point. For the first Among all the baseline control points corresponding to the nth optimized control point, the nth The Gaussian attenuation weights for each baseline control point relative to the optimized control point are chosen as the design standard for attenuation weights because Gaussian attenuation weights have significant advantages in spatial attenuation applications. The larger it is, the more likely it is to be the first Among all the baseline control points corresponding to the nth optimized control point, the nth The greater the distance between the sampling point corresponding to the first baseline control point and the sampling point corresponding to the optimized control point, the more it indicates that the first... The optimized control point and its corresponding _ ... The greater the difference in illumination at each reference control point, the more significant the difference in illumination conditions, indicating that the corresponding first... The less meaningful the thickness value is at each reference control point, the less important the Gaussian attenuation weight becomes. The smaller it is; In the formula, For the first Among all the baseline control points corresponding to the nth optimized control point, the nth The attenuation weights of each baseline control point relative to the optimized control point are determined by... The results were obtained through normalization, which provides a basis for the weighted fusion of the thickness values ​​of each benchmark control point in the following text. In the formula, For the first Among all the baseline control points corresponding to the nth optimized control point, the nth Thickness values ​​at each reference control point For the first The reference thickness value at the optimization control point is obtained by adjusting the _th ... The thickness values ​​at all reference control points corresponding to the i-th optimized control point are obtained by weighted fusion, which integrates all reference control points close to the i-th optimized control point, and is the i-th... The thickness value at the first optimized control point provides a reasonable reference. When generating new solutions subsequently, the thickness value at that optimized control point is generated based on this reasonable reference, rather than based on the first optimized control point. The setting of generating new values ​​for unreasonable thickness values ​​at each optimization control point significantly improves the quality of new solution generation and enhances the search performance of the simulated annealing algorithm. The logic for obtaining the energy value of the new solution is as follows: Based on the double interpolation algorithm, a three-dimensional mathematical model corresponding to the new solution is established on the three-dimensional modeling software, and this three-dimensional mathematical model is input into the optical simulation software for secondary simulation to obtain the light intensity distribution map of the target surface under the three-dimensional mathematical model corresponding to the new solution, and the energy value of the new solution is calculated accordingly. It should be noted that the three-dimensional mathematical model based on the new solution is an existing technology. The specific logic is as follows: a plane identical to the lens sample plane is created on conventional three-dimensional modeling software such as MATLAB and Blender. The spatial position of each control point is determined outside this plane based on the thickness value of each control point in the new solution. The control points are analyzed using bilinear interpolation or bicubic interpolation algorithms to generate a freeform surface. The plane and the freeform surface are merged through Boolean calculations to generate the three-dimensional mathematical model corresponding to the new solution. Both bilinear interpolation algorithms and Boolean calculations are conventional techniques for those skilled in the art and will not be elaborated here. It should be noted that the logic for obtaining the energy value of the new solution is the same as the logic for obtaining the energy value of the initial solution. Specifically, the target surface light intensity distribution map corresponding to the new solution is first divided into sub-regions to be evaluated based on the same method, and then the energy value of the new solution is calculated using the same method as the method for solving the energy value of the initial solution. The specific process has been discussed in the previous text and will not be repeated here. The logic for comparing the current solution and the new solution in the current annealing optimization round to determine the current solution for the next annealing optimization round is as follows: If the energy value of the new solution in the current annealing optimization round is lower than that of the current solution, it means the new solution is superior to the current solution, and the new solution is accepted as the current solution for the next annealing optimization round. Otherwise, the acceptance probability of the new solution is calculated using the Boltzmann formula, and the new solution is probabilistically accepted based on the acceptance probability, according to the Boltzmann formula. Calculating the acceptance probability of a new solution using formulas is a conventional technique for those skilled in the art and will not be elaborated here. Furthermore, after accepting the new solution as the current solution for the next annealing optimization round, when executing the next annealing optimization round, it is necessary to use the same method to find the region to be improved in the light intensity distribution map corresponding to the current solution of the next annealing optimization round, so as to update the control points of the current solution of the next annealing optimization round, to ensure the real-time updating of the control points, thereby ensuring the efficiency of annealing optimization. Specifically, it is also based on the ray tracing algorithm to analyze the region to be improved corresponding to the current solution of the next annealing optimization round, so as to classify the control points set on the free surface corresponding to the current solution of the next annealing optimization round into optimization control points and reference control points according to type, and to perform reduction processing on the reference control points. Example 2

[0026] The present invention also provides a uniform light distribution lens system based on a freeform surface, including a freeform surface lens manufactured using the design method in Embodiment 1 above, and further including a light source for providing illumination, a bracket for fixing the light source, and a light intensity adjustment device for adjusting the brightness of the light source. The light source can be an LED, a laser lamp, or a halogen lamp, etc., and the light intensity adjustment device can specifically be a dimmer for adjusting the brightness of the light source. The light source, the bracket, and the light intensity adjustment device can all use existing devices, which will not be described in detail here.

[0027] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0028] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0029] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0030] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A design method for a freeform surface lens, characterized in that, Includes the following steps: S1. Model the lens sample containing the freeform surface to generate its three-dimensional mathematical model, and input the three-dimensional mathematical model into the optical simulation software to obtain the light intensity distribution map of the target surface, and identify the areas to be improved in the light intensity distribution map that are uneven in light intensity distribution based on the uniform light distribution standard. S2, based on the ray tracing algorithm, analyzes the area to be improved, classifies the control points set on the freeform surface into optimized control points and reference control points according to their type, and performs reduction processing on the reference control points; S3, after randomly perturbing the thickness value based on the type of the control point, the three-dimensional mathematical model is reconstructed by combining the double interpolation algorithm and the second simulation is performed based on optical simulation software. The optimization objectives are to maximize the uniformity of light intensity distribution on the target surface and maximize the smoothness of the freeform surface. The optimal three-dimensional mathematical model is determined by the simulated annealing algorithm.

2. The design method for a freeform surface lens according to claim 1, characterized in that, The logic for identifying the areas to be improved is as follows: 1) Determine the size of the evaluation grid based on the height and width of the light intensity distribution map, and then generate an evaluation network covering the light intensity distribution map, which consists of several evaluation grids. The area of ​​the light intensity distribution map that falls within each evaluation grid is treated as a separate evaluation sub-region. 2) A uniform light distribution standard consisting of minimum light intensity constraint, light intensity extreme ratio constraint, and light intensity variation constraint is preset. Each evaluation sub-region is judged to determine whether it meets the above three constraints at the same time. If it does, it is judged as a qualified region; otherwise, it is judged as a region to be improved.

3. The design method for a freeform surface lens according to claim 2, characterized in that, The logic for dividing the light intensity distribution map into several evaluation sub-regions is as follows: 1.1) Preset a scaling factor that is a positive decimal, calculate the product of the light intensity distribution map height value and the scaling factor as the evaluation grid height value, and calculate the product of the light intensity distribution map width value and the scaling factor as the evaluation grid width value, thereby determining the size of the evaluation grid; 1.2) Lay the first evaluation grid on the light intensity distribution map, and make sure that the geometric center of the evaluation grid coincides with the geometric center of the light intensity distribution map, so as to determine the initial laying position of the evaluation grid; 1.3) Using the newly laid evaluation grid as the center grid, the evaluation grid laying is performed based on the following rules: Rule 1: Determine whether there is a light intensity distribution area directly above the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly above the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid upward by one evaluation grid height value. Rule 2: Determine whether there is a light intensity distribution area directly below the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid directly below the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid down by one evaluation grid height value. Rule 3: Determine whether there is a light intensity distribution area to the left of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the left of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the left by one evaluation grid width value. Rule 4: Determine whether there is a light intensity distribution area to the right of the central grid that is not covered by the existing evaluation grid. If so, lay a new evaluation grid to the right of the central grid, and the position of this evaluation grid is the same as the position after shifting the central grid to the right by one evaluation grid width value. 1.4) Repeat step 1.3) until no new evaluation grid can be laid, thereby generating an evaluation network that covers the light intensity distribution map. Then, the area of ​​the light intensity distribution map that falls within each evaluation grid is treated as a separate evaluation sub-region.

4. The design method for a freeform surface lens according to claim 2, characterized in that, The minimum light intensity constraint condition is: the light intensity value at each location in the evaluation sub-region is not less than the minimum light intensity threshold. The constraint condition for the extreme value ratio of light intensity is: the ratio of the maximum light intensity to the minimum light intensity in the evaluation sub-region is not higher than the extreme value ratio threshold, and the extreme value ratio threshold is greater than one; The constraint condition for the degree of light intensity variation is: the ratio of the standard deviation of light intensity to the average value of light intensity within the evaluation sub-region is not higher than the light intensity variation threshold, and the light intensity variation threshold is a positive decimal less than one.

5. The design method for a freeform surface lens according to claim 2, characterized in that, The logic for setting control points on a freeform surface is as follows: uniform sampling is performed on the plane of the lens sample at a fixed sampling interval to generate several sampling points. For each sampling point, its projection position on the freeform surface along the direction perpendicular to the plane is used as a control point set on the freeform surface.

6. The design method for a freeform surface lens according to claim 5, characterized in that, The logic for classifying control points is as follows: For any control point, the ray tracing function in the optical simulation software is used to simulate the position of the light rays after passing through the control point and incident on the light intensity distribution map. If this position is located in the area to be improved, the control point is used as the optimization control point; otherwise, the control point is used as the reference control point. The logic for reducing reference control points is as follows: Define the sampling point corresponding to the reference control point as the reference sampling point. Randomly select a reference sampling point from all reference sampling points on the plane, and remove all reference sampling points whose distance from the reference sampling point is less than the reduction radius. Iterate in this way until the distance between all remaining reference sampling points on the plane is not less than the reduction radius. Retain the reference control points that correspond one-to-one with the remaining reference sampling points on the plane, and the reduction radius is greater than the sampling interval.

7. The design method for a freeform surface lens according to claim 6, characterized in that, The logic for determining the optimal three-dimensional mathematical model is as follows: 1) Set the initial temperature, final temperature, and cooling rate; 2) Using the thickness value of each control point in the lens sample as the initial solution, and combining it with the light intensity distribution map of the target surface under the initial solution, the energy value of the initial solution is determined. It is generated by weighted fusion of the uniform light distribution index, which characterizes the degree of adaptation of the light intensity distribution map of the target surface to the uniform light distribution standard, and the smoothness index, which characterizes the smoothness of the freeform surface. 3) The initial solution is used as the current solution for the first round of annealing optimization. The thickness values ​​of each control point in the current solution are randomly perturbed based on the control point type to generate a new solution. The energy values ​​of the current solution and the new solution are compared to determine the current solution for the next round of annealing optimization. The iteration is repeated until the annealing temperature is no greater than the termination temperature. The current solution under the last round of annealing optimization is taken as the optimal solution, and the optimal three-dimensional mathematical model is determined based on the optimal solution.

8. The design method for a freeform surface lens according to claim 7, characterized in that, For the initial solution, the calculation logic of its corresponding uniform light distribution index is as follows: In the light intensity distribution map of the target surface corresponding to the initial solution, each evaluation sub-region is subjected to a fit analysis with the three constraints in the uniform light distribution standard to obtain the fit value of each evaluation sub-region relative to each constraint. The fit values ​​of the same evaluation sub-region relative to each constraint are weighted and fused to obtain the fit index of the evaluation sub-region relative to the uniform light distribution standard. The average value of the fit index of all evaluation sub-regions is taken as the uniform light distribution index corresponding to the initial solution. For the initial solution, the corresponding smoothness index is calculated as follows: calculate the standard deviation and average value of the thickness values ​​of each control point in the initial solution, and use the ratio of the standard deviation and average value of the thickness values ​​as the smoothness index.

9. The design method for a freeform surface lens according to claim 7, characterized in that, For any given round of annealing optimization, the logic for generating a new solution based on the current solution is as follows: 1) Calculate the ratio of the annealing temperature to the initial temperature under this round of annealing optimization, and use the product of this ratio and the initial value of the perturbation amplitude as the perturbation amplitude under this round of annealing optimization, and the initial value of the perturbation amplitude is a positive decimal less than one. 2) Multiply the perturbation amplitude under this round of annealing optimization by a scaling factor with a positive decimal value to generate the reference perturbation amplitude under this round of annealing optimization. Use the reference perturbation amplitude as the upper limit value and the product of the reference perturbation amplitude and -1 as the lower limit value to construct the reference perturbation interval. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, multiply its thickness value by a random value in the reference perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point in the current solution under this round of annealing optimization that belongs to the reference control point, sum its thickness value and thickness perturbation value to obtain the thickness value of the control point in the new solution under this round of annealing optimization. 3) The perturbation amplitude under this round of annealing optimization is used as the upper limit, and the product of the perturbation amplitude under this round of annealing optimization and -1 is used as the lower limit to construct the optimization perturbation interval. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, if there exists a reference control point whose distance between its corresponding sampling point and its corresponding sampling point is less than the search radius, then this reference control point is used as its baseline control point, and the search radius is greater than the reduction radius. Based on the distance between the corresponding sampling points of each baseline control point and its corresponding sampling point, the attenuation weight of each baseline control point relative to it is determined based on the Gaussian function. The thickness values ​​of each baseline control point are weighted and fused based on the attenuation weight to obtain the baseline thickness value of the control point. The baseline thickness value is multiplied by a random value in the optimization perturbation interval under this round of annealing optimization to obtain the thickness perturbation value of the control point under this round of annealing optimization. For any control point belonging to the optimization control point in the current solution under this round of annealing optimization, its thickness value and thickness perturbation value are summed to obtain the thickness value of this control point in the new solution under this round of annealing optimization.

10. A uniform light distribution lens system based on a freeform surface, comprising a freeform surface lens manufactured using the design method of any one of claims 1-9, characterized in that: It also includes a light source for providing illumination, a bracket for fixing the light source, and a light intensity adjustment device for adjusting the brightness of the light source.

Images