Optimization method and system for multi-target lens arrangement of light spot shaping

Abstract

The application discloses a kind of multi-target lens arrangement optimization method and system of light spot shaping, comprising: according to the input target light spot parameter, the mapping model between target light spot and lens optical parameter is established, and the transmission relationship of input and output light spot is obtained;Multi-objective optimization model is constructed with the minimum system volume, the maximum energy efficiency, the maximum light spot shaping precision as target, quantitative energy loss model is introduced, non-dominated sorting genetic algorithm is used to solve, and the optimal lens combination, arrangement order and space interval are obtained;Real-time acquisition output light spot image and collimation degree deviation are calculated, when collimation degree deviation exceeds threshold value, proportional integral derivative adaptive algorithm is used to drive fine adjustment mechanism to carry out position correction, and stable shaping light spot is obtained.The application realizes the automatic arrangement of lens, without manual intervention, effectively reduces the system volume, improves energy efficiency and ensures light spot collimation degree, and can be widely applied in precision instrument optical experiment, scientific research and other fields.

Description

Technical Field

[0001] This invention belongs to the field of laser optics and intelligent optimization technology, and particularly relates to a method and system for optimizing the arrangement of multi-target lenses for spot shaping. Background Technology

[0002] Laser spot shaping is an indispensable and crucial step in laser processing, precision measurement, medical equipment, and scientific research. After laser beams exit the light source, they often exhibit problems such as shape asymmetry, large differences in divergence angles, and uneven energy distribution, directly affecting the coupling efficiency, processing accuracy, and experimental results of subsequent optical systems. Especially in precision instrument optical experiments, the shape, size, and collimation of the laser spot must be precisely controllable; otherwise, measurement errors, energy loss, or instrument performance degradation will occur. Existing technologies include several spot shaping solutions, such as using two cylindrical lenses closely arranged and manually adjusting their positions, or using a controller in conjunction with mechanical components like apertures and guide rails to adjust the divergence angle, and using an entrance aperture control device in conjunction with a scanning galvanometer to achieve shaping. These solutions improve spot quality to some extent, but they share the common characteristic of relying on fixed hardware combinations, manual adjustment, or preset mechanical drives, lacking an intelligent, automated arrangement framework.

[0003] However, existing technologies still suffer from the following problems: First, determining the lens combination and arrangement order relies on manual selection and adjustment, resulting in low operational efficiency and difficulty in guaranteeing optimality. Second, there is a lack of a multi-objective optimization framework that simultaneously considers system miniaturization, maximizing energy efficiency, and spot shaping accuracy (including collimation). Often, only one aspect can be optimized, leading to large system size, high energy loss, or unstable collimation. Third, a quantitative energy loss model incorporating Fresnel reflection, material absorption, optical path loss, and surface scattering has not been established, making it impossible to evaluate energy utilization in real time during optimization. Fourth, closed-loop control capabilities are weak; most schemes only provide limited feedback on the divergence angle and do not cover real-time sub-micron level correction of spot collimation, making it difficult to meet the long-term stability requirements of precision instrument optical experiments. Therefore, there is an urgent need for a spot shaping scheme that can automatically select lens combinations, perform multi-objective collaborative optimization, and possess real-time closed-loop control capabilities. Summary of the Invention

[0004] This invention aims to address the problems of existing spot shaping technologies, such as heavy reliance on manual intervention, inability to balance system miniaturization, energy efficiency and collimation, lack of automatic optimization framework and quantitative energy loss model. It provides a multi-target lens arrangement optimization method and system for spot shaping, which automatically selects lens combinations and optimizes the arrangement order and spacing after inputting target spot parameters, without manual intervention, effectively reducing system size, improving energy efficiency and ensuring spot collimation.

[0005] To achieve the above objectives, the present invention provides a method and system for optimizing the arrangement of multiple target lenses for beam reshaping. Specifically, the method for optimizing the arrangement of multiple target lenses for beam reshaping includes: Based on the input target spot parameters, a mapping model between the target spot and the lens optical parameters is established using ray tracing or matrix optical transmission theory to obtain the transmission relationship between the input and output spots. Based on the transmission relationship, a multi-objective optimization model is constructed with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam reshaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. Based on the optimal lens combination, arrangement order, and spatial spacing, the output light spot image is acquired in real time and the collimation deviation is calculated. When the collimation deviation exceeds the threshold, the proportional-integral-derivative adaptive algorithm is used to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

[0006] Preferably, the process of obtaining the transmission relationship includes: The mapping between input and output spot parameters is described using the overall system transmission matrix. The total transmission matrix of the system is obtained by cascading and multiplying the transmission matrices of each lens. The transmission matrix of each lens is determined by the focal length, tilt angle, and distance between the lens and its adjacent lenses. The input spot parameters include the size and divergence angle of the input spot in two vertical directions; The output spot parameters include the size and divergence angle of the output spot in two vertical directions; By employing matrix optical transmission theory, the refraction, propagation, and tilting effects of a single lens are integrated into a single transmission matrix. The total transmission matrix of the system is obtained by cascading multiple lens transmission matrices, enabling numerical prediction of the size, energy distribution, and collimation deviation of the output spot.

[0007] Preferably, the process of constructing the multi-objective optimization model includes: With minimizing the total system volume as the primary objective, the total system volume is calculated by summing the axial lengths of each lens and the spacing between each lens. With maximizing energy efficiency as the second objective, the energy efficiency is obtained by dividing the difference between the total incident energy and the total energy loss by the total incident energy. With maximizing the accuracy of light spot shaping as the third objective, the light spot shaping error is calculated by combining the deviation between the actual light intensity and the target light intensity at each sampling point of the output light spot and the collimation deviation.

[0008] Preferably, the process of obtaining the optimal lens combination, arrangement order, and spatial spacing includes: In each iteration of the non-dominated sorting genetic algorithm, the energy loss rate and energy efficiency are calculated for each lens arrangement scheme in the population, and invalid schemes that do not meet the preset energy efficiency lower limit constraint are eliminated.

[0009] Preferably, the process of acquiring and calculating the collimation deviation in real time through the closed-loop control module includes: The system acquires and outputs light spot images, and performs preprocessing such as noise reduction, threshold segmentation, and edge extraction. Calculate the actual size and energy distribution of the light spot, and calculate the collimation deviation based on the equivalent effective focal length of the system.

[0010] Preferably, the process of calculating the collimation deviation includes: The collimation deviation is obtained by dividing the change in the spot radius by the equivalent effective focal length of the system. A proportional-integral-derivative (PID) control law is used to generate lens fine-tuning displacement commands. The proportional coefficient, integral coefficient, and derivative coefficient in the PID control law are adaptively adjusted based on the collimation error signal.

[0011] The present invention also provides a multi-target lens arrangement optimization system for beam shaping, comprising: A laser source is used to output an initial laser beam to obtain the original beam to be shaped. The lens library is used to store the optical parameters of various optical elements, and to obtain a dataset of lenses to choose from. The mapping model module is used to establish a mapping model between the target spot and the lens optical parameters based on the input target spot parameters and the lens optical parameters in the lens library, using ray tracing or matrix optical transmission theory, to obtain the transmission relationship between the input and output spots. The multi-objective optimization module is used to construct a multi-objective optimization model based on the transmission relationship, with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam pattern shaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. The energy loss calculation module is used to calculate the energy loss rate and energy efficiency of each lens arrangement scheme in real time during the solution process of the multi-objective optimization module, and to obtain energy efficiency verification data and feed it back to the multi-objective optimization module. The closed-loop control module is used to acquire and output spot images in real time and calculate the collimation deviation to obtain the collimation error signal. The controller is bidirectionally connected to the mapping model module, the multi-objective optimization module, the energy loss calculation module, and the closed-loop control module, respectively. It is used to determine whether the deviation exceeds the threshold based on the collimation error signal, and when the deviation exceeds the threshold, it uses a proportional-integral-derivative adaptive algorithm to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

[0012] Preferably, the mapping model module includes: The transfer matrix construction unit is used to integrate the refraction effect, propagation effect and tilting effect of a single lens into a single transfer matrix, thereby obtaining the independent transfer matrix of each lens. The cascaded operation unit is used to cascade and multiply the independent transmission matrices of each lens according to the lens arrangement order to obtain the total transmission matrix of the system. The parameter prediction unit is used to numerically predict the size, energy distribution, and collimation deviation of the output spot based on the total transmission matrix of the system and the input spot parameters, thereby obtaining the transmission relationship between the input and output spots.

[0013] Preferably, the multi-objective optimization module includes: The objective function construction unit is used to construct three core objective functions with the goal of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam spot shaping accuracy, thereby obtaining the optimization objective set. The algorithm initialization unit is used to encode the lens type, quantity, arrangement order, spacing and tilt angle, randomly generate an initial scheme population, and obtain a set of lens arrangement schemes to be evaluated. The fitness calculation unit is used to send each lens arrangement scheme in the population to the energy loss calculation module, obtain the energy loss rate and energy efficiency of each scheme, and calculate the fitness value of each individual to obtain the non-dominated ranking and crowding results. The iterative evolutionary unit is used to perform selection, crossover, and mutation operations to generate offspring populations. After merging the parent and offspring populations, the non-dominated sorting is performed again and elite individuals are selected to obtain the Pareto optimal solution set. The scheme selection unit is used to select the lens arrangement scheme with the best overall performance from the Pareto optimal solution set according to the preset weight coefficients, so as to obtain the optimal lens combination, arrangement order and spatial spacing.

[0014] Preferably, the energy loss calculation module includes: The parameter reading unit is used to receive the lens arrangement scheme issued by the multi-objective optimization module, and read the material refractive index, laser incident angle, material absorption coefficient, optical path length, surface roughness and aperture size of each lens in the scheme to obtain the basic data for loss calculation of each lens. The Fresnel reflection calculation unit is used to calculate the reflection loss of the two optical surfaces of each lens based on the material refractive index and the laser incident angle, and to obtain the Fresnel reflectivity of each lens. The material absorption calculation unit is used to calculate the material absorption attenuation factor based on the material absorption coefficient and the optical path length inside the lens, using the Beer-Lambert law, to obtain the material absorption loss of each lens. The surface scattering calculation unit is used to calculate the surface scattering loss rate based on the surface roughness and working wavelength of each lens, and obtain the scattering loss of each lens; The total loss synthesis unit is used to multiply the Fresnel reflectivity, material absorption attenuation factor and surface scattering loss rate of each lens, and then subtract the product from one to obtain the total energy loss rate. Based on the total energy loss rate, the energy efficiency is calculated to obtain energy efficiency verification data.

[0015] Compared with the prior art, the present invention has the following advantages and technical effects: First, by constructing a mapping model and adopting a multi-objective optimization algorithm, intelligent automatic selection and arrangement of lens type, quantity, arrangement order and spatial spacing are realized, completely eliminating manual debugging and selection, and significantly improving operational efficiency and optimal solution.

[0016] Second, by constructing a multi-objective optimization model with the core objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam shaping accuracy, and by introducing a quantitative energy loss model during the optimization process, a synergistic balance between system miniaturization, high energy efficiency, and high beam shaping accuracy is achieved, avoiding the shortcomings of traditional solutions that sacrifice one aspect for another.

[0017] Third, by establishing a quantitative energy loss model covering all elements such as Fresnel reflection, material absorption, optical path loss, and surface scattering, the energy efficiency of each scheme can be calculated in real time during the optimization iteration, and invalid schemes can be eliminated, ensuring that the final selected lens arrangement scheme achieves optimal energy utilization.

[0018] Fourth, the closed-loop control module collects and outputs the light spot image in real time, calculates the collimation deviation, and uses a proportional-integral-derivative adaptive algorithm to drive the sub-micron fine-tuning mechanism to perform position correction when the deviation exceeds the threshold. This forms a real-time closed-loop feedback mechanism, which effectively ensures the long-term stability of the light spot shaping accuracy and collimation, and meets the requirements of high precision and high stability for optical experiments of precision instruments. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1This is a schematic diagram of the system structure according to an embodiment of the present invention; Figure 2 This is a flowchart of the mapping model module in an embodiment of the present invention; Figure 3 This is a flowchart of the multi-objective optimization module according to an embodiment of the present invention; Figure 4 This is a flowchart of the energy loss calculation module in an embodiment of the present invention; Figure 5 This is a flowchart of the closed-loop control module in an embodiment of the present invention. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0022] Example 1 This embodiment provides a multi-target lens arrangement optimization method for spot shaping, including: Based on the input target spot parameters, a mapping model between the target spot and the lens optical parameters is established using ray tracing or matrix optical transmission theory to obtain the transmission relationship between the input and output spots. Based on the transmission relationship, a multi-objective optimization model is constructed with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam pattern shaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. Based on the optimal lens combination, arrangement order and spatial spacing, the output light spot image is acquired in real time and the collimation deviation is calculated. When the collimation deviation exceeds the threshold, the proportional-integral-derivative adaptive algorithm is used to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

[0023] This embodiment achieves intelligent automatic lens arrangement by constructing a mapping model, a multi-objective optimization framework, and a quantitative energy loss model, eliminating the need for manual adjustment and selection. While meeting the requirements for beam shaping, this embodiment effectively reduces system size, improves energy efficiency, and ensures beam collimation, thereby enhancing the stability and repeatability of optical experiments on precision instruments.

[0024] This embodiment constructs a unified multi-objective optimization framework, which, unlike traditional methods that rely on fixed hardware combinations or manual adjustments, enables intelligent and automatic selection and arrangement of lens type, order, and spacing based on target spot parameters. Simultaneously, it introduces a quantitative energy loss model and a closed-loop collimation feedback mechanism, achieving a synergistic balance between minimizing system volume, maximizing energy efficiency, and optimizing spot shaping accuracy (including collimation). Furthermore, this embodiment requires no manual intervention or debugging throughout the entire process, significantly improving the operational efficiency and accuracy of spot shaping. It exhibits clear advantages in automation, energy loss control, and collimation stability, possessing greater versatility and technological transfer potential.

[0025] Furthermore, the process of obtaining the transmission relationship includes: The mapping between input and output spot parameters is described using the overall system transmission matrix. The overall system transmission matrix is ​​obtained by cascading and multiplying the transmission matrices of each lens. The transmission matrix of each lens is determined by the lens's focal length, tilt angle, and distance from adjacent lenses. The input spot parameters include the size and divergence angle of the input spot in two vertical directions; The output spot parameters include the size and divergence angle of the output spot in two vertical directions; By employing matrix optical transmission theory, the refraction, propagation, and tilting effects of a single lens are integrated into a single transmission matrix. The total transmission matrix of the system is obtained by cascading multiple lens transmission matrices, enabling numerical prediction of the size, energy distribution, and collimation deviation of the output spot.

[0026] Furthermore, the process of constructing a multi-objective optimization model includes: With minimizing the total system volume as the primary objective, the total system volume is calculated by summing the axial lengths of each lens and the spacing between each lens. With maximizing energy efficiency as the secondary objective, energy efficiency is obtained by dividing the difference between the total incident energy and the total energy loss by the total incident energy. With maximizing the accuracy of light spot shaping as the third objective, the light spot shaping error is calculated by combining the deviation between the actual light intensity and the target light intensity at each sampling point of the output light spot and the collimation deviation.

[0027] Furthermore, the process of obtaining the optimal lens combination, arrangement order, and spatial spacing includes: In each iteration of the non-dominated sorting genetic algorithm, the energy loss rate and energy efficiency are calculated for each lens arrangement scheme in the population, and invalid schemes that do not meet the preset energy efficiency lower limit constraint are eliminated.

[0028] Furthermore, the process of acquiring and calculating the collimation deviation in real time through the closed-loop control module includes: The system acquires and outputs light spot images, and performs preprocessing such as noise reduction, threshold segmentation, and edge extraction. Calculate the actual size and energy distribution of the light spot, and calculate the collimation deviation based on the equivalent effective focal length of the system.

[0029] Furthermore, the process of calculating the collimation deviation includes: The collimation deviation is obtained by dividing the change in the spot radius by the equivalent effective focal length of the system. The lens fine-tuning displacement command is generated using a proportional-integral-derivative (PID) control law. The proportional coefficient, integral coefficient, and derivative coefficient in the PID control law are adaptively adjusted based on the collimation error signal.

[0030] As a preferred implementation, the multi-target lens arrangement optimization method for beam spot shaping in this embodiment sequentially arranges a laser source, a lens library, a beam sampling beam splitter, and an optical output window along the optical path. The transmission optical path of the beam sampling beam splitter is connected to the optical output window, and the reflection optical path is connected to a high-resolution CCD camera. The signal output terminal of the high-resolution CCD camera is connected to the signal input terminal of the closed-loop control module, and the signal output terminal of the closed-loop control module is connected to the controller. The signal output terminal of the parameter input terminal is connected to the controller, and the controller is bidirectionally connected to the mapping model module, the multi-target optimization module, and the energy loss calculation module. Simultaneously, the drive output terminal of the controller is connected to the sub-micron level fine-tuning mechanism built into the lens library, forming a complete optical transmission link and electrical signal control closed loop. Specifically, it includes the following steps: Step 1: Based on the input target spot parameters, establish a mapping model between the target spot and the lens optical parameters using ray tracing or ABCD matrix optical transmission theory to determine the transmission relationship between the input and output spots. Step 2: Construct a multi-objective optimization model with the core objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam pattern shaping accuracy. Use the NSGA-II non-dominated sorting genetic algorithm to solve for the optimal lens combination, arrangement order, and spatial spacing. Step 3: Establish a quantitative energy loss model that integrates Fresnel reflection, material absorption, optical path loss, and surface scattering. During the optimization process, calculate the energy loss rate and energy efficiency of each lens arrangement scheme in real time. Step 4: The closed-loop control module acquires and outputs the light spot image in real time, calculates the collimation deviation of the light spot, and uses a PID adaptive algorithm to drive the fine-tuning mechanism to complete sub-micron level position correction when the deviation exceeds the threshold, so as to ensure the accuracy of light spot shaping and the stability of collimation.

[0031] Furthermore, step 1 includes the following mapping equation: in, For the system's total transmission matrix, Let be the transfer matrix of the i-th lens. Let be the focal length of the i-th lens. Let be the tilt angle of the i-th lens. Let be the distance between the i-th lens and its adjacent lens. , To input the dimensions of the light spot in the X and Y directions, , Input the divergence angles of the light spot in the X and Y directions. , To output the size of the light spot in the X and Y directions, , To output the divergence angles of the light spot in the X and Y directions.

[0032] Specifically, this step is executed through the mapping model module. Its core is to establish a precise mapping relationship between the input light spot, lens optical parameters, and the output light spot using ray tracing or ABCD matrix optical transmission theory, thus determining the transmission relationship between the input and output light spots and providing a numerical basis for subsequent optimization. The specific implementation process is as follows: First, the mapping model module receives the initial laser beam parameters and target spot parameters from the controller, and reads the X-direction spot size of the initial laser beam. Y-direction spot size X-direction divergence angle Y-direction divergence angle Simultaneously read the shape and size of the target light spot. Target divergence angle , Target energy distribution curve.

[0033] Secondly, the optical transmission calculation method can be selected according to the application scenario. For paraxial optics scenarios, the ABCD matrix optical transmission theory is chosen, while for non-paraxial or complex curved lens scenarios, ray tracing is used. This embodiment uses the ABCD matrix optical transmission theory as its core, retrieving the basic parameters of all optical elements in the lens library to construct an independent transmission matrix for each lens. This matrix integrates the refraction, propagation, and tilt effects of a lens, with matrix elements derived from the lens's focal length. Inclination angle Spacing with adjacent lenses The only certainty.

[0034] Subsequently, the total system transmission matrix is ​​calculated through the cascade operation of the multi-lens transmission matrices. This leads to the establishment of a mapping equation between the input and output light spots. Using this equation, the mapping model module can rapidly predict the characteristics of the output light spot under arbitrary lens combinations, arrangements, and spacing through numerical simulation. This includes the light spot size, divergence angle, and energy distribution. It also calculates the size deviation, energy distribution deviation, and collimation deviation between the simulated output light spot and the target light spot. The mapping relationship and deviation data are transmitted in real-time to the multi-objective optimization module and simultaneously uploaded to the controller for archiving, thus completing the construction of the mapping model.

[0035] Specifically, in step 1, the mapping model module adopts the ABCD matrix optical transmission theory to integrate the refraction effect, propagation effect and tilt effect of a single lens into a single transmission matrix. Then, the total transmission matrix of the system is obtained by cascading multiple lens transmission matrices, so as to realize the rapid numerical prediction of input spot parameters and output spot characteristics (size, energy distribution, collimation deviation).

[0036] Furthermore, step 2 includes a multi-objective optimization objective function, expressed as follows: Where V is the total volume of the system, L i Let L be the axial length of the i-th lens, η be the energy efficiency, and L be the axial length of the i-th lens. loss Here, E represents the total energy loss rate, E represents the spot shaping error, and M represents the number of spot energy sampling points. Output the light intensity for the j-th sampling point. Let be the target light intensity at the j-th sampling point, and Δθ be the collimation divergence angle deviation.

[0037] This step is executed through a multi-objective optimization module. Its core is to construct a multi-objective optimization model focused on minimizing the overall system volume, maximizing energy efficiency, and maximizing beam pattern shaping accuracy. The NSGA-II non-dominated sorting genetic algorithm is used to solve for the optimal lens combination, arrangement order, and spatial spacing. The specific implementation process is as follows: First, the multi-objective optimization module receives the mapping data and bias data output by the mapping model module and constructs three core objective functions. Second, it sets the operating parameters of the NSGA-II algorithm and, based on the input constraints, sets hard constraints for the system's maximum volume, minimum energy efficiency, and maximum permissible bias.

[0038] Subsequently, the algorithm initializes the population, encoding the lens type, quantity, arrangement order, spacing, and tilt angle, and randomly generates an initial scheme population. All schemes within the population are sent to the energy loss calculation module to obtain the energy loss rate and energy efficiency of each scheme, eliminating invalid schemes that do not meet the hard constraint of energy efficiency. The fitness value of each individual in the population is calculated, non-dominated sorting and crowding calculation are completed, and selection, crossover, and mutation operations are performed to generate a new generation of offspring. The parent and offspring populations are merged, non-dominated sorting is performed again, and elite individuals are selected to be retained for the next generation. During the iteration process, it is continuously checked whether the algorithm has reached the maximum number of iterations or the convergence threshold. If not, the population is updated and the above iteration process is repeated. If the convergence condition is met, the Pareto optimal solution set is output. Based on the weight coefficients preset by the operator, the lens arrangement scheme with the best overall performance is selected from the optimal solution set to determine the optimal lens type and quantity, arrangement order, spatial spacing, and tilt angle. The optimal scheme is then output to the controller and the energy loss calculation module to complete the optimization solution process.

[0039] Furthermore, step 3 includes a quantitative energy loss model, expressed as follows: in, Let be the Fresnel reflectivity of the i-th lens. Let L be the absorption coefficient of the i-th lens material. i Let S be the optical path length within the i-th lens. i Let be the scattering loss rate of the i-th lens surface.

[0040] This step is executed through the energy loss calculation module. Its core is to establish a quantitative energy loss model that integrates Fresnel reflection, material absorption, optical path loss, and surface scattering. During the optimization process, the energy loss rate and energy efficiency of each lens arrangement scheme are calculated in real time, providing the objective function basis for multi-objective optimization. Simultaneously, the energy efficiency of the final optimal scheme is verified. The specific implementation process is as follows: First, the energy loss calculation module receives the lens arrangement scheme from the multi-objective optimization module and reads the parameters of each lens in the scheme, including the material refractive index, laser incident angle, and material absorption coefficient. , the optical path length L inside the lens i Surface roughness and aperture size.

[0041] Secondly, calculate the loss components of each lens individually: one is Fresnel reflectivity. The reflection loss of the two optical surfaces of the lens is calculated using Fresnel's formula based on the refractive index of the lens material and the laser incident angle; the absorption loss is calculated using Beer-Lambert's law. Thirdly, the surface scattering loss rate S iThe energy loss rate is calculated based on the lens surface roughness and the operating wavelength. Then, based on the individual lens loss component, the total energy loss rate is calculated. and energy efficiency The system determines whether the energy efficiency meets the preset minimum constraint threshold. If it does not, the solution is marked as invalid, and a rejection instruction is sent to the multi-objective optimization module. If it does, it is marked as valid, and the energy efficiency data is fed back to the multi-objective optimization module in real time. Simultaneously, the loss component data, total loss rate, and energy efficiency of the solution are uploaded to the controller for archiving. In each iteration of the multi-objective optimization, the energy loss calculation module performs the above calculation process to ensure that the optimization process always considers energy efficiency indicators, avoiding the problem of traditional design focusing only on shaping effects while ignoring energy waste. After the optimal solution is determined, the energy loss calculation module performs a full-element energy efficiency review of the optimal solution and outputs the final energy efficiency report to ensure that the solution meets the design requirements.

[0042] Furthermore, step 4 includes the collimation deviation calculation formula and the PID control law, the expressions of which are as follows: Where Δθ is the collimation deviation, Δw is the change in spot radius, and f eff Let u(t) be the equivalent effective focal length of the system, u(t) be the lens fine-tuning displacement command, e(t) be the collimation error signal, and K be the collimation error signal. p K is the proportionality coefficient. i K is the integral coefficient. d is the differential coefficient.

[0043] This step is executed collaboratively by the closed-loop control module and the controller. The core mechanism involves the closed-loop control module acquiring the output light spot image in real time, calculating the collimation deviation, and then using a PID adaptive algorithm to drive the fine-tuning mechanism to complete sub-micron level position correction when the deviation exceeds a threshold. This ensures the long-term stability of the light spot shaping accuracy and collimation. The specific implementation process is as follows: First, the controller, based on the optimal scheme, drives the sub-micron level fine-tuning mechanism of the lens library to complete the assembly, initial positioning, and spacing adjustment of the lenses. The laser source outputs an initial laser beam, which, after being shaped by the optimized lens combination within the lens library, is split by the beam sampling beam splitter. The transmitted light is output through the optical output window, and the reflected light is incident on the high-resolution CCD camera. Second, the image processing unit of the closed-loop control module receives the spot image acquired by the CCD camera, performs preprocessing such as noise reduction, threshold segmentation, and edge extraction, and calculates the actual parameters of the spot, including the X / Y direction dimensions, two-dimensional energy distribution, and the spot radius change Δw. The collimation calculation unit 12 calculates the collimation based on the system's equivalent effective focal length f. effThe collimation deviation Δθ is calculated. Simultaneously, combining the actual energy distribution of the light spot with the target energy distribution, the total light spot shaping error E is calculated, and it is determined whether the total error E or the collimation deviation Δθ exceeds a preset threshold. If the deviation does not exceed the threshold, no adjustment action is performed, and the spot image is continuously acquired in real time to maintain the monitoring status; if the deviation exceeds the threshold, the PID adaptive control unit 13 is activated, the current error signal e(t) is read, and the PID parameters, including the proportional coefficient K, are adaptively adjusted. p Integral coefficient K i Differential coefficient K d The controller calculates the lens fine-tuning displacement command u(t) based on the PID control law. After receiving the displacement command, the controller drives the submicron-level fine-tuning mechanism to complete the submicron-level precision correction of the position, spacing, or tilt angle of the corresponding lens. After the correction is completed, the spot image is re-acquired, and the deviation is judged and corrected again to form a real-time closed-loop feedback until the spot deviation returns to the threshold range. Then, the lens position is locked, the spot status is continuously monitored, and the fully automatic spot shaping process is completed.

[0044] This embodiment achieves automatic optimization and arrangement of lens combinations, arrangement order, and spatial spacing by establishing a precise mapping model between the target spot and lens parameters, a multi-objective intelligent optimization algorithm, a full-element energy loss model, and a real-time closed-loop collimation control mechanism. High-precision spot shaping can be completed without manual adjustment. This embodiment can achieve a synergistic balance between minimizing system volume, maximizing energy transmission efficiency, and optimizing shaping accuracy while meeting the requirements for target spot shape, size, energy distribution, and collimation. It is compatible with various spot shapes such as circles, squares, rings, and ellipses, and is suitable for applications such as precision instrument optical experiments and quantum precision measurement, providing an efficient, stable, and universal automated solution for high-precision laser application technology.

[0045] Example 2 like Figure 1 As shown, based on the same inventive concept, this embodiment also provides a multi-target lens arrangement optimization system for spot shaping, including: A laser source is used to output an initial laser beam to obtain the original beam to be shaped. The lens library is used to store the optical parameters of various optical elements, and to obtain a dataset of lenses to choose from. The mapping model module is used to establish a mapping model between the target spot and the lens optical parameters based on the input target spot parameters and the lens optical parameters in the lens library, using ray tracing or matrix optical transmission theory, to obtain the transmission relationship between the input and output spots. The multi-objective optimization module is used to construct a multi-objective optimization model based on the transmission relationship, with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam pattern shaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. The energy loss calculation module is used to calculate the energy loss rate and energy efficiency of each lens arrangement scheme in real time during the solution process of the multi-objective optimization module, and to obtain energy efficiency verification data to be fed back to the multi-objective optimization module. The closed-loop control module is used to acquire and output spot images in real time and calculate the collimation deviation to obtain the collimation error signal. The controller is bidirectionally connected to the mapping model module, the multi-objective optimization module, the energy loss calculation module, and the closed-loop control module. It is used to determine whether the deviation exceeds the threshold based on the collimation error signal, and when the deviation exceeds the threshold, it uses a proportional-integral-derivative adaptive algorithm to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

[0046] Furthermore, the system in this embodiment includes a laser source, a lens library, a mapping model module, a multi-objective optimization module, an energy loss calculation module, a closed-loop control module, and a controller. The laser source is a semiconductor laser or a solid-state laser, providing the initial laser beam. The lens library stores optical parameters of various optical elements, including cylindrical lenses, aspherical lenses, freeform lenses, and prisms. For each lens, detailed information such as focal length, radius of curvature, material refractive index, aperture size, and surface coating is recorded.

[0047] In this embodiment, the system is arranged along the optical path as follows: a laser source, a lens library, a mapping model module, a multi-objective optimization module, an energy loss calculation module, and a closed-loop control module. Each module is connected to the controller, forming a complete signal transmission and control closed loop. The laser source outputs the initial laser beam, providing the shaping object for the system; the lens library provides standardized optical components and parameter databases, supporting optimized selection and automatic assembly; the mapping model module establishes the mathematical mapping relationship between the input light spot, lens parameters, and the output light spot, enabling rapid prediction of the light spot output effect; the multi-objective optimization module intelligently seeks optimization based on volume, efficiency, and accuracy, outputting the optimal lens arrangement scheme; the energy loss calculation module verifies the energy efficiency of the optimized scheme, ensuring that the system's energy utilization rate meets requirements; the closed-loop control module collects light spot information in real time and performs collimation correction, maintaining long-term stability of the output light spot; the controller uniformly schedules the work of each module, realizing fully automated operation of the target parameter input, modeling, optimization, verification, and fine-tuning process.

[0048] This embodiment, through its modular design, is compatible with various target spot shapes such as circles, squares, and rings, making it widely applicable. It can be used in multiple fields such as precision instrument optical experiments and cutting-edge scientific research, and can improve the shortcomings of existing technologies such as low efficiency, high energy loss, and insufficient collimation stability.

[0049] Furthermore, the mapping model module includes: The transfer matrix construction unit is used to integrate the refraction effect, propagation effect and tilting effect of a single lens into a single transfer matrix, thereby obtaining the independent transfer matrix of each lens. The cascaded operation unit is used to cascade and multiply the independent transmission matrices of each lens according to the lens arrangement order to obtain the total transmission matrix of the system. The parameter prediction unit is used to numerically predict the size, energy distribution, and collimation deviation of the output spot based on the total transmission matrix of the system and the input spot parameters, thereby obtaining the transmission relationship between the input and output spots.

[0050] Furthermore, the mapping model module in this embodiment employs ray tracing or ABCD matrix optical transmission theory to establish the relationship between the target light spot and lens parameters (focal length) based on the input target light spot parameters (shape: circular, square, or annular; size: diameter or side length; collimation: divergence angle deviation). Inclination angle ,spacing The exact mapping relationship (order) is satisfied by the following mapping equation: Among them, The transmission matrix of the i-th lens includes refraction, propagation, and tilt effects. This module rapidly predicts the output spot characteristics under arbitrary lens combinations through numerical simulation, providing a basic mapping for subsequent optimization.

[0051] like Figure 2 As shown, the mapping model module, based on the ABCD matrix optical transmission theory, integrates the refraction, propagation, and tilting effects of a single lens into a single transmission matrix. This is then achieved by cascading multiple lens transmission matrices to obtain the overall system transmission matrix, establishing a definite mapping relationship between the input and output light spots. During operation, the mapping model module reads the initial laser beam parameters and the target light spot parameters. Through numerical simulation, it rapidly calculates the output light spot characteristics corresponding to any lens combination. Data such as light spot size error, energy distribution deviation, and collimation deviation are transmitted in real-time to the multi-objective optimization module, providing reliable data support for optimization calculations and ensuring that the optimization process has a clear numerical basis and convergence direction.

[0052] Furthermore, the multi-objective optimization module includes: The objective function construction unit is used to construct three core objective functions with the goal of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam spot shaping accuracy, thereby obtaining the optimization objective set. The algorithm initialization unit is used to encode the lens type, quantity, arrangement order, spacing and tilt angle, randomly generate an initial scheme population, and obtain a set of lens arrangement schemes to be evaluated. The fitness calculation unit is used to send each lens arrangement scheme in the population to the energy loss calculation module, obtain the energy loss rate and energy efficiency of each scheme, and calculate the fitness value of each individual to obtain the non-dominated ranking and crowding results. The iterative evolutionary unit is used to perform selection, crossover, and mutation operations to generate offspring populations. After merging the parent and offspring populations, the non-dominated sorting is performed again and elite individuals are selected to obtain the Pareto optimal solution set. The scheme selection unit is used to select the lens arrangement scheme with the best overall performance from the Pareto optimal solution set according to the preset weight coefficients, so as to obtain the optimal lens combination, arrangement order and spatial spacing.

[0053] Furthermore, the multi-objective optimization module in this embodiment constructs an optimization model containing three core objectives: minimizing the overall system volume, maximizing energy efficiency, and maximizing spot shaping accuracy (including minimizing collimation deviation). The NSGA-II (Non-Dominated Sorting Genetic Algorithm II) is used to solve for the Pareto optimal solution, automatically selecting the optimal lens combination, arrangement order, and spatial spacing. The objective function is defined as: in, The total system volume is calculated by summing the lens dimensions and spacing. For energy efficiency, Spot error (including intensity distribution deviation and collimation deviation) ), Let be the length of the i-th lens. Where M is the loss rate and M is the number of sampling points.

[0054] like Figure 3 As shown, the multi-objective optimization module uses the NSGA-II non-dominated sorting genetic algorithm as its core. The process begins by receiving the mapping data and bias data from the mapping model. First, it constructs three core objective functions: minimizing system volume, maximizing energy efficiency, and maximizing shaping accuracy. After configuring the algorithm parameters and constraints, it initializes the population. After multiple rounds of iterative genetic operations, non-dominated sorting, and elite selection, it outputs the Pareto optimal solution set after convergence. Finally, it selects the comprehensive optimal lens arrangement scheme to achieve a cooperative balance of multiple objectives.

[0055] Furthermore, the energy loss calculation module includes: The parameter reading unit is used to receive the lens arrangement scheme issued by the multi-objective optimization module, and read the material refractive index, laser incident angle, material absorption coefficient, internal optical path length, surface roughness and aperture size of each lens in the scheme to obtain the basic data for loss calculation of each lens. The Fresnel reflection calculation unit is used to calculate the reflection loss of the two optical surfaces of each lens based on the material refractive index and the laser incident angle, and to obtain the Fresnel reflectivity of each lens. The material absorption calculation unit is used to calculate the material absorption attenuation factor based on the material absorption coefficient and the optical path length inside the lens, using the Beer-Lambert law, to obtain the material absorption loss of each lens. The surface scattering calculation unit is used to calculate the surface scattering loss rate based on the surface roughness and working wavelength of each lens, and obtain the scattering loss of each lens; The total loss synthesis unit is used to multiply the Fresnel reflectivity, material absorption attenuation factor and surface scattering loss rate of each lens, and then subtract the product from one to obtain the total energy loss rate. Based on the total energy loss rate, the energy efficiency is calculated to obtain energy efficiency verification data.

[0056] Furthermore, the energy loss calculation module in this embodiment establishes a comprehensive quantitative model, taking into account factors such as Fresnel reflection, material absorption, optical path loss, and surface scattering. The model is defined as follows: in, Let be the Fresnel reflectivity of the i-th lens (determined by the material and the angle of incidence). The absorption coefficient is... Optical path length This represents the scattering loss rate (related to surface roughness). This module calculates the energy efficiency of each scheme in real time during the optimization process, ensuring that the selected lens arrangement achieves optimal energy utilization.

[0057] like Figure 4 As shown, the energy loss calculation module is based on a comprehensive quantitative model that integrates Fresnel reflection, material absorption, optical path loss, and surface scattering. The process begins by receiving the lens arrangement scheme from the multi-objective optimization module, reading the lens parameters, calculating each loss component, solving for the total energy loss rate and energy efficiency of the scheme, simultaneously completing constraint threshold verification, feeding back energy efficiency data and invalid scheme elimination instructions to the multi-objective optimization module, and finally uploading all data to the controller for archiving, providing real-time energy efficiency support for the optimization process.

[0058] Furthermore, the closed-loop control module includes: A high-resolution charge-coupled device camera is used to acquire output spot images and obtain raw spot image data. The image processing unit is connected to the signal of the high-resolution charge-coupled device camera and is used to perform noise reduction, threshold segmentation and edge extraction preprocessing on the raw spot image data to obtain the actual size and energy distribution parameters of the spot. The collimation calculation unit is connected to the image processing unit and is used to calculate the change in the spot radius based on the actual size of the spot, and divide the change in the spot radius by the equivalent effective focal length of the system to obtain the collimation deviation. The proportional-integral-derivative adaptive control unit is connected to the collimation calculation unit. It is used to generate an error signal based on the collimation deviation and adaptively adjust the proportional coefficient, integral coefficient, and derivative coefficient to obtain the lens fine-tuning displacement command.

[0059] Furthermore, the closed-loop control module also includes: The submicron-level fine-tuning mechanism is signal-connected to the proportional-integral-derivative adaptive control unit and the controller, respectively. It is used to receive lens fine-tuning displacement commands and make submicron-level precise adjustments to the position, spacing or tilt angle of the lens to obtain the corrected lens spatial pose.

[0060] Furthermore, the closed-loop control module in this embodiment includes a high-resolution CCD camera, an image processing unit, and a collimation calculation unit, which acquires and outputs light spot images in real time and calculates the collimation deviation. when At this time, the controller drives the fine-tuning mechanism (using a PID algorithm, with adaptive adjustment of proportional, integral, and derivative parameters) to perform sub-micron level precision adjustments to the lens position, achieving real-time feedback optimization. The PID control law is: in, For the sake of quasi-institutional error, This is for fine-tuning displacement commands. The controller is an embedded industrial computer or FPGA module that coordinates the operation of various modules to achieve a fully automated closed-loop process from target parameter input, mapping modeling, multi-target optimization, energy verification to closed-loop fine-tuning. The entire system adopts a modular design and is compatible with various target spot shapes (circular, square, ring, elliptical, etc.). After inputting parameters through the software interface, the optimal solution is generated with one click.

[0061] Specifically, the closed-loop control module uses a PID adaptive algorithm to fine-tune the lens position in real time, and the fine-tuning mechanism drives the lens to achieve sub-micron level precision adjustment. When the collimation deviation Δθ exceeds the threshold, the controller outputs a fine-tuning displacement command u(t) based on the error signal e(t) to maintain the long-term stability of the output light spot. The system as a whole is arranged along the optical path as follows: laser source, lens library, mapping model module, multi-objective optimization module, energy loss calculation module, and closed-loop control module. Each module is connected to the controller to form a complete signal transmission and control closed loop.

[0062] like Figure 5 As shown, the closed-loop control module is based on high-resolution image acquisition, high-precision deviation calculation and adaptive PID control. The process begins with the CCD camera acquiring and outputting the light spot image in real time. After preprocessing by the image processing unit, the actual parameters of the light spot are calculated. The collimation calculation unit solves for the collimation deviation and the total shaping error. When the deviation exceeds the threshold, the adaptive PID algorithm is activated to generate fine-tuning instructions. The controller drives the submicron level mechanism to complete the lens correction, forming a real-time closed-loop feedback to ensure the long-term stability of the light spot shaping accuracy and collimation.

[0063] Furthermore, the system also includes: A beam sampling beam splitter is set in the optical path after the lens library to split the shaped laser beam into transmitted light and reflected light. The transmitted light is transmitted along the original optical path to the optical output window, and the reflected light is transmitted to the high-resolution charge-coupled device camera after changing direction. An optical output window is set in the transmission optical path of the beam sampling beam splitter to output the final shaped stable beam spot; The parameter input terminal is connected to the controller signal and is used to receive the target spot parameters input by the user to obtain the initial optimization conditions.

[0064] Furthermore, the lens library has a built-in submicron-level fine-tuning mechanism. The optical elements stored in the lens library include cylindrical lenses, aspherical lenses, freeform lenses, and prisms. Each lens has recorded information on its focal length, radius of curvature, material refractive index, aperture size, and surface coating.

[0065] Furthermore, the controller is an embedded industrial computer or a field-programmable gate array module, used to uniformly schedule the operation of the mapping model module, multi-objective optimization module, energy loss calculation module and closed-loop control module, realizing a fully automatic closed-loop process from target parameter input, mapping modeling, multi-objective optimization, energy verification to closed-loop fine-tuning.

[0066] Furthermore, the system sequentially arranges a laser source, a lens library, a beam sampling beam splitter, and an optical output window along the optical path; a high-resolution charge-coupled device (CCD) camera is arranged on the reflected optical path of the beam sampling beam splitter; the signal output terminal of the high-resolution CCD camera is connected to the signal input terminal of the closed-loop control module; the signal output terminal of the closed-loop control module is connected to the controller; the signal output terminal of the parameter input terminal is connected to the controller; and the drive output terminal of the controller is connected to the sub-micron fine-tuning mechanism built into the lens library.

[0067] The multi-target lens arrangement optimization system for beam spot shaping includes a laser source, a lens library, a mapping model module, a multi-target optimization module, an energy loss calculation module, a closed-loop control module, and a controller. The lens library stores the optical parameters of optical elements such as cylindrical lenses, aspherical lenses, freeform lenses, and prisms. The closed-loop control module includes a high-resolution CCD camera, an image processing unit, a collimation calculation unit, and a sub-micron fine-tuning mechanism. The laser source outputs an initial laser beam, which is shaped by the optimized lens combination in the lens library. The closed-loop control module then detects and corrects the beam in real time. The controller coordinates the operation of each module to achieve a fully automated closed-loop process from target parameter input, mapping modeling, multi-target optimization, energy verification to closed-loop fine-tuning.

[0068] This embodiment establishes a mapping model between the target light spot and the lens optical parameters by inputting the target light spot parameters. It uses a multi-objective optimization algorithm to automatically select and arrange lens combinations and their order. At the same time, it introduces a quantitative energy loss model and combines it with a closed-loop control module to achieve the optimal balance between minimizing system volume, maximizing energy efficiency, and light spot shaping accuracy. It is suitable for fields such as precision instrument optical experiments and scientific research.

[0069] The multi-target lens arrangement optimization system for spot shaping provided in this embodiment has all the advantages of the multi-target lens arrangement optimization method for spot shaping provided in Embodiment 1.

[0070] Example 3 This embodiment also discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method of Embodiment 1.

[0071] Example 4 This embodiment also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method of Embodiment 1.

[0072] Example 5 This embodiment also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method of Embodiment 1.

[0073] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations 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. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for optimizing the arrangement of multi-target lenses to shape light spots, characterized in that, include: Based on the input target spot parameters, a mapping model between the target spot and the lens optical parameters is established using ray tracing or matrix optical transmission theory to obtain the transmission relationship between the input and output spots. Based on the transmission relationship, a multi-objective optimization model is constructed with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam reshaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. Based on the optimal lens combination, arrangement order, and spatial spacing, the output light spot image is acquired in real time and the collimation deviation is calculated. When the collimation deviation exceeds the threshold, the proportional-integral-derivative adaptive algorithm is used to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

2. The method according to claim 1, characterized in that, The process of obtaining the transmission relationship includes: The mapping between input and output spot parameters is described using the overall system transmission matrix. The total transmission matrix of the system is obtained by cascading and multiplying the transmission matrices of each lens. The transmission matrix of each lens is determined by the focal length, tilt angle, and distance between the lens and its adjacent lenses. The input spot parameters include the size and divergence angle of the input spot in two vertical directions; The output spot parameters include the size and divergence angle of the output spot in two vertical directions; By employing matrix optical transmission theory, the refraction, propagation, and tilting effects of a single lens are integrated into a single transmission matrix. The total transmission matrix of the system is obtained by cascading multiple lens transmission matrices, enabling numerical prediction of the size, energy distribution, and collimation deviation of the output spot.

3. The method according to claim 1, characterized in that, The process of constructing the multi-objective optimization model includes: With minimizing the total system volume as the primary objective, the total system volume is calculated by summing the axial lengths of each lens and the spacing between each lens. With maximizing energy efficiency as the second objective, the energy efficiency is obtained by dividing the difference between the total incident energy and the total energy loss by the total incident energy. With maximizing the accuracy of light spot shaping as the third objective, the light spot shaping error is calculated by combining the deviation between the actual light intensity and the target light intensity at each sampling point of the output light spot and the collimation deviation.

4. The method according to claim 1, characterized in that, The process of obtaining the optimal lens combination, arrangement order, and spatial spacing includes: In each iteration of the non-dominated sorting genetic algorithm, the energy loss rate and energy efficiency are calculated for each lens arrangement scheme in the population, and invalid schemes that do not meet the preset energy efficiency lower limit constraint are eliminated.

5. The method according to claim 1, characterized in that, The process of acquiring and calculating the collimation deviation in real time through the closed-loop control module includes: The system acquires and outputs light spot images, and performs preprocessing such as noise reduction, threshold segmentation, and edge extraction. Calculate the actual size and energy distribution of the light spot, and calculate the collimation deviation based on the equivalent effective focal length of the system.

6. The method according to claim 1, characterized in that, The process of calculating the collimation deviation includes: The collimation deviation is obtained by dividing the change in the spot radius by the equivalent effective focal length of the system. A proportional-integral-derivative (PID) control law is used to generate lens fine-tuning displacement commands. The proportional coefficient, integral coefficient, and derivative coefficient in the PID control law are adaptively adjusted based on the collimation error signal.

7. A multi-target lens arrangement optimization system for beam spot shaping, characterized in that, include: A laser source is used to output an initial laser beam to obtain the original beam to be shaped. The lens library is used to store the optical parameters of various optical elements, and to obtain a dataset of lenses to choose from. The mapping model module is used to establish a mapping model between the target spot and the lens optical parameters based on the input target spot parameters and the lens optical parameters in the lens library, using ray tracing or matrix optical transmission theory, to obtain the transmission relationship between the input and output spots. The multi-objective optimization module is used to construct a multi-objective optimization model based on the transmission relationship, with the objectives of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam pattern shaping accuracy. A quantitative energy loss model is introduced, and a non-dominated sorting genetic algorithm is used to solve the problem to obtain the optimal lens combination, arrangement order, and spatial spacing. The energy loss calculation module is used to calculate the energy loss rate and energy efficiency of each lens arrangement scheme in real time during the solution process of the multi-objective optimization module, and to obtain energy efficiency verification data and feed it back to the multi-objective optimization module. The closed-loop control module is used to acquire and output spot images in real time and calculate the collimation deviation to obtain the collimation error signal. The controller is bidirectionally connected to the mapping model module, the multi-objective optimization module, the energy loss calculation module, and the closed-loop control module, respectively. It is used to determine whether the deviation exceeds the threshold based on the collimation error signal, and when the deviation exceeds the threshold, it uses a proportional-integral-derivative adaptive algorithm to drive the fine-tuning mechanism to perform position correction and obtain a stable shaped light spot.

8. The system according to claim 7, characterized in that, The mapping model module includes: The transfer matrix construction unit is used to integrate the refraction effect, propagation effect and tilting effect of a single lens into a single transfer matrix, thereby obtaining the independent transfer matrix of each lens. The cascaded operation unit is used to cascade and multiply the independent transmission matrices of each lens according to the lens arrangement order to obtain the total transmission matrix of the system. The parameter prediction unit is used to numerically predict the size, energy distribution, and collimation deviation of the output spot based on the total transmission matrix of the system and the input spot parameters, thereby obtaining the transmission relationship between the input and output spots.

9. The system according to claim 7, characterized in that, The multi-objective optimization module includes: The objective function construction unit is used to construct three core objective functions with the goal of minimizing the total system volume, maximizing energy efficiency, and maximizing the beam spot shaping accuracy, thereby obtaining the optimization objective set. The algorithm initialization unit is used to encode the lens type, quantity, arrangement order, spacing and tilt angle, randomly generate an initial scheme population, and obtain a set of lens arrangement schemes to be evaluated. The fitness calculation unit is used to send each lens arrangement scheme in the population to the energy loss calculation module, obtain the energy loss rate and energy efficiency of each scheme, and calculate the fitness value of each individual to obtain the non-dominated ranking and crowding results. The iterative evolutionary unit is used to perform selection, crossover, and mutation operations to generate offspring populations. After merging the parent and offspring populations, the non-dominated sorting is performed again and elite individuals are selected to obtain the Pareto optimal solution set. The scheme selection unit is used to select the lens arrangement scheme with the best overall performance from the Pareto optimal solution set according to the preset weight coefficients, so as to obtain the optimal lens combination, arrangement order and spatial spacing.

10. The system according to claim 7, characterized in that, The energy loss calculation module includes: The parameter reading unit is used to receive the lens arrangement scheme issued by the multi-objective optimization module, and read the material refractive index, laser incident angle, material absorption coefficient, optical path length, surface roughness and aperture size of each lens in the scheme to obtain the basic data for loss calculation of each lens. The Fresnel reflection calculation unit is used to calculate the reflection loss of the two optical surfaces of each lens based on the material refractive index and the laser incident angle, and to obtain the Fresnel reflectivity of each lens. The material absorption calculation unit is used to calculate the material absorption attenuation factor based on the material absorption coefficient and the optical path length inside the lens, using the Beer-Lambert law, to obtain the material absorption loss of each lens. The surface scattering calculation unit is used to calculate the surface scattering loss rate based on the surface roughness and working wavelength of each lens, and obtain the scattering loss of each lens; The total loss synthesis unit is used to multiply the Fresnel reflectivity, material absorption attenuation factor and surface scattering loss rate of each lens, and then subtract the product from one to obtain the total energy loss rate. Based on the total energy loss rate, the energy efficiency is calculated to obtain energy efficiency verification data.

Images