Beam quality test setup for far-field lasers

Abstract

The present invention discloses a beam quality testing device for far-field lasers, which belongs to the field of beam quality testing and includes: a user end, which is used to receive beam quality test requests; a data center, which includes: a dynamic weighting component, which is used to generate beam quality evaluation factor weight distribution results; a function configuration component, which is used to configure a beam quality evaluation function; a virtual test component, which is used to obtain theoretical qualified control parameters; an actual test component, which is used to obtain actual test qualified control parameters; and a data transmission component, which sends the actual test qualified control parameters to the user end when the number of actual test qualified control parameters is not equal to 0. The present invention solves the technical problems of low efficiency and high actual test cost in the prior art of beam quality testing. By introducing a virtual test link and combining a dynamically weighted beam quality evaluation function, the present invention achieves the technical effect of improving test efficiency and reducing actual test costs.

Description

Technical Field

[0001] The invention relates to the field of beam quality testing, and in particular to a beam quality testing device for far-field lasers. Background Art

[0002] In order to ensure the normal operation of far-field laser equipment and achieve the expected functions, the quality of the laser beam must be strictly tested and controlled. In the prior art, the beam quality test usually adopts the actual test method, that is, by adjusting the laser emission control parameters, emitting the laser multiple times and evaluating the beam quality until qualified control parameters are found. This test method has the following shortcomings: actual testing often requires a large number of attempts and iterations, resulting in a time-consuming and inefficient test process; frequent actual testing not only consumes a lot of energy and materials, but may also cause a certain degree of loss and wear on the test equipment, and the actual test cost is high. Summary of the invention

[0003] The present application aims to solve the technical problems of low efficiency and high actual testing cost of beam quality testing in the prior art by providing a beam quality testing device for far-field lasers.

[0004] The present application discloses a beam quality testing device for far-field lasers, comprising: a user end, for receiving a beam quality test request, wherein the beam quality test request includes a far-field laser application scenario type; a data center, comprising: a dynamic weighting component, for receiving a far-field laser application scenario type from a user end, weighting a beam quality evaluation factor, and generating a beam quality evaluation factor weight distribution result; a function configuration component, for receiving a beam quality evaluation factor weight distribution result from a dynamic weighting component, and configuring a beam quality evaluation function; a virtual test component, for activating a beam quality evaluation factor prediction network, and performing a virtual test on a far-field laser emission control parameter in combination with a beam quality evaluation function to obtain a theoretical qualified control parameter; an actual test component, for receiving a theoretical qualified control parameter from a virtual test component, and performing an actual test in combination with a beam quality evaluation function to obtain an actual test qualified control parameter; and a data transmission component, for sending the actual test qualified control parameter to the user end when the number of actual test qualified control parameters is not equal to 0.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The user end receives the beam quality test request, including the far-field laser application scenario type, and clarifies the test requirements and target scenarios; the dynamic weighting component of the data center dynamically weights the beam quality evaluation factor according to the far-field laser application scenario type, generates the beam quality evaluation factor weight distribution result, and establishes an evaluation index system suitable for the application scenario; the function configuration component of the data center configures the beam quality evaluation function according to the evaluation factor weight distribution result to form a complete beam quality evaluation model; the virtual test component of the data center combines the beam quality evaluation factor prediction network and the beam quality evaluation function to test and screen the control parameters in a virtual environment, obtain theoretically qualified control parameters, and pass the virtual test. The actual test component of the data center combines the evaluation function to conduct actual test verification on the theoretical qualified control parameters, and obtains the actual test qualified control parameters, so as to carry out actual test under the guidance of theory and further screen the optimal parameters; if the number of the actual test qualified control parameters is not zero, it is output to the user end through the data transmission component, so as to provide the laser control parameters that meet the requirements to the user, and guide the technical solution for practical application, which solves the technical problems of low efficiency of beam quality test and high actual test cost in the prior art. By introducing the virtual test link and combining the dynamically weighted beam quality evaluation function, the present invention achieves the technical effect of improving test efficiency and reducing actual test cost.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A structural schematic diagram of a beam quality testing device for far-field lasers is provided for an embodiment of the present application;

[0009] Figure 2 A flow chart of generating a beam quality evaluation factor prediction network in a beam quality testing device for a far-field laser is provided for an embodiment of the present application.

[0010] Explanation of the reference numerals: user end 11 , data center 12 , dynamic weighting component 121 , function configuration component 122 , virtual test component 123 , actual test component 124 , data transmission component 125 . DETAILED DESCRIPTION

[0011] The overall idea of ​​the technical solution provided by this application is as follows:

[0012] The embodiment of the present application provides a beam quality testing device for far-field lasers. Taking the far-field laser application scenario type as the entry point, through the collaborative work of the user end and the data center, a beam quality evaluation system is adaptively established, and virtual testing is introduced to optimize and guide the actual testing, and finally the laser emission control parameters that meet the specific application requirements are output.

[0013] Specifically, first, the user side receives a beam quality test request, including the far-field laser application scenario type. The dynamic weighting component of the data center dynamically weights the beam quality evaluation factors according to the far-field laser application scenario type to form the evaluation factor weight distribution result. The function configuration component then configures the corresponding beam quality evaluation function based on the weight distribution result to build a complete beam quality evaluation model. Subsequently, the virtual test component combines the prediction network and the evaluation function to test and screen the control parameters in a virtual environment to obtain a set of theoretical qualified control parameters. The actual test component further combines the evaluation function to perform actual test verification on this set of theoretical qualified parameters, and finally obtains the control parameters that are qualified in the actual test. If the number of actual qualified parameters is not zero, it is output to the user side through the data transmission component for actual application.

[0014] Through the above scheme, the present invention constructs a full-process closed-loop optimization mechanism from test requirement input to qualified parameter output. The user terminal initiates a test request and provides application scenario information, which triggers the data center to carry out adaptive test optimization. The data center dynamically configures the beam quality evaluation system and efficiently screens out the optimal control parameters through the collaboration of virtual and actual tests. The screening results are fed back to the user terminal in a timely manner, forming a test closed loop, effectively improving the efficiency of beam quality testing and reducing testing costs.

[0015] After introducing the basic principles of the present application, the non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the drawings in the specification.

[0016] like Figure 1 As shown, an embodiment of the present application provides a beam quality testing device for far-field lasers, the device comprising:

[0017] The user terminal 11 is used to receive a beam quality test request, wherein the beam quality test request includes a far-field laser application scenario type.

[0018] Specifically, the user submits a beam quality test request through the user terminal 11, and the request includes the application scenario type information of the far-field laser. The far-field laser application scenario type refers to the specific application field or environment of the far-field laser technology, such as laser cutting, laser welding, laser ranging, laser communication, etc. Different application scenarios have different requirements for the quality indicators of the laser beam.

[0019] After receiving the request containing the application scenario type, the user terminal 11 sends the request information to the data center 12 for subsequent beam quality evaluation processing. By receiving the request and transmitting the scenario information by the user terminal 11, the beam quality test can be combined with the actual application scenario to meet the specific needs of different fields and improve the flexibility of the test.

[0020] The data center 12 is the core functional module for implementing beam quality testing. The data center 12 receives the beam quality test request from the user end 11, and performs a series of data processing and analysis operations based on the far-field laser application scenario type information contained in the request, and finally completes the evaluation and testing of the beam quality. Specifically, the data center 12 includes a dynamic weighting component 121, a function configuration component 122, a virtual test component 123, an actual test component 124, and a data transmission component 125. These components work together to complete the various tasks of the beam quality test.

[0021] By dividing different functions into dedicated components, Data Center 12 achieves modular design, improving maintainability and scalability. Each component communicates and exchanges data through clear interfaces to ensure smooth execution of the beam quality test process.

[0022] The dynamic weighting component 121 is used to receive the far-field laser application scenario type from the user terminal, weight the beam quality evaluation factor, and generate a beam quality evaluation factor weight distribution result.

[0023] Specifically, the dynamic weighting component 121 is one of the components of the data center 12, which dynamically weights the beam quality evaluation factors according to the application scenario type of the far-field laser and generates a weight distribution result.

[0024] First, the dynamic weighting component 121 obtains the far-field laser application scenario type information from the beam quality test request received by the user terminal 11. Different application scenarios have different degrees of importance for various indicators of beam quality (i.e., beam quality evaluation factors). For example, in the laser welding scenario, the energy distribution and focusing performance of the beam are more critical; while in the laser communication scenario, the divergence angle and spot quality of the beam are more concerned. Therefore, the dynamic weighting component 121 assigns weights to each beam quality evaluation factor according to the far-field laser application scenario type, and obtains the weight coefficient of each factor by calculating the correlation between different beam quality evaluation factors and the application scenario, thereby generating a beam quality evaluation factor weight distribution result, which reflects the importance of each beam quality evaluation factor in a specific scenario, making the beam quality assessment more flexible and adaptable, ensuring the matching degree between the assessment results and actual needs, and improving the effectiveness of the test.

[0025] The function configuration component 122 is used to receive the beam quality evaluation factor weight distribution result from the dynamic weighting component 121 and configure the beam quality evaluation function.

[0026] Specifically, the function configuration component 122 receives the beam quality evaluation factor weight distribution result from the dynamic weighting component 121, and generates an evaluation function suitable for the current application scenario based on the result.

[0027] First, the function configuration component 122 obtains the beam quality evaluation factor weight distribution result from the dynamic empowerment component 121 through the interface. The result includes the weight coefficient of each beam quality evaluation factor, reflecting the importance of each beam quality evaluation factor in the current application scenario. The function configuration component 122 uses the weight coefficient of each beam quality evaluation factor as a parameter, substitutes it into the general beam quality evaluation function template, generates a targeted evaluation function, and completes the configuration of the beam quality evaluation function. The beam quality evaluation function comprehensively considers the differences between the measured values ​​and ideal values ​​of multiple beam quality evaluation factors, as well as their respective weights, calculates a quantitative beam quality score, intuitively reflects the quality of the beam, facilitates comparison and screening, and provides a reliable basis for subsequent virtual tests and actual tests.

[0028] The virtual test component 123 is used to activate the beam quality evaluation factor prediction network, and in combination with the beam quality evaluation function, to perform a virtual test on the far-field laser emission control parameters to obtain theoretical qualified control parameters.

[0029] Specifically, the virtual test component 123 is used to pre-screen potential theoretical qualified control parameters through theoretical calculation and simulation before actual testing, so as to provide guidance for subsequent actual testing.

[0030] First, the virtual test component 123 activates the pre-trained beam quality evaluation factor prediction network, which is a machine learning model. Through the training of a large amount of historical data, a mapping relationship between far-field laser emission control parameters and beam quality evaluation factors is established. In the virtual test phase, the virtual test component 123 inputs a set of control parameters to be tested into the beam quality evaluation factor prediction network, and obtains the corresponding beam quality evaluation factor prediction value after forward propagation calculation. Then, these predicted values ​​are substituted into the beam quality evaluation function generated by the function configuration component 122 to calculate the beam quality score. Through a large number of parameter combinations and iterative calculations, the virtual test component 123 finds a series of control parameters that can make the score meet the preset qualified threshold. These parameters are theoretical qualified control parameters, which means that at the level of theoretical calculation and simulation prediction, there is a high probability that a high-quality laser beam can be generated.

[0031] Through virtual testing, using machine learning prediction and numerical simulation methods, we achieved pre-screening of far-field laser emission control parameters, effectively improved test efficiency, reduced test costs, and provided reliable reference and guidance for subsequent actual tests.

[0032] The actual test component 124 is used to receive the theoretical qualified control parameters from the virtual test component 123, perform actual testing in combination with the beam quality evaluation function, and obtain actual test qualified control parameters.

[0033] Specifically, the actual test component 124 receives the theoretical qualified control parameters from the virtual test component 123 and performs tests on the actual laser device to determine the effectiveness of these parameters in practice.

[0034] First, the actual test component 124 obtains one or more sets of theoretical qualified control parameters from the virtual test component 123 through the interface. These parameters are obtained through theoretical calculation and simulation prediction in the virtual test stage, and meet the beam quality score requirements at the numerical level, but their actual effect still needs to be verified by actual measurement. The actual test component 124 applies the theoretical qualified control parameters to the laser device to control the emission process of the laser. During the emission process, the measured data of the beam, such as the spot radius, divergence angle, energy distribution, etc., are collected using a beam quality measurement device (such as a wavefront sensor, an energy meter, etc.). Then, the actual test component 124 inputs the collected measured data into the beam quality evaluation function provided by the function configuration component 122 to calculate the comprehensive beam quality score. The scoring result is compared with the preset qualified threshold to determine whether the current control parameters can meet the quality requirements in actual applications. After a series of actual tests and data analysis, the actual test component 124 selects truly feasible and effective actual test qualified control parameters from the theoretical qualified control parameters. These parameters are not only feasible in theory, but can also stably produce high-quality laser beams in actual tests, and can serve as the optimal reference for subsequent applications.

[0035] Through actual testing, actual verification was carried out on the basis of theoretical predictions, and reliable actual test qualified control parameters were obtained, which effectively improved the credibility and application value of the test results and provided important reference for subsequent laser applications.

[0036] The data transmission component 125 sends the actual test qualified control parameters to the user end when the number of the actual test qualified control parameters is not equal to 0.

[0037] Specifically, the data transmission component 125 is responsible for transmitting the test results between the data center 12 and the user end 11, and serves as a bridge connecting the internal processing module and the external user interface, ensuring that the valid parameters obtained from the test can be fed back to the user in a timely and accurate manner.

[0038] First, the data transmission component 125 obtains a set of actual test qualified control parameters by monitoring the output of the actual test component 124. When the set is not empty, that is, some control parameters that meet the quality requirements are indeed screened out during the actual test phase, the data transmission component 125 will start the data transmission process. During the transmission process, the data transmission component 125 packages the actual test qualified control parameters into a data packet in a standard format through the data interface and communication protocol pre-agreed with the user terminal 11. Then, using the network communication module, the data packet is sent to the user terminal 11 through a wired or wireless data transmission channel. After receiving the data packet, the user terminal 11 extracts the actual test qualified control parameters contained therein by parsing the data format. The user sets and adjusts these preferred actual test qualified control parameters in actual laser applications in order to obtain the best beam quality.

[0039] Through the data transmission component 125, the test results are efficiently transmitted between the data center and the user terminal, ensuring the closed loop of the test process, so that the optimized control parameters can be truly applied to actual laser usage scenarios.

[0040] Furthermore, the steps performed by the dynamic weighting component 121 include:

[0041] The far-field laser application scenario type is identified through the operation effect evaluation index calibration table, and an operation effect evaluation index matching result is generated, wherein the operation effect evaluation index calibration table is an evaluation index pre-set for different application scenario types;

[0042] Traversing the matching results of the operation effect evaluation indicators, performing correlation analysis on the far-field laser application scenario types, and obtaining several groups of correlation analysis results;

[0043] The plurality of groups of correlation analysis results are fused to generate the beam quality evaluation factor weight distribution result.

[0044] In a preferred embodiment, first, the dynamic weighting component 121 identifies and calibrates the current far-field laser application scenario type by accessing a pre-set calibration table of operating effect evaluation indicators. The calibration table is a set of evaluation index systems that are pre-set according to the characteristics and requirements of different application scenarios, and specifies which operating effect indicators should be focused on in specific scenarios. For example, in the laser welding scenario, the calibration table lists "processing accuracy" and "energy concentration" as important evaluation indicators; in the laser communication scenario, the calibration table emphasizes indicators such as "signal quality", "transmission efficiency" and "receiving intensity". By looking up the current far-field laser application scenario type in the calibration table, the dynamic weighting component 121 can quickly determine the operating effect indicators that are focused on and generate operating effect evaluation indicator matching results.

[0045] Next, the dynamic weighting component 121 traverses and analyzes the obtained operation effect evaluation index matching results one by one. For each matched evaluation index, the correlation between the index and the beam quality evaluation factor is examined using the correlation analysis method. Specifically, the correlation coefficient between each beam quality evaluation factor (such as spot radius, divergence angle, M2 factor, etc.) and the current operation effect index is calculated to obtain a quantitative correlation analysis result. This process is repeated until all matched operation effect indicators are analyzed and several groups of correlation analysis results are accumulated. After that, the dynamic weighting component 121 performs a comprehensive processing on all the correlation analysis results to generate the final beam quality evaluation factor weight distribution. First, all the correlation coefficients of each evaluation factor are summed to obtain the total correlation of the factor. Then, it divides the total correlation of each factor by the sum of the total correlations of all factors to obtain the weight coefficient of the factor. In this way, the sum of the weight coefficients of all evaluation factors is exactly 1, forming the beam quality evaluation factor weight distribution result, which reflects the importance of each beam quality evaluation factor in the current application scenario and provides a basis for the subsequent evaluation function construction.

[0046] Through the above steps, the dynamic weighting component 121 realizes the adaptive allocation of beam quality evaluation factor weights, comprehensively considers the characteristics and requirements of the application scenario, and the correlation between each evaluation factor and the operation effect, and obtains the beam quality evaluation factor weight distribution result. The dynamic weighting method overcomes the limitations of traditional fixed weight allocation, so that subsequent beam quality evaluation can better adapt to different scenarios and improve the reliability and applicability of test results.

[0047] Furthermore, the embodiment of the present application also includes:

[0048] Extracting a first operation effect evaluation indicator according to the operation effect evaluation indicator matching result;

[0049] Taking the beam quality evaluation factor as an independent variable and the first operation effect evaluation index as a dependent variable, collecting multiple sets of beam quality evaluation factor record data and multiple first operation effect evaluation index record data;

[0050] Performing normalization processing on the plurality of first operation effect evaluation index record data to construct an operation effect evaluation index characteristic value sequence;

[0051] Performing same-attribute normalization processing on the multiple groups of beam quality evaluation factor record data to construct multiple beam quality evaluation factor eigenvalue sequences;

[0052] Grey correlation processing is performed on the characteristic value sequence of the operation effect evaluation index and the characteristic value sequence of the multiple beam quality evaluation factors to obtain a first group of correlation analysis results, which are added to the plurality of groups of correlation analysis results.

[0053] In a preferred embodiment, first, the dynamic weighting component 121 extracts a specific evaluation index from the obtained operation effect evaluation index matching results as the target of this analysis, which is called the first operation effect evaluation index. Then, the dynamic weighting component 121 starts data collection and preparation. The beam quality evaluation factor is used as the independent variable, and the first operation effect evaluation index is used as the dependent variable, and a certain number of data records are collected from the historical database. These records include the values ​​of each beam quality evaluation factor under different conditions, as well as the performance values ​​of the corresponding first operation effect evaluation index, and multiple groups of beam quality evaluation factor record data and multiple first operation effect evaluation index record data are obtained, and the original data samples required for correlation analysis are obtained. Next, in order to eliminate the differences in the dimensions and value ranges of different data indicators, the dynamic weighting component 121 normalizes the collected data. For the dependent variable data, that is, the multiple record values ​​of the first operation effect evaluation index, a unified scale and benchmark are used to scale and translate them to convert them into a sequence of characteristic values ​​of the operation effect evaluation index in the interval [0, 1]. Similarly, for the independent variable data, that is, multiple sets of recorded values ​​of the beam quality evaluation factors, the data subsets of each evaluation factor are normalized respectively to obtain multiple beam quality evaluation factor eigenvalue sequences that match the dependent variable sequence, thus ensuring the comparability between different indicator data and laying the foundation for the subsequent correlation calculation.

[0054] After the normalization process is completed, the dynamic weighting component 121 uses the grey correlation analysis method to examine the correlation between the independent variable eigenvalue sequence and the dependent variable eigenvalue sequence, and reflects the correlation strength between the beam quality evaluation factor and the operation effect evaluation index by calculating the degree of proximity of the sequences in geometric shape. The range of the correlation is [0, 1]. The larger the value, the more consistent the change trends of the two sequences are, and the stronger the correlation is. By calculating the grey correlation between each beam quality evaluation factor eigenvalue sequence and the first operation effect evaluation index eigenvalue sequence, the dynamic weighting component 121 obtains a set of quantitative correlation analysis results, namely the first set of correlation analysis results, which reflects the correlation between each beam quality evaluation factor and the first operation effect evaluation index. Afterwards, the dynamic weighting component 121 adds the obtained first correlation analysis results to several groups of correlation analysis results as part of several groups of correlation analysis results.

[0055] By looping through each operation effect evaluation indicator in the matching results and analyzing the correlation of each operation effect evaluation indicator, we finally get several groups of correlation analysis results, which comprehensively characterize the intrinsic connection between the beam quality evaluation factor and the operation effect in the current application scenario, and provide data support for the subsequent weight calculation.

[0056] Furthermore, the function configuration component 122 executes the following steps:

[0057] Construct a beam quality evaluation function:

[0058]

[0059] in, Characterize the beam quality evaluation value, Characterizes the measurement value of the i-th evaluation factor, Characterizes the ideal threshold of the i-th evaluation factor, The exponential adjustment coefficient representing the i-th factor is used to control the rate of exponential growth. Represents the weight of the i-th evaluation factor;

[0060] The beam quality evaluation function is configured according to the beam quality evaluation factor weight distribution result.

[0061] In a preferred embodiment, the function configuration component 122 adopts a weighted exponential function form that comprehensively considers multiple beam quality evaluation factors when constructing the beam quality evaluation function, specifically:

[0062] in, It represents the beam quality evaluation value, a real number in the interval (0, 1]. The larger the value, the higher the beam quality and the better the overall performance. Represents the total number of beam quality evaluation factors considered. represents the actual measured value of the i-th evaluation factor, Represents the ideal target value or threshold of the i-th evaluation factor. It reflects the degree of deviation between the actual performance of the i-th evaluation factor and the expected performance. is the exponential adjustment coefficient of the i-th evaluation factor, which is used to control the rate at which the deviation affects the evaluation value. When smaller, Close to 1, the evaluation value is less affected by this factor; when When larger, Rapidly increases, and the evaluation value is greatly affected by this factor. The value of can control the sensitivity of different factors to the comprehensive quality.

[0063] It is a logarithmic function that grows slowly as the degree of deviation increases, complementing the exponential function. When the degree of deviation is large, it weakens the excessive growth of the exponential function, making the change of the evaluation value smoother and more reasonable. The function configuration component 122 receives the beam quality evaluation factor weight distribution result from the dynamic weighting component 121, extracts the weight value of each factor, and assigns it to the corresponding , thus completing the configuration of the beam quality evaluation function.

[0064] In the beam quality evaluation function, The weighted items of the evaluation factors appear in the denominator in a cumulative form. When the actual measured values ​​of all factors are close to the ideal threshold, each item approaches , the cumulative result is close to the sum of the weights 1, at this time When the value is close to 1, it means that the comprehensive beam quality is optimal. When some factors deviate greatly, the corresponding exponential function term will increase rapidly, and the denominator of the sum will increase sharply, resulting in The value decreases, indicating that the overall quality decreases. In this way, the beam quality evaluation function realizes the comprehensive evaluation of beam quality, while highlighting the influence of key factors, providing a clear evaluation standard for subsequent virtual tests and actual tests. At the same time, since the parameters of the evaluation function are dynamically generated according to the application scenario, it can adaptively reflect the quality requirements in different scenarios, ensuring the pertinence and effectiveness of the evaluation results.

[0065] Furthermore, the virtual test component 123 executes the steps including:

[0066] The beam quality evaluation factors include far-field spot radius, far-field divergence angle, diffraction limit multiplication factor, Strehl ratio, M2 factor, and surrounding energy ratio;

[0067] The far-field laser emission control parameters include at least wavelength, power, pulse duration, repetition rate, beam diameter, and working distance;

[0068] Taking the wavelength, the power, the pulse duration, the repetition rate, the beam diameter, and the working distance as input variables, traversing the far-field spot radius, the far-field divergence angle, the diffraction limit multiplication factor, the Strehl ratio, the M2 factor, and the surrounding energy, a beam quality evaluation factor prediction network is constructed;

[0069] Performing a virtual test on the far-field laser emission control parameters according to the beam quality evaluation factor prediction network to obtain a beam quality evaluation factor prediction value set;

[0070] Processing the beam quality evaluation factor prediction value set according to the beam quality evaluation function to generate a beam quality evaluation value;

[0071] When the beam quality evaluation value meets the beam quality evaluation threshold, the far-field laser emission control parameter is added to the theoretical qualified control parameter.

[0072] In a feasible implementation, first, the virtual test component 123 determines a set of key beam quality evaluation factors, including far-field spot radius, far-field divergence angle, diffraction limit multiple factor, Strehl ratio, M2 factor and surrounding energy ratio. These factors reflect the key characteristics of the laser beam from different aspects and are important indicators for judging beam quality. Secondly, the virtual test component 123 identifies the main types of far-field laser emission control parameters, including at least wavelength, power, pulse duration, repetition rate, beam diameter and working distance. These parameters directly determine the emission characteristics of the far-field laser and are key factors affecting beam quality.

[0073] After determining the beam quality evaluation factor and the far-field laser emission control parameters, the virtual test component 123 begins to build a beam quality evaluation factor prediction network. The network uses a supervised learning algorithm in machine learning, with wavelength, power, pulse duration, repetition rate, beam diameter and working distance as input variables, and far-field spot radius, far-field divergence angle, diffraction limit multiple factor, Strehl ratio, M2 factor and ring energy ratio as output targets. Through training with a large amount of historical data, a nonlinear mapping relationship between control parameters and evaluation factors is established. The structure of the prediction network can adopt common models such as multi-layer perceptron and convolutional neural network, and continuously improve the accuracy and generalization ability of prediction through repeated iterative optimization. After the beam quality evaluation factor prediction network is established, the virtual test component 123 uses the network to quickly evaluate the far-field laser emission control parameter combination to be tested. The specific value of the control parameter is input into the prediction network, and the corresponding beam quality evaluation factor prediction value is obtained through forward propagation calculation to form a beam quality evaluation factor prediction value set. This process can complete the evaluation of a large number of parameter combinations in a very short time, greatly accelerating the test progress.

[0074] Next, the virtual test component 123 substitutes the predicted beam quality evaluation factor prediction value set into the beam quality evaluation function constructed by the function configuration component 122, and calculates a comprehensive beam quality evaluation value, which reflects the overall quality level of the laser beam under the current control parameter settings. Afterwards, the virtual test component 123 compares the calculated beam quality evaluation value with the preset beam quality evaluation threshold. If the evaluation value meets or exceeds the threshold requirement, it is considered that the current control parameter combination is theoretically feasible, and it is added to the theoretical qualified control parameter set as an alternative for subsequent actual testing. If the evaluation value does not reach the threshold, it indicates that the current parameter combination is not ideal and needs to be further optimized and adjusted.

[0075] By quickly evaluating and screening a large number of laser emission control parameter combinations, using machine learning prediction models to estimate the feasibility of parameters at the theoretical level, and combining them with the beam quality evaluation function, objective screening results can be obtained. A large number of obviously unqualified parameters can be eliminated before actual testing, narrowing the scope of actual testing, thereby significantly improving test efficiency and reducing test costs.

[0076] Further, such as Figure 2 As shown, a beam quality evaluation factor prediction network is constructed, including:

[0077] According to the wavelength, the power, the pulse duration, the repetition rate, the beam diameter, and the working distance as independent variables, and the far-field spot radius as a dependent variable, collecting an independent variable array set and a dependent variable data set;

[0078] According to the independent variable array set and the dependent variable data set, configuring a far-field spot radius prediction subnetwork;

[0079] Traversing the far-field divergence angle, the diffraction limit multiplication factor, the Strehl ratio, the M2 factor, and the surrounding energy, constructing a far-field divergence angle prediction subnetwork, a diffraction limit multiplication factor prediction subnetwork, a Strehl ratio prediction subnetwork, an M2 factor prediction subnetwork, and a surrounding energy prediction subnetwork;

[0080] The input layers of the far-field spot radius prediction subnetwork, the far-field divergence angle prediction subnetwork, the diffraction limit multiple factor prediction subnetwork, the Strehl ratio prediction subnetwork, the M2 factor prediction subnetwork and the surrounding energy prediction subnetwork are merged to generate the beam quality evaluation factor prediction network.

[0081] In a preferred embodiment, first, the virtual test component 123 uses wavelength, power, pulse duration, repetition rate, beam diameter and working distance as independent variables, and far-field spot radius as dependent variable, and collects paired independent variable array sets and dependent variable data sets from the historical database. Among them, each independent variable array contains a complete control parameter combination of a laser emission, and the corresponding dependent variable data is the far-field spot radius value measured under the parameter combination, and the training data required for constructing the far-field spot radius prediction subnetwork is obtained. Next, the virtual test component 123 uses the collected independent variable array set and dependent variable data set to configure and train a special far-field spot radius prediction subnetwork. The subnetwork adopts common regression models such as feedforward neural networks and support vector machines, and establishes a nonlinear mapping relationship from control parameters to far-field spot radius by setting network structure and hyperparameters, as well as sufficient training iterations. After the training is completed, the subnetwork can predict the corresponding theoretical value of the far-field spot radius according to any given control parameter combination.

[0082] After constructing the far-field spot radius prediction subnetwork, the virtual test component 123 uses the same method to traverse several other key beam quality evaluation factors, namely the far-field divergence angle, the diffraction limit multiple factor, the Strehl ratio, the M2 factor and the surrounding energy ratio, and respectively establishes prediction subnetworks between each beam quality evaluation factor and the control parameters. The inputs of these subnetworks are all the same control parameter combination, and the outputs are the corresponding evaluation factor prediction values. Through a one-to-many modeling approach, the virtual test component 123 trains a dedicated prediction subnetwork for each evaluation factor, including a far-field divergence angle prediction subnetwork, a diffraction limit multiple factor prediction subnetwork, a Strehl ratio prediction subnetwork, an M2 factor prediction subnetwork and a surrounding energy prediction subnetwork, thereby achieving independent estimation of different beam quality indicators.

[0083] Afterwards, in order to form a complete beam quality evaluation factor prediction network, the virtual test component 123 merges the input layers of all sub-networks constructed previously to form a unified input interface. This interface receives a complete combination of control parameters, distributes the parameters to each sub-network, and triggers the prediction process at the same time. After each sub-network completes the calculation independently, it outputs the prediction results to its respective output nodes. In this way, the entire prediction network forms a mapping relationship, which can predict the values ​​of all beam quality evaluation factors at one time given the control parameters.

[0084] Through modular construction, virtual test component 123 realizes a highly flexible and scalable beam quality evaluation factor prediction network. It adopts the idea of ​​sub-network integration to decompose the complex multi-factor prediction problem into multiple independent single-factor prediction tasks, which not only reduces the difficulty of each sub-task, but also can fully explore the intrinsic connection between different factors.

[0085] Furthermore, the data transmission component 125 performs the following steps:

[0086] When the number of the control parameters that are actually tested and qualified is equal to 0, a far-field laser emission control parameter adjustment instruction is generated;

[0087] Sending the far-field laser emission control parameter adjustment instruction to the virtual test component 123, adjusting the theoretical qualified control parameter, and obtaining the theoretical qualified updated control parameter;

[0088] The control parameters are updated according to the theoretical qualifications and an iterative cycle is performed.

[0089] In a feasible implementation, the data transmission component 125 starts the process of parameter adjustment and iterative optimization when the actual test results are not ideal. First, the data transmission component 125 determines whether the number of qualified control parameters in the actual test is equal to 0. If it is equal to 0, it means that in the current set of theoretical qualified control parameters, no set of parameters has passed the verification of the actual test, that is, all parameter combinations have failed to meet the expected beam quality requirements in actual applications. In this case, the data transmission component 125 automatically generates a far-field laser emission control parameter adjustment instruction, indicating that the control parameters need to be further optimized and adjusted.

[0090] Next, the data transmission component 125 sends the generated control parameter adjustment instruction to the virtual test component 123. After receiving the instruction, the latter analyzes and corrects the current set of theoretical qualified control parameters. Specifically, the virtual test component 123 fine-tunes certain key parameters or weights in the theoretical prediction model according to the feedback results of the actual test, in order to better match the characteristics of the actual laser system. At the same time, the search range of the control parameters is appropriately expanded to explore more possible parameter combinations. Through these adjustment and optimization measures, the virtual test component 123 can generate updated theoretical qualified control parameters, called theoretical qualified updated control parameters, which absorb the feedback information of the actual test and make targeted improvements on the original basis, which can bring better actual test results.

[0091] Afterwards, the theoretical qualified updated control parameters are sent again to the actual test component 124, triggering a new round of actual test process. The actual test component 124 performs parameter settings and beam quality measurements on the actual laser device according to the updated parameter set, and returns the test results to the data transmission component 125. If the updated parameter set can produce actual test results that meet the requirements, the optimization process ends, and the data transmission component 125 sends the final qualified parameters to the user end; if the actual test results are still not ideal, the data transmission component 125 will generate parameter adjustment instructions again and send them to the virtual test component 123 to start a new round of parameter adjustment and iterative optimization. Repeat until a control parameter combination that meets the actual requirements is found, or the preset upper limit of the number of iterations is reached.

[0092] Through parameter adjustment and iterative optimization mechanisms, the data transmission component 125 implements a closed-loop adaptive control strategy. Based on the feedback from actual tests, it automatically calls the virtual test component 123 to dynamically adjust and optimize the control parameters, continuously narrowing the gap between theoretical predictions and actual results, and ultimately finding the optimal control parameter combination.

[0093] Furthermore, the data transmission component 125 executes the steps further including:

[0094] When the number of the actual test qualified control parameters is not equal to 0, and is greater than or equal to 2, the actual test qualified control parameters are sorted from large to small according to the beam quality evaluation value to obtain a list of actual test qualified control parameters;

[0095] The actual test qualified control parameter list is sent to the user terminal.

[0096] In a feasible implementation, the data transmission component 125 performs result sorting and recommendation when the actual test results are ideal. First, the data transmission component 125 determines whether the number of qualified control parameters in the actual test is not equal to 0 and is greater than or equal to 2. If this condition is met, it means that in the current set of theoretical qualified control parameters, at least two sets of parameters have passed the verification of the actual test, that is, these parameter combinations have met the expected beam quality requirements in practical applications. In this case, although all qualified parameter combinations can meet basic usage requirements, different parameter combinations may still have differences in performance and advantages and disadvantages. In order to provide users with better parameter recommendations, the data transmission component 125 further screens and sorts the qualified control parameters in the actual test.

[0097] Specifically, the data transmission component 125 extracts the beam quality evaluation value corresponding to each actual test qualified control parameter. The data transmission component 125 arranges all actual test qualified control parameters in descending order according to the size of the evaluation value, that is, the larger the evaluation value, the higher the comprehensive beam quality, the better the performance of the parameter combination, and the higher the ranking, forming an actual test qualified control parameter list. Afterwards, the data transmission component 125 sends the generated actual test qualified control parameter list to the user terminal 11. After receiving the actual test qualified control parameter list, the user terminal 11 can intuitively see the ranking of the advantages and disadvantages of each parameter combination.

[0098] In summary, the beam quality testing device for far-field lasers provided in the embodiments of the present application has the following technical effects:

[0099] The user end is used to receive beam quality test requests, where the beam quality test request includes the far-field laser application scenario type, clarifies the test requirements and target scenarios, and provides important input and reference basis for subsequent test optimization. The dynamic weighting component is used to receive the far-field laser application scenario type from the user end, weight the beam quality evaluation factor, generate the beam quality evaluation factor weight distribution result, adaptively adjust the evaluation factor weight for different application scenarios, establish an evaluation index system that matches the application requirements, and lay the foundation for test optimization. The function configuration component is used to receive the beam quality evaluation factor weight distribution result from the dynamic weighting component, configure the beam quality evaluation function, and build a complete beam quality evaluation mathematical model based on the evaluation factor weight to guide subsequent virtual tests and actual tests. The virtual test component is used to activate the beam quality evaluation factor prediction network, combine the beam quality evaluation function, perform virtual tests on the far-field laser emission control parameters, obtain theoretical qualified control parameters, and provide optimization directions for actual tests. The actual test component is used to receive the theoretical qualified control parameters from the virtual test component, conduct actual tests in combination with the beam quality evaluation function, obtain the qualified control parameters for the actual test, and conduct test verification in the actual environment guided by the virtual test results, further optimize the control parameters that ultimately meet the requirements of the evaluation function, and complete the test closed loop. The data transmission component sends the qualified control parameters for the actual test to the user end when the number of qualified control parameters for the actual test is not equal to 0, and promptly feeds back the final results of the test optimization to the user, providing guidance for actual applications, thereby improving test efficiency and reducing actual test costs.

[0100] Any steps of the above-described device may be stored as computer instructions or programs in an unlimited computer memory, and may be called and recognized by an unlimited computer processor to implement any device in the embodiments of the present application, without any unnecessary restrictions.

[0101] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these changes and variations.

Claims

1. A beam quality test device for far-field lasers, characterized in that: include: A user end, configured to receive a beam quality test request, wherein the beam quality test request includes a far-field laser application scenario type; Data centers, including: A dynamic weighting component, used for receiving the far-field laser application scenario type from the user terminal, weighting the beam quality evaluation factor, and generating a beam quality evaluation factor weight distribution result; A function configuration component, used to receive the beam quality evaluation factor weight distribution result from the dynamic weighting component and configure a beam quality evaluation function; A virtual test component is used to activate the beam quality evaluation factor prediction network, and to perform a virtual test on the far-field laser emission control parameters in combination with the beam quality evaluation function to obtain theoretical qualified control parameters; An actual test component, used for receiving the theoretical qualified control parameters from the virtual test component, performing actual testing in combination with the beam quality evaluation function, and obtaining actual test qualified control parameters; The data transmission component sends the actual test qualified control parameters to the user end when the number of the actual test qualified control parameters is not equal to 0.

2. The device according to claim 1, characterized in that The steps of executing the dynamic weighting component include: The far-field laser application scenario type is identified through the operation effect evaluation index calibration table, and an operation effect evaluation index matching result is generated, wherein the operation effect evaluation index calibration table is an evaluation index pre-set for different application scenario types; Traversing the matching results of the operation effect evaluation index, performing correlation analysis on the far-field laser application scenario types, and obtaining several groups of correlation analysis results; The plurality of groups of correlation analysis results are fused to generate the beam quality evaluation factor weight distribution result.

3. The device according to claim 2, characterized in that The dynamic weighting component executes the steps of traversing the matching results of the operation effect evaluation index, performing a correlation analysis on the far-field laser application scenario type, and obtaining several groups of correlation analysis results. The execution steps include: Extracting a first operation effect evaluation indicator according to the operation effect evaluation indicator matching result; Taking the beam quality evaluation factor as an independent variable and the first operation effect evaluation index as a dependent variable, collecting multiple sets of beam quality evaluation factor record data and multiple first operation effect evaluation index record data; Performing normalization processing on the plurality of first operation effect evaluation index record data to construct an operation effect evaluation index characteristic value sequence; Performing same-attribute normalization processing on the multiple groups of beam quality evaluation factor record data to construct multiple beam quality evaluation factor eigenvalue sequences; Grey correlation processing is performed on the characteristic value sequence of the operation effect evaluation index and the characteristic value sequence of the multiple beam quality evaluation factors to obtain a first group of correlation analysis results, which are added to the plurality of groups of correlation analysis results.

4. The device according to claim 1, characterized in that The function configuration component execution steps include: Construct a beam quality evaluation function: ， in, Characterize the beam quality evaluation value, Characterizes the measurement value of the i-th evaluation factor, Characterizes the ideal threshold of the i-th evaluation factor, The exponential adjustment coefficient representing the i-th evaluation factor is used to control the rate of exponential growth. represents the weight of the i-th evaluation factor, represents the total number of beam quality evaluation factors considered, and i represents the i-th evaluation factor; The beam quality evaluation function is configured according to the beam quality evaluation factor weight distribution result.

5. The device according to claim 1, characterized in that The virtual test component execution steps include: The beam quality evaluation factors include far-field spot radius, far-field divergence angle, diffraction limit multiplication factor, Strehl ratio, M2 factor and surrounding energy ratio; The far-field laser emission control parameters include at least wavelength, power, pulse duration, repetition rate, beam diameter and working distance; Taking the wavelength, the power, the pulse duration, the repetition rate, the beam diameter and the working distance as input variables, traversing the far-field spot radius, the far-field divergence angle, the diffraction limit multiplication factor, the Strehl ratio, the M2 factor and the surrounding energy ratio, and constructing the beam quality evaluation factor prediction network; Performing a virtual test on the far-field laser emission control parameters according to the beam quality evaluation factor prediction network to obtain a beam quality evaluation factor prediction value set; Processing the beam quality evaluation factor prediction value set according to the beam quality evaluation function to generate a beam quality evaluation value; When the beam quality evaluation value meets the beam quality evaluation threshold, the far-field laser emission control parameter is added to the theoretical qualified control parameter.

6. The device according to claim 5, characterized in that The virtual test component performs the steps of: taking the wavelength, the power, the pulse duration, the repetition rate, the beam diameter and the working distance as input variables, traversing the far-field spot radius, the far-field divergence angle, the diffraction limit multiplication factor, the Strehl ratio, the M2 factor and the surrounding energy ratio, and constructing the beam quality evaluation factor prediction network, including: According to the wavelength, the power, the pulse duration, the repetition rate, the beam diameter and the working distance as independent variables, and the far-field spot radius as a dependent variable, collecting an independent variable array set and a dependent variable data set; According to the independent variable array set and the dependent variable data set, configuring a far-field spot radius prediction subnetwork; Traversing the far-field divergence angle, the diffraction limit multiplication factor, the Strehl ratio, the M2 factor and the surrounding energy ratio, constructing a far-field divergence angle prediction subnetwork, a diffraction limit multiplication factor prediction subnetwork, a Strehl ratio prediction subnetwork, an M2 factor prediction subnetwork and a surrounding energy ratio prediction subnetwork; The input layers of the far-field spot radius prediction subnetwork, the far-field divergence angle prediction subnetwork, the diffraction limit multiple factor prediction subnetwork, the Strehl ratio prediction subnetwork, the M2 factor prediction subnetwork and the surrounding energy ratio prediction subnetwork are merged to generate the beam quality evaluation factor prediction network.

7. The device according to claim 1, characterized in that The data transmission component performs the steps including: When the number of the control parameters that are actually tested and qualified is equal to 0, a far-field laser emission control parameter adjustment instruction is generated; Sending the far-field laser emission control parameter adjustment instruction to the virtual test component, adjusting the theoretical qualified control parameter, and obtaining the theoretical qualified updated control parameter; The control parameters are updated according to the theoretical qualifications and an iterative cycle is performed.

8. The device according to claim 4, characterized in that The data transmission component execution step also includes: When the number of the actual test qualified control parameters is not equal to 0, and is greater than or equal to 2, the actual test qualified control parameters are sorted from large to small according to the beam quality evaluation value to obtain a list of actual test qualified control parameters; The actual test qualified control parameter list is sent to the user terminal.

Images