Self-adaptive control method and system for laser marking of wiring terminal shell

Abstract

The invention discloses a self-adaptive control method and system for laser marking of a wiring terminal shell, and relates to the technical field of laser processing and intelligent manufacturing, and the method comprises the steps: obtaining surface image information, extracting initial surface state data, building a mapping relation between laser energy density and shell surface change, and building an initial marking control model; setting an initial control value according to the initial marking control model; marking information of the marking area is collected and subjected to fusion processing, and comprehensive marking state data is generated; outputting a real-time marking quality error and a change rate thereof according to the comprehensive marking state data, and obtaining an adjustment direction and amplitude of the laser data under a preset physical constraint condition; according to the adjustment direction and amplitude of the laser data, the physical constraint conditions are dynamically corrected in combination with historical marking data, and an optimal laser data combination is output; according to the method, through surface state self-adaption and historical data optimization, the problems of uneven laser marking quality and batch difference are solved.

Description

Technical Field

[0001] This invention relates to the field of laser processing and intelligent manufacturing technology, and more specifically, to an adaptive control method and system for laser marking on terminal block housings. Background Technology

[0002] In the modern electronics and electrical industry, terminal blocks are crucial components for circuit connections, and their quality and marking clarity directly impact the reliability and maintenance efficiency of terminal equipment. With the continuous improvement of industrial automation, laser marking technology has gradually become the mainstream method for marking terminal blocks due to its advantages such as non-contact operation, high precision, durability, and fast processing speed. However, existing technologies still face numerous challenges and limitations in practical applications, necessitating urgent improvement and optimization.

[0003] For example, the invention patent with publication number CN108406124A discloses a laser marking method and system, relating to the field of laser processing technology. The system includes a laser marking component, a marking test component, and a control module. The laser marking component emits laser light to mark the image, and the marking test component detects the relative grayscale value of the image to be inspected, which is marked in real-time on the actual material. This invention, during laser marking, obtains the relative grayscale value of the actual material by actually marking it. Based on this value, suitable laser marking parameters are selected, allowing the laser marking component to adjust its marking program and emit appropriate laser light to mark the corresponding material. This solves the problem of inconsistent marking effects caused by differences in actual materials. Therefore, it ensures the consistency of the marked image effect, improves production efficiency, and reduces the scrap rate.

[0004] For example, the invention patent with publication number CN110788489A discloses a method for automatically correcting the position of laser marking, including: providing a marking template; conducting a marking test according to the marking template to obtain a marking sample; acquiring the test results using an image acquisition system; comparing the test results with the marking template to obtain a deviation value; and marking based on the deviation value; wherein, the marking template includes the coordinates of n marking contents x1, y1, x2, y2, ..., xn, yn, where n is... The test results are natural numbers greater than or equal to 1; the actual marked coordinates of n marked contents are x1', y1', x2', y2', ..., xn', yn'; the deviation values ​​are △x1, △y1, △x2, △y2, ..., △xn, △yn, where △x1 = x1 - x1', △y1 = y1 - y1', △x2 = x2 - x2', △y2 = y2 - y2', ..., △xn = xn - xn', △yn = yn - yn'. Using this laser marking method, the marking results are highly accurate, require no post-processing adjustments, and have high marking efficiency.

[0005] The above-disclosed technical solutions have at least the following technical problems:

[0006] Traditional laser marking techniques typically employ fixed energy and preset parameters, lacking real-time sensing and adaptive adjustment of the shell surface condition. Furthermore, when adjusting laser parameters, there is a lack of real-time constraints on material tolerance limits and laser safety ranges, easily leading to over-marking or damage. To address these issues, this invention proposes a solution. Summary of the Invention

[0007] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an adaptive control method and system for laser marking of terminal housings, which solves the problems of uneven laser marking quality and batch differences by adapting to surface conditions and optimizing historical data.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] An adaptive control method for laser marking of terminal block housings includes: acquiring surface image information of the target terminal block housing, extracting initial surface state data, establishing a mapping relationship between laser energy density and surface changes, and constructing an initial marking control model; setting initial control values ​​according to the initial marking control model and driving the laser to perform marking; during the marking process, collecting marking information of the marking area and performing fusion processing to generate comprehensive marking state data; outputting real-time marking quality error and its rate of change based on the comprehensive marking state data, and obtaining the adjustment direction and amplitude of the laser data under preset physical constraints; dynamically correcting the physical constraints based on the adjustment direction and amplitude of the laser data and combining historical marking data, and outputting the optimal laser data combination.

[0010] In a preferred embodiment, the steps of acquiring surface image information of the target terminal housing, extracting initial surface state data, establishing a mapping relationship between laser energy density and housing surface changes, and constructing an initial marking control model are as follows: The area to be marked on the target terminal housing is located, initial image information of the housing surface is acquired, and the initial image information is preprocessed; based on the preprocessed surface image information, initial surface state data is extracted; the initial surface state data undergoes a unified scale transformation, and weights are assigned according to the intensity of the influence of laser marking to form a comprehensive surface state vector; based on the comprehensive surface state vector, the initial interval of the housing surface's response to laser energy is output, and the range of laser energy density adjustment is constrained within different intervals; the housing is marked within the constrained laser energy density range, and corresponding surface change feature data is extracted; based on the laser energy density, the comprehensive surface state vector, and the corresponding surface change feature data, a mapping relationship between laser input and housing surface change trends is established; the mapping relationship between laser input and housing surface change trends, along with the comprehensive surface state vector, is introduced into the marking control logic to construct the initial marking control model.

[0011] In a preferred embodiment, the step of outputting the initial range of the shell surface's response to laser energy based on the comprehensive surface state vector, and constraining the adjustment range of laser energy density within different ranges, is as follows: Based on the comprehensive surface state vector, the response of the shell surface to laser energy input is quantitatively analyzed to generate an energy response sensitivity; based on the energy response sensitivity, the response of the shell surface to laser energy input is divided into several initial energy response ranges; the laser energy density adjustment boundary corresponding to the initial energy response range is constrained, so that the laser energy density is limited to the adjustment boundary matching the current energy response range; during the calibration process, the shell surface features are collected in real time and the comprehensive surface state vector is updated, the energy response sensitivity is re-output and the corresponding initial energy response range is determined, and the corresponding laser energy density adjustment boundary is called according to the current range.

[0012] In a preferred embodiment, the step of setting initial control values ​​according to the initial marking control model and driving the laser for marking is as follows: In the initial marking control model, feature analysis is performed on the comprehensive surface state vector to extract key states; the key states are matched with the energy response judgment rules pre-established in the model; based on the matching result, the energy response sensitivity of the current shell is output, and the initial energy response range of the shell in the initial marking control model is determined based on the energy response sensitivity; the laser energy density adjustment boundary pre-associated with the initial energy response range is called as the laser energy density adjustment boundary allowed in the initial stage of the current shell marking; within the adjustment boundary, an initial control value is generated based on the energy response sensitivity and the regional response differences, and the initial control value is dynamically corrected based on the material batch information and historical marking data; the laser is driven to start marking along a preset path according to the initial control value.

[0013] In a preferred embodiment, within the adjustment boundary, an initial control value is generated based on energy response sensitivity and regional response differences. This initial control value is then dynamically corrected based on material batch information and historical marking data. Specifically: The material batch information and historical marking data of the current shell are obtained; statistical analysis of the historical marking data is performed to extract the deviation trend and drift characteristics of different batches of materials in response to laser data, obtaining the deviation range of the current batch; the marking area is divided based on the shell surface feature image and energy response sensitivity, and the response differences of different areas to changes in laser data are identified, i.e., regional response differences; the reference laser data of the historical marking data is obtained, and the reference laser data is corrected based on the batch deviation range and regional response differences to generate an initial control value; the initial control value of each sensitive area is controlled to ensure that the initial marking of each sensitive area meets the expected requirements; the initial control value after regional correction is matched with the safe operating range of the laser equipment and the material tolerance limit, and control values ​​exceeding the limits are constrained to obtain the final control value; based on the corrected control value, the laser is driven to begin marking along a preset path.

[0014] In a preferred embodiment, the step of collecting and fusing the marking information of the marking area to generate comprehensive marking status data specifically involves: collecting marking process data of the marking area and associating the marking process data with the laser scanning path and time axis; using the laser scanning timing sequence as a reference, performing time synchronization calibration on marking process data from different sources, and resampling data with inconsistent sampling frequencies and acquisition delays; using the spatial coordinates of the marking area as a reference, performing spatial mapping on the time-synchronized calibrated marking process data to achieve spatial alignment, and performing noise suppression on the aligned marking process data. The process involves: applying trend constraints to the data on reflected light intensity changes and temperature rise trends during the marking process, and marking abnormal sampling points that deviate from the current marking state; extracting process feature data based on the noise-suppressed marking process data; associating the process feature data with the marking path location and time nodes and storing them in a structured manner to form a spatiotemporal feature sequence of the marking process; extracting the process feature data corresponding to the current marking segment from the spatiotemporal feature sequence of the marking process, normalizing and uniformly scaling the extracted process feature data, and assigning weights according to their intensity of marking quality representation to construct a comprehensive marking state vector.

[0015] In a preferred embodiment, the step of outputting the real-time marking quality error and its rate of change based on the comprehensive marking state data, and obtaining the adjustment direction and amplitude of the laser data under preset physical constraints, specifically includes: determining the marking quality reference state according to the expected requirements of the marking task; comparing the comprehensive marking state data acquired at the current moment with the marking quality reference state to calculate the real-time marking quality error; performing time-series analysis on the real-time marking quality error at continuous marking times to output the rate of change of the marking quality error as the marking path progresses; extracting the state characteristics of the marking process evolution trend from the comprehensive marking state data; and adjusting the laser data according to the real-time marking quality... The system generates a comprehensive judgment result of the marking quality deviation by considering the measurement error, the rate of change of the marking quality error, and the state characteristics, distinguishing between transient fluctuations and persistent deviations. Based on the comprehensive judgment result and the mapping relationship between the laser input and the changing trend of the shell surface, the system determines the adjustment direction of the laser data. Based on the determined laser data adjustment direction, and considering the energy response range of the current shell and the corresponding laser energy density adjustment boundary, the system outputs the adjustment amplitude of the laser data. Based on preset physical constraints, the system restricts the adjustment direction and amplitude of the laser data. Under the premise of satisfying the physical constraints, the system outputs the adjustment direction and amplitude of the laser data.

[0016] In a preferred embodiment, the step of generating a comprehensive judgment result of the marking quality deviation based on the real-time marking quality error, the rate of change of the marking quality error, and state characteristics, and distinguishing between transient fluctuations and persistent deviations, is as follows: Based on the real-time marking quality error, the current marking quality deviation is compared with a preset allowable deviation range. When the real-time marking quality error is within the allowable deviation range, the current deviation is determined to be acceptable. When the real-time marking quality error exceeds the allowable deviation range, a trend judgment is made based on the rate of change of the marking quality error. When the marking quality error shows an increasing trend in the same direction or does not show a significant decrease at multiple consecutive marking times, the deviation is determined to have a continuous development trend. If the deviation exceeds the allowable deviation range but the rate of change of the marking quality error shows a rapid decline, the deviation is determined to be a transient fluctuation. The direction of change of state characteristics related to the marking quality deviation is extracted from the comprehensive marking status data. It is determined whether the direction of change of state characteristics is consistent with the direction of change of the real-time marking quality error. If the direction is consistent, the deviation is confirmed to originate from the actual change in the marking process status. If the real-time marking quality error exceeds the allowable deviation range, the rate of change of the marking quality error shows a continuous trend, and the direction of change of state characteristics is consistent with the deviation direction, the current marking quality deviation is determined to be a persistent deviation. If the above persistent deviation determination conditions are not met, the current marking quality deviation is determined to be a transient fluctuation deviation.

[0017] In a preferred embodiment, the step of dynamically correcting the physical constraints based on the adjustment direction and amplitude of the laser data, combined with historical marking data, and outputting the optimal laser data combination is as follows: Historical marking data corresponding to the current shell material type, material batch, and surface condition is retrieved; based on the historical marking data, the effective boundaries of the preset physical constraints are corrected; the corrected physical constraints are used as new constraint boundaries and matched with the current laser data adjustment direction and amplitude to determine whether the adjustment triggers over-marking, material damage, or marking instability risks; when the adjusted laser data falls within the corrected physical constraint boundaries, the laser data adjustment direction and amplitude are confirmed as effective adjustments; when the adjusted laser data exceeds the corrected physical constraint boundaries, the laser data adjustment amplitude is redistributed according to the safe adjustment path for the corresponding working condition in the historical marking data, and a suboptimal adjustment direction is output; the corrected laser data is combined and optimized to generate a laser data combination that meets the marking quality requirements and is within the corrected physical constraints; the laser data combination is output as the optimal laser data combination for the current marking stage.

[0018] A system for adaptive control of laser marking on terminal block housings includes a data acquisition module, an initial control value module, a fusion module, a constraint module, and an output module, with interconnections between the modules. The data acquisition module acquires surface image information of the target terminal block housing, extracts initial surface state data, establishes a mapping relationship between laser energy density and surface changes, and constructs an initial marking control model. The initial control value module sets initial control values ​​based on the initial marking control model and drives the laser for marking. The fusion module acquires marking information of the marking area during the marking process, performs fusion processing, and generates comprehensive marking state data. The constraint module outputs real-time marking quality error and its rate of change based on the comprehensive marking state data, and obtains the adjustment direction and amplitude of the laser data under preset physical constraints. The output module dynamically corrects the physical constraints based on the adjustment direction and amplitude of the laser data, combined with historical marking data, and outputs the optimal laser data combination.

[0019] The technical effects and advantages of the adaptive control method and system for laser marking of terminal housings according to the present invention are as follows:

[0020] 1. This invention sets initial control values ​​through an initial marking control model and dynamically corrects them by combining material batch information and historical marking data, achieving regional optimized control of laser energy. The differences in surface features and energy response sensitivity of different regions can be accurately identified and matched, enabling the laser output to adapt to the physical properties of different regions and ensuring optimal marking depth, clarity, and consistency. Simultaneously, by real-time acquisition and fusion processing of the marking area, comprehensive marking status data is generated, achieving full perception of the marking process and enabling rapid detection of abnormal deviations or quality fluctuations.

[0021] 2. This invention calculates the marking quality error and its rate of change, distinguishing between transient fluctuations and persistent deviations. By combining the real-time deviation with the laser input-shell response mapping relationship, the system outputs laser data to adjust the direction and amplitude under preset physical constraints, ensuring both precise and safe adjustment. Simultaneously, the system can dynamically correct the physical constraints by calling historical marking data, achieving optimal matching of laser data combinations and avoiding over-marking, material damage, or marking instability. This closed-loop adaptive adjustment mechanism significantly improves the stability and reliability of the marking process. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating an adaptive control method for laser marking on a terminal block housing according to the present invention.

[0023] Figure 2 This is a schematic diagram of the system structure of an adaptive control method for laser marking on a terminal block housing according to the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1, Figure 1 An adaptive control method for laser marking on terminal block housings is provided by the present invention, comprising:

[0026] S1. Obtain surface image information of the target terminal housing, extract initial surface state data, establish the mapping relationship between laser energy density and housing surface changes, and construct the initial marking control model;

[0027] In this embodiment, surface image information of the target terminal housing is acquired, initial surface state data is extracted, and a mapping relationship between laser energy density and changes in the housing surface is established to construct an initial marking control model, as detailed below:

[0028] Before laser marking begins, the area to be marked on the target terminal housing is located, and initial image information of the housing surface is acquired under fixed optical conditions. The initial image information is then subjected to multi-dimensional preprocessing, which includes noise reduction, illumination compensation, and color space conversion. It also includes adaptive enhancement processing for high-reflection and low-contrast areas to highlight the feature information of the housing surface related to the laser action.

[0029] Based on the preprocessed surface image information, initial surface state data is extracted. The initial surface state data includes grayscale distribution for characterizing laser absorption characteristics, texture feature data for characterizing local heat accumulation trends, and brightness variation for characterizing surface reflection differences.

[0030] The initial surface state data is subjected to a uniform scale transformation, and weights are assigned according to the intensity of the influence of laser marking to form a comprehensive surface state vector characterizing the sensitivity of the shell to the initial marking.

[0031] Based on the comprehensive surface state vector, the initial range of the shell surface response to laser energy is output, and the range of laser energy density control is constrained within different ranges;

[0032] Within the constrained laser energy density range, the shell is marked, and the corresponding surface change feature data is extracted. The surface change feature data is used as the marking output variable to characterize the shell's response to the laser. The surface change feature data is used to characterize the response state of the shell surface after laser marking. It corresponds to the initial surface state data, but is not the same as the amount of change in the initial surface state data. Instead, it is the result data obtained by extracting the image features of the marked area and the unmarked area, including grayscale and contrast feature data of the optical property changes in the marked area and texture and edge feature data of the marking contour changes.

[0033] Based on laser energy density, comprehensive surface state vector and corresponding surface change feature data, a mapping relationship between laser input and shell surface change trend is established. The mapping relationship is used to limit the directionality and boundary conditions of laser data adjustment. The shell surface change trend includes increased or decreased reflected light intensity, accelerated or slowed temperature rise, and deepened or shallowed texture.

[0034] The mapping relationship between laser input and the changing trend of the shell surface, along with the comprehensive surface state vector, is introduced into the marking control logic to construct an initial marking control model. This model enables the model to distinguish between different shell surface states at the beginning of the marking process.

[0035] In this embodiment, based on the comprehensive surface state vector, the initial range of the shell surface response to laser energy is output, and the control range of laser energy density within different ranges is constrained, as follows:

[0036] Based on the comprehensive surface state vector, the response of the shell surface to laser energy input is quantitatively analyzed to generate energy response sensitivity. Specifically, each feature of the comprehensive surface state vector is compared with pre-established laser response calibration data. The calibration data records the influence of laser input on the marking results under different surface state features, including indicators such as surface grayscale change, texture evolution, contour stability, and contrast change. By calculating the response intensity of each feature to laser input and weighting and summarizing them according to their influence on the marking quality, a normalized comprehensive response index is obtained, thereby quantifying and generating energy response sensitivity.

[0037] Based on the energy response sensitivity, the response of the shell surface to laser energy input is divided into at least one initial energy response interval. Each interval represents the intensity level of the shell to the laser action in the initial stage of calibration. The initial energy response interval is obtained through calibration test. The shell is tested under different laser energies, the surface response after calibration is measured, and the energy interval corresponding to different sensitivity levels is determined.

[0038] The laser energy density adjustment boundary corresponding to the initial energy response range is constrained so that the laser energy density is limited to the control boundary that matches the current energy response range;

[0039] During the marking process, the surface features of the shell are acquired in real time and the comprehensive surface state vector is updated. The energy response sensitivity is re-output and the initial energy response range is determined. The corresponding laser energy density adjustment boundary is called according to the current range to realize the dynamic adaptive adjustment of the laser data.

[0040] In this embodiment, the mapping relationship between the laser input and the changing trend of the shell surface, along with the comprehensive surface state vector, is introduced into the marking control logic to construct an initial marking control model, as follows:

[0041] By combining the comprehensive surface state vector and the mapping relationship between laser input and shell surface change trend, the energy response sensitivity and initial energy response range corresponding to the current shell surface are determined, thereby determining the initial laser setting value and the adjustable range.

[0042] Based on the judgment result, the control logic generates an initial laser data combination, which includes laser power, scanning speed, pulse frequency, etc., so that the current shell surface state can be adaptively controlled at the beginning of the marking stage.

[0043] In the control logic, the control range and energy response interval are used as boundary conditions, so that any initial control value and subsequent adjustment are limited to the preset safety and effective range, avoiding under-marking, over-marking or material damage.

[0044] By integrating the judgment logic, initial laser data generation, and boundary conditions, an initial marking control model is formed, enabling the system to classify and control different shell surface states at the beginning of the marking stage.

[0045] S2, set the initial control value according to the initial marking control model, and drive the laser to perform marking;

[0046] In this embodiment, initial control values ​​are set according to the initial marking control model, and the laser is driven to perform marking, as follows:

[0047] A comprehensive surface state vector is generated based on the multidimensional features extracted from the shell surface image, and combined with the energy response sensitivity, the shell surface's response capability to laser energy input is quantified.

[0048] In the initial calibration control model, feature analysis is performed on the comprehensive surface state vector to extract key states such as shell laser energy absorption capacity, heat accumulation tendency and reflection characteristics;

[0049] The key states are matched with the energy response determination rules pre-established in the model. These energy response determination rules are used to describe the laser energy response levels corresponding to different combinations of surface states.

[0050] Based on the matching results, the energy response sensitivity of the current shell is output, and the initial energy response range of the shell in the initial calibration control model is determined based on the energy response sensitivity.

[0051] Call the laser energy density adjustment boundary that is pre-associated with the initial energy response range as the laser energy density adjustment boundary allowed in the current initial stage of shell marking;

[0052] Within the adjustment boundary, initial control values ​​are generated based on energy response sensitivity and regional response differences, and the initial control values ​​are dynamically corrected based on material batch information and historical calibration data.

[0053] The laser is driven to begin marking along a preset path based on the initial control value.

[0054] In this embodiment, within the adjustment boundary, an initial control value is generated based on the energy response sensitivity and regional response differences. This initial control value is then dynamically corrected based on material batch information and historical calibration data, as detailed below:

[0055] Obtain the material batch information of the current shell, including composition, thickness, surface treatment method and finish;

[0056] Retrieve historical marking data, including the marking effects of different batches of materials under various laser data, such as color changes, texture features, and contour depth;

[0057] Statistical analysis of historical marking data is performed to extract the deviation trend and drift characteristics of different batches of materials in response to laser data, and the deviation range of the current batch is obtained.

[0058] Based on the surface feature image of the shell and the energy response sensitivity, the marked area is divided into sensitive areas and low-sensitivity areas. The response differences of different areas to changes in laser data are also identified, i.e., regional response differences.

[0059] The reference laser data of historical marking data is obtained, and the reference laser data is corrected by combining the batch deviation range and the response difference of each region to generate initial control values. Among them, the laser power, scanning speed or pulse characteristics of the energy response sensitive regions are adjusted according to the deviation direction and amplitude, the low sensitivity regions are adjusted by coordinated constraints, and the adjustment amplitude of each sub-region is limited to meet the material tolerance limit and equipment safety range. Finally, the laser parameters of each sub-region are combined and integrated into the initial control values.

[0060] The initial control value of each sensitive area is controlled to ensure that the initial marking of each sensitive area meets the expected requirements, while the low-sensitivity area is coordinated and constrained to ensure the overall marking consistency.

[0061] The initial control values, after regional correction, are matched with the safe operating range of the laser equipment and the material tolerance limit. Control values ​​that exceed the limits are constrained to obtain the final control values.

[0062] Based on the corrected control values, the laser is driven to begin marking along a preset path.

[0063] S3, during the marking process, collect marking information of the marking area, perform fusion processing, and generate comprehensive marking status data;

[0064] In this embodiment, the marking information of the marking area is collected and fused to generate comprehensive marking status data, as follows:

[0065] During the laser marking process, the marking process data of the marking area is collected in real time. The marking process data includes process image data of the marking area, data on the change of reflected light intensity of the shell surface to the laser, data on the temperature rise trend of the marking area and equivalent thermal response data, as well as equivalent energy feedback from the laser output side.

[0066] The marking process data is associated with the laser scanning path and time axis, so that each acquisition moment corresponds to a clear marking position and laser action state;

[0067] Based on the laser scanning timing sequence, time synchronization calibration is performed on marking process data from different sources, and data with inconsistent sampling frequencies and acquisition delays are resampled or interpolated.

[0068] Using the spatial coordinates of the marking area as a reference, spatial mapping is performed on the marking process data after time synchronization calibration to achieve alignment of the marking process data in the spatial dimension.

[0069] Noise suppression processing is performed on the marking process data that has achieved time and space alignment, including temporal filtering of the process image to reduce the abnormal effects caused by splashing, transient reflections or plasma disturbances;

[0070] Trend constraint processing is performed on the data of the calibration process of reflected light intensity change and temperature rise trend to suppress discontinuous or non-physically meaningful abrupt changes, and abnormal sampling points that deviate significantly from the current calibration state are marked to reduce the weight of abnormal sampling points in subsequent feature extraction and fusion processes.

[0071] Based on the marking process data after noise suppression processing, process feature data for characterizing the marking process state is extracted. The process feature data includes grayscale change features, contrast change features, texture evolution features, edge continuity features, contour stability features, and reflection intensity change features or equivalent thermal response features of the marking area.

[0072] The process feature data is associated with the location and time node of the marking path and stored in a structured manner to form a spatiotemporal feature sequence of the marking process;

[0073] The process feature data corresponding to the current marking segment is extracted from the spatiotemporal feature sequence of the marking process. The extracted process feature data is normalized and uniformly scaled, and weights are assigned according to their intensity of marking quality representation to construct a comprehensive marking state vector. The intensity of marking quality representation refers to the degree of correlation between each process feature data and the expected marking quality. The marking quality includes at least one or more of marking contrast, marking sharpness, marking outline integrity, marking consistency, and marking depth stability. The degree of correlation between each process feature data and the marking quality is obtained through statistical analysis of historical marking data.

[0074] S4 outputs the real-time marking quality error and its rate of change based on the comprehensive marking status data, and obtains the adjustment direction and amplitude of the laser data under preset physical constraints;

[0075] In this embodiment, the real-time marking quality error and its rate of change are output based on the comprehensive marking status data, and the adjustment direction and amplitude of the laser data are obtained under preset physical constraints, as follows:

[0076] The marking quality reference state is determined according to the expected requirements of the marking task. The marking quality reference state includes marking contrast, marking sharpness, marking outline integrity, and marking consistency.

[0077] The comprehensive marking status data acquired at the current moment is compared with the marking quality reference status to calculate the real-time marking quality error, which is used to characterize the degree of deviation between the current marking process status and the expected marking quality.

[0078] A time-series analysis of the real-time marking quality error is performed at continuous marking times, and the rate of change of the marking quality error as the marking path progresses is output to characterize the development trend of the marking quality deviation.

[0079] State features of the marking process evolution trend are extracted from the comprehensive marking state data, including the optical change trend, texture evolution trend and contour stability change trend of the marking area, which are used to reflect the dynamic response state of the marking process.

[0080] Based on the real-time marking quality error, the rate of change of marking quality error, and state characteristics, a comprehensive judgment result of marking quality deviation is generated to distinguish between transient fluctuations and persistent deviations.

[0081] Based on the comprehensive judgment results and the mapping relationship between the laser input and the change trend of the shell surface, the adjustment direction of the laser data is determined. The mapping relationship is used to limit the effective direction of laser data adjustment for improving the marking quality.

[0082] Based on the determined laser data adjustment direction, and combined with the energy response range of the current shell and the corresponding laser energy density adjustment boundary, the adjustment range of the laser data is output, where different energy response ranges correspond to different adjustment sensitivities.

[0083] Based on preset physical constraints, the adjustment direction and adjustment range of laser data are restricted. The physical constraints include at least the safe operating range of the laser equipment, the material tolerance limit, and the marking stability requirements.

[0084] Under the premise of satisfying the physical constraints, the adjustment direction and adjustment range of the output laser data are determined.

[0085] In this embodiment, a comprehensive judgment result of the marking quality deviation is generated based on the real-time marking quality error, the rate of change of the marking quality error, and the state characteristics, distinguishing between transient fluctuations and persistent deviations, as detailed below:

[0086] Based on the real-time marking quality error, the current marking quality deviation is compared with the preset allowable deviation range. When the real-time marking quality error is within the allowable deviation range, the current deviation is determined to be acceptable.

[0087] When the real-time marking quality error exceeds the allowable deviation range, the trend is determined by combining the rate of change of the marking quality error. When the marking quality error increases in the same direction or does not show a significant decline at multiple consecutive marking times, the deviation is determined to have a continuous development trend.

[0088] When the real-time marking quality error exceeds the allowable deviation range, but the rate of change of the marking quality error shows a rapid decline or a reversal of direction, the deviation is judged as a transient fluctuation.

[0089] Extract the direction of change of state features related to the marking quality deviation from the comprehensive marking state data;

[0090] Determine whether the direction of change of state characteristics is consistent with the direction of change of real-time marking quality error. If the direction of change is consistent, confirm that the deviation originates from the actual change of state during the marking process.

[0091] When the real-time marking quality error exceeds the allowable deviation range, the rate of change of marking quality error shows a continuous trend, and the direction of change of state characteristics is consistent with the direction of deviation, the current marking quality deviation is determined to be a continuous deviation.

[0092] If the above-mentioned persistent deviation determination conditions are not met, the current marking quality deviation is determined to be a transient fluctuation deviation.

[0093] In this embodiment, based on the comprehensive judgment results and the mapping relationship between the laser input and the changing trend of the shell surface, the adjustment direction of the laser data is determined as follows:

[0094] The direction of the surface change trend of the shell is matched with the pre-established mapping relationship to determine the direction of laser action that causes the deviation in the marking quality under the current laser input state;

[0095] When the comprehensive judgment results indicate that the marking quality deviation is a persistent deviation, the laser input adjustment direction is determined in reverse according to the mapping relationship to counteract the changing trend of the shell surface, so as to suppress the further development of the marking quality deviation.

[0096] If the comprehensive judgment results indicate that the marking quality deviation is a transient fluctuation, maintain the current laser input direction or adjust the output direction gently to avoid over-adjustment that introduces new marking instability factors;

[0097] The determined laser input adjustment direction is output as the laser data adjustment direction.

[0098] In this embodiment, based on determining the laser data adjustment direction, and combining the energy response range of the current housing and the corresponding laser energy density adjustment boundary, the adjustment range of the laser data is output as follows:

[0099] Obtain the energy response range to which the current shell belongs, and call the laser energy density adjustment boundary corresponding to the energy response range. The adjustment range is limited to the minimum and maximum adjustment allowed during the marking process.

[0100] The deviation level of the marking quality is determined based on the absolute value and rate of change of the real-time marking quality error. The severity level is used to characterize whether the current marking quality deviation is in a state of slight deviation, moderate deviation or significant deviation.

[0101] The deviation level of the marking quality is mapped to the adjustment ratio within the energy response range, where a smaller adjustment ratio corresponds to an energy response range with higher energy response sensitivity, and a larger adjustment ratio corresponds to an energy response range with lower energy response sensitivity.

[0102] Based on the adjustment ratio coefficient, the basic adjustment range of the output laser data within the corresponding laser energy density adjustment boundary is made so that the basic adjustment range increases with the increase of the marking quality deviation level.

[0103] Based on the changing trend of the marking quality deviation, the basic adjustment range is adjusted accordingly. Specifically, when the marking quality deviation shows an aggravating trend, the adjustment range is increased, and when the marking quality deviation shows a weakening trend, the adjustment range is decreased.

[0104] The corrected adjustment range is matched with the minimum resolvable adjustment step size of the laser device, and adjustment ranges smaller than the minimum resolvable adjustment step size are merged.

[0105] Under the premise of satisfying the boundary constraints of laser energy density adjustment, the output laser data adjustment amplitude corresponds to the laser data adjustment direction.

[0106] S5, based on the adjustment direction and amplitude of the laser data, combined with historical marking data, dynamically corrects the physical constraints and outputs the optimal combination of laser data;

[0107] In this embodiment, the physical constraints are dynamically corrected based on the adjustment direction and amplitude of the laser data, combined with historical marking data, to output the optimal combination of laser data, as follows:

[0108] Obtain the laser data output from the previous stage to adjust the direction and adjustment range;

[0109] Retrieve historical marking data corresponding to the current shell material type, material batch, and surface condition. The historical marking data includes marking quality results formed under different laser data combinations and their corresponding physical constraint triggering conditions.

[0110] Based on historical calibration data, the effective boundaries of preset physical constraints are corrected. The correction includes fine-tuning of material tolerance limits, safe energy upper limits, stable calibration lower limits, or continuous operation constraint ranges.

[0111] The revised physical constraints are used as the new constraint boundaries and matched with the current laser data adjustment direction and adjustment range to determine whether there is a risk of over-marking, material damage or unstable marking.

[0112] When the adjusted laser data falls within the corrected physical constraint boundary, the adjustment direction and adjustment range of the laser data are confirmed to be effective.

[0113] When the adjusted laser data exceeds the corrected physical constraint boundary, the laser data adjustment range is redistributed according to the safe adjustment path of the corresponding working condition in the historical calibration data, and the original adjustment direction is kept unchanged or the suboptimal adjustment direction is output.

[0114] The corrected laser data is combined and optimized to generate a laser data combination that meets the marking quality requirements and is within the corrected physical constraints.

[0115] The combined laser data output is used as the optimal laser data combination for the current marking stage to drive the laser equipment to continue marking.

[0116] Example 2, Figure 2 The present invention provides a system for an adaptive control method for laser marking on terminal housings, comprising a data acquisition module, an initial control value module, a fusion module, a constraint module, and an output module, with connections between the modules;

[0117] The data acquisition module is used to acquire surface image information of the target terminal housing, extract initial surface state data, establish the mapping relationship between laser energy density and housing surface changes, and construct an initial marking control model.

[0118] The initial control value module is used to set the initial control value according to the initial marking control model and drive the laser to perform marking.

[0119] The fusion module is used to collect the marking information of the marking area during the marking process, and perform fusion processing to generate comprehensive marking status data;

[0120] The constraint module is used to output the real-time marking quality error and its rate of change based on the comprehensive marking status data, and to obtain the adjustment direction and amplitude of the laser data under preset physical constraint conditions;

[0121] The output module is used to dynamically correct the physical constraints based on the adjustment direction and amplitude of the laser data and historical marking data, and output the optimal combination of laser data.

[0122] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0123] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0124] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0125] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0126] The above description is merely a specific embodiment 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.

[0127] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive control method for laser marking on terminal block housings, characterized in that, include: Acquire surface image information of the target terminal housing, extract initial surface state data, establish the mapping relationship between laser energy density and housing surface changes, and construct an initial marking control model; Initial control values ​​are set according to the initial marking control model, and the laser is driven to perform marking. During the marking process, marking information of the marking area is collected and fused to generate comprehensive marking status data; Based on the comprehensive marking status data, the real-time marking quality error and its rate of change are output, and under preset physical constraints, the adjustment direction and amplitude of the laser data are obtained; Based on the adjustment direction and amplitude of the laser data, and combined with historical marking data, the physical constraints are dynamically corrected to output the optimal combination of laser data.

2. The adaptive control method for laser marking on a terminal block housing according to claim 1, characterized in that, The process involves acquiring surface image information of the target terminal housing, extracting initial surface state data, establishing a mapping relationship between laser energy density and changes in the housing surface, and constructing an initial marking control model, as detailed below: The area to be marked on the target terminal housing is located, the initial image information of the housing surface is obtained, and the initial image information is preprocessed. Based on the preprocessed surface image information, the initial surface state data is extracted; The initial surface state data is subjected to a uniform scale transformation, and weights are assigned according to the intensity of the influence of the laser marking to form a comprehensive surface state vector; Based on the comprehensive surface state vector, the initial range of the shell surface response to laser energy is output, and the range of laser energy density control is constrained within different ranges; The shell is marked within the constrained laser energy density range, and the corresponding surface change feature data is extracted. Based on laser energy density, comprehensive surface state vector and corresponding surface change characteristic data, a mapping relationship between laser input and shell surface change trend is established. The mapping relationship between laser input and the changing trend of the shell surface, along with the comprehensive surface state vector, is introduced into the marking control logic to construct the initial marking control model.

3. The adaptive control method for laser marking on a terminal block housing according to claim 2, characterized in that, The initial range of the shell surface response to laser energy is output based on the comprehensive surface state vector, and the control range of laser energy density is constrained within different ranges, as follows: Based on the comprehensive surface state vector, the response of the shell surface to laser energy input is quantitatively analyzed to generate energy response sensitivity; Based on the energy response sensitivity, the response of the shell surface to laser energy input is divided into several initial energy response intervals; The laser energy density adjustment boundary corresponding to the initial energy response range is constrained so that the laser energy density is limited to the control boundary that matches the current energy response range; During the marking process, the surface features of the shell are acquired in real time and the comprehensive surface state vector is updated. The energy response sensitivity is re-output and the initial energy response range is determined. The corresponding laser energy density adjustment boundary is called according to the current range.

4. The adaptive control method for laser marking on a terminal block housing according to claim 1, characterized in that, The process of setting initial control values ​​based on the initial marking control model and driving the laser for marking is as follows: In the initial calibration control model, feature analysis is performed on the comprehensive surface state vector to extract key states; Match the critical states with the energy response determination rules pre-established in the model; Based on the matching results, the energy response sensitivity of the current shell is output, and the initial energy response range of the shell in the initial calibration control model is determined based on the energy response sensitivity. Call the laser energy density adjustment boundary that is pre-associated with the initial energy response range as the laser energy density adjustment boundary allowed in the current initial stage of shell marking; Within the adjustment boundary, initial control values ​​are generated based on energy response sensitivity and regional response differences, and the initial control values ​​are dynamically corrected based on material batch information and historical calibration data. The laser is driven to begin marking along a preset path based on the initial control value.

5. The adaptive control method for laser marking on a terminal block housing according to claim 4, characterized in that, Within the adjustment boundary, an initial control value is generated based on energy response sensitivity and regional response differences. This initial control value is then dynamically corrected based on material batch information and historical calibration data, as detailed below: Obtain the current shell's material batch information and historical marking data; Statistical analysis of historical marking data is performed to extract the deviation trend and drift characteristics of different batches of materials in response to laser data, and the deviation range of the current batch is obtained. Based on the surface feature image of the shell and the energy response sensitivity, the marked area is divided, and the response differences of different areas to changes in laser data are identified, i.e., the regional response differences. The baseline laser data of historical marking data is obtained, and the baseline laser data is corrected by combining the batch deviation range and regional response differences to generate initial control values; The initial control value of each sensitive area is controlled to ensure that the initial marking of each sensitive area meets the expected requirements; The initial control values, after regional correction, are matched with the safe operating range of the laser equipment and the material tolerance limit. Control values ​​that exceed the limits are constrained to obtain the final control values. Based on the corrected control values, the laser is driven to begin marking along a preset path.

6. The adaptive control method for laser marking on a terminal block housing according to claim 1, characterized in that, The marking information of the collected marking area is then fused to generate comprehensive marking status data, as detailed below: Collect marking process data in the marking area and correlate the marking process data with the laser scanning path and time axis; Based on the laser scanning timing sequence, time synchronization calibration is performed on marking process data from different sources, and data with inconsistent sampling frequencies and acquisition delays are resampled. Using the spatial coordinates of the marking area as a reference, spatial mapping is performed on the marking process data after time synchronization calibration to achieve spatial alignment, and noise suppression processing is performed on the aligned marking process data; Trend constraints are applied to the calibration process data of reflected light intensity change and temperature rise trend, and abnormal sampling points that deviate from the current calibration state are marked. Based on the marking process data after noise suppression processing, process feature data is extracted; The process feature data is associated with the location and time node of the marking path and stored in a structured manner to form a spatiotemporal feature sequence of the marking process; Extract process feature data corresponding to the current marking segment from the spatiotemporal feature sequence of the marking process, normalize and uniformly scale the extracted process feature data, and assign weights according to their intensity of marking quality representation to construct a comprehensive marking state vector.

7. The adaptive control method for laser marking on a terminal block housing according to claim 1, characterized in that, The process involves outputting real-time marking quality error and its rate of change based on comprehensive marking status data, and obtaining the adjustment direction and amplitude of laser data under preset physical constraints, as detailed below: Determine the reference state for marking quality based on the expected requirements of the marking task; The real-time marking quality error is calculated by comparing the comprehensive marking status data acquired at the current moment with the marking quality reference status. A time-series analysis of the real-time marking quality error is performed at continuous marking times, and the rate of change of the marking quality error as the marking path progresses is output. Extract state features of the marking process evolution trend from comprehensive marking state data; Based on the real-time marking quality error, the rate of change of marking quality error, and state characteristics, a comprehensive judgment result of marking quality deviation is generated to distinguish between transient fluctuations and persistent deviations. Based on the comprehensive judgment results and the mapping relationship between the laser input and the changing trend of the shell surface, the direction of laser data adjustment is determined. Based on the determined laser data adjustment direction, and combined with the energy response range of the current shell and the corresponding laser energy density adjustment boundary, the adjustment range of the laser data is output. Based on preset physical constraints, the adjustment direction and adjustment range of the laser data are restricted; Under the premise of satisfying the physical constraints, the adjustment direction and amplitude of the output laser data are determined.

8. The adaptive control method for laser marking on a terminal block housing according to claim 7, characterized in that, The process involves generating a comprehensive judgment result for marking quality deviation based on real-time marking quality error, the rate of change of marking quality error, and state characteristics, distinguishing between transient fluctuations and persistent deviations, as detailed below: Based on the real-time marking quality error, the current marking quality deviation is compared with the preset allowable deviation range. When the real-time marking quality error is within the allowable deviation range, the current deviation is determined to be acceptable. When the real-time marking quality error exceeds the allowable deviation range, the trend is determined by combining the rate of change of the marking quality error. When the marking quality error increases in the same direction or does not show a significant decline at multiple consecutive marking times, the deviation is determined to have a continuous development trend. When the real-time marking quality error exceeds the allowable deviation range, but the rate of change of the marking quality error shows a rapid decline, the deviation is judged to be a transient fluctuation. Extract the direction of change of state features related to the marking quality deviation from the comprehensive marking state data; Determine whether the direction of change of state characteristics is consistent with the direction of change of real-time marking quality error. If the direction of change is consistent, confirm that the deviation originates from the actual change of state during the marking process. When the real-time marking quality error exceeds the allowable deviation range, the rate of change of marking quality error shows a continuous trend, and the direction of change of state characteristics is consistent with the direction of deviation, the current marking quality deviation is determined to be a continuous deviation. If the above-mentioned persistent deviation determination conditions are not met, the current marking quality deviation is determined to be a transient fluctuation deviation.

9. The adaptive control method for laser marking on a terminal block housing according to claim 1, characterized in that, The process involves adjusting the direction and amplitude of the laser data, combining it with historical marking data to dynamically correct the physical constraints, and outputting the optimal combination of laser data, as detailed below: Retrieve historical marking data corresponding to the current shell material type, material batch, and surface condition; Based on historical marking data, the effective boundaries of the preset physical constraints are corrected; The revised physical constraints are used as the new constraint boundaries and matched with the current laser data adjustment direction and adjustment range to determine whether there is a risk of over-marking, material damage or unstable marking. When the adjusted laser data falls within the corrected physical constraint boundary, the adjustment direction and adjustment range of the laser data are confirmed to be effective. When the adjusted laser data exceeds the corrected physical constraint boundary, the laser data adjustment range is redistributed according to the safe adjustment path of the corresponding working condition in the historical calibration data, and the suboptimal adjustment direction is output. The corrected laser data is combined and optimized to generate a laser data combination that meets the marking quality requirements and is within the corrected physical constraints. The combined laser data output is taken as the optimal laser data combination for the current marking stage.

10. A system using an adaptive control method for laser marking of a terminal housing as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, an initial control value module, a fusion module, a constraint module, and an output module, and the modules are interconnected. The data acquisition module is used to acquire surface image information of the target terminal housing, extract initial surface state data, establish the mapping relationship between laser energy density and housing surface changes, and construct an initial marking control model. The initial control value module is used to set the initial control value according to the initial marking control model and drive the laser to perform marking. The fusion module is used to collect the marking information of the marking area during the marking process, and perform fusion processing to generate comprehensive marking status data; The constraint module is used to output the real-time marking quality error and its rate of change based on the comprehensive marking status data, and to obtain the adjustment direction and amplitude of the laser data under preset physical constraint conditions; The output module is used to dynamically correct the physical constraints based on the adjustment direction and amplitude of the laser data and historical marking data, and output the optimal combination of laser data.

Images