Intelligent control method and system for cutting of blue laser

Abstract

The invention discloses an intelligent cutting control method and system for blue laser, and the method comprises the steps: carrying out the time domain analysis of the power time sequence data of a blue laser in real time, and obtaining the power fluctuation data; matching the power fluctuation data with a preset defect feature model of the blue laser to determine the defect occurrence probability of the blue laser; under the condition that the defect occurrence probability is determined to be greater than a preset probability threshold value, according to the cutting parameters and the material parameters, determining an initial power compensation parameter and a kerf quality optimization parameter through a compensation parameter calculation model; performing multi-target iterative optimization on the initial power compensation parameter, the kerf quality optimization parameter and the power fluctuation data through a self-adaptive deviation adjustment algorithm to generate a dynamic power control signal; and adjusting the output power and the beam focusing position of the blue laser according to the dynamic power control signal so as to correct the cutting path of the blue laser in real time.

Description

Technical Field

[0001] The present application relates to the field of laser technology, and more specifically, to a blue laser cutting intelligent control method and system. Background Art

[0002] As a core process of modern manufacturing, laser cutting technology is of irreplaceable importance in fields such as aerospace, automobile manufacturing, and electronic devices due to its high precision, high efficiency, and wide applicability. Blue laser, due to its short wavelength and high energy density, can achieve finer cutting effects, making it the first choice for high-precision processing. However, existing laser cutting control methods have significant limitations in dealing with real-time fluctuations and instabilities in laser output power. Traditional control systems mostly rely on static calibration or manual adjustment, which makes it difficult to adapt to power deviations in dynamic processing environments, resulting in reduced cutting quality or material waste. In addition, existing methods lack the ability to adaptively optimize parameters when dealing with different material properties and cutting requirements, which limits the improvement of the intelligence level of existing cutting technologies.

[0003] Regarding related technologies, existing laser cutting control methods have significant limitations in dealing with real-time fluctuations and instabilities in laser output power, and lack the ability to adaptively optimize parameters, resulting in problems such as reduced cutting quality, material waste, and limited intelligence levels. No effective solutions have yet been proposed. Summary of the Invention

[0004] The embodiments of the present application provide a blue laser cutting intelligent control method and system to at least solve the problems in the related art, in which the existing laser cutting control method has significant limitations in dealing with the real-time fluctuation and instability of the laser output power, and lacks the ability to adaptively optimize parameters, resulting in reduced cutting quality, material waste and limited intelligence level.

[0005] According to one embodiment of the present application, a blue laser cutting intelligent control method is provided, comprising: performing real-time time-domain analysis on power timing data of a blue laser to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting blue laser light, and the power fluctuation data is used to indicate power fluctuations of the blue laser; matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; when it is determined that the defect occurrence probability is greater than a preset probability threshold, obtaining cutting parameters corresponding to a current cutting task from a preset processing parameter database and obtaining material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; determining initial power compensation parameters and kerf quality optimization parameters using a compensation parameter calculation model based on the cutting parameters and the material parameters; performing multi-objective iterative optimization on the initial power compensation parameters, the kerf quality optimization parameters, and the power fluctuation data using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; and adjusting the output power and beam focus position of the blue laser based on the dynamic power control signal to correct the cutting path of the blue laser in real time.

[0006] According to another embodiment of the embodiment of the present application, a blue laser cutting intelligent control system is also provided, including: a time domain analysis module, which is used to perform time domain analysis on the power timing data of the blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut the target material by outputting blue laser, and the power fluctuation data is used to indicate the power fluctuation of the blue laser; a matching module, which is used to match the power fluctuation data with a preset defect feature model of the blue laser to determine the defect occurrence probability of the blue laser; an acquisition module, which is used to obtain the current cutting power from a preset processing parameter database when it is determined that the defect occurrence probability is greater than a preset probability threshold. The invention relates to a method for obtaining cutting parameters corresponding to a task, and material parameters of the target material obtained from a material parameter database, wherein the current cutting task is used to cut the target material; a determination module, used to determine, according to the cutting parameters and the material parameters, an initial power compensation parameter and a slit quality optimization parameter through a compensation parameter calculation model; an optimization module, used to perform multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter and the power fluctuation data through an adaptive deviation adjustment algorithm to generate a dynamic power control signal; an adjustment module, used to adjust the output power and beam focusing position of the blue light laser according to the dynamic power control signal to correct the cutting path of the blue light laser in real time.

[0007] According to another aspect of an embodiment of the present application, a computer-readable storage medium is further provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned blue light laser cutting intelligent control method when running.

[0008] According to another aspect of the embodiment of the present application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the processor executes the above-mentioned blue light laser cutting intelligent control method through the computer program.

[0009] In an embodiment of the present application, during the operation of the blue laser, the power timing data of the blue laser is subjected to time series analysis in real time to obtain power fluctuation data, wherein the blue laser is used to perform laser cutting on the target material by outputting the blue laser, and the power fluctuation data indicates the power fluctuation of the blue laser; the power fluctuation data is matched with a preset defect characteristic model to determine the defect occurrence probability of the blue laser; if the defect occurrence probability is greater than a preset probability threshold, the cutting parameters corresponding to the current cutting task are obtained from a preset processing parameter database, and the material parameters of the target material are obtained from a material parameter database, wherein the current cutting task is used to cut the target material; based on the cutting parameters and the material parameters, the initial power compensation parameters and the slit quality optimization parameters are determined by a compensation parameter calculation model; then, the initial power compensation parameters, the slit quality optimization parameters and the power fluctuation data are subjected to multi-objective iterative optimization through an adaptive deviation adjustment algorithm to generate a dynamic dynamic power control signal; finally, the output power and beam focus position of the blue laser are adjusted in real time according to the dynamic power control signal to correct the cutting path of the blue laser in real time; by adopting the above scheme, by real-time monitoring of the power fluctuation data of the blue laser and matching it with the preset defect characteristic model, it is possible to predict in advance the defects that may occur in the cutting process, so that when the probability of defect occurrence exceeds the threshold, the output power and beam focus position of the laser are dynamically adjusted; this intelligent control method not only improves the cutting quality and reduces material waste, but also enhances the adaptive ability of the system, so that it can automatically optimize the cutting parameters according to different materials and process conditions, significantly improving the intelligence level and efficiency of laser cutting; thereby solving the problem in the related art that the existing laser cutting control method has significant limitations in dealing with the real-time fluctuation and instability of the laser output power, and lacks the ability to optimize adaptive parameters, resulting in reduced cutting quality, material waste and limited intelligence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 This is a hardware structure block diagram of a camera device for an optional blue laser cutting intelligent control method according to an embodiment of the present application; Figure 2 This is a flow chart of an optional blue laser cutting intelligent control method according to an embodiment of the present application; Figure 3 This is a structural block diagram of a blue laser cutting intelligent control system according to an embodiment of the present application. DETAILED DESCRIPTION

[0011] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] The method embodiments provided in the embodiments of the present application can be executed on a blue laser or similar computing system. Taking running on a blue laser as an example, Figure 1 This is a hardware block diagram of a blue laser cutting intelligent control method of the embodiment of the present application. Figure 1 As shown, the blue laser may include one or more ( Figure 1 Only one is shown in the figure) processor 102 (processor 102 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing systems) and a memory 104 for storing data. In an exemplary embodiment, the above-mentioned blue laser may also include a transmission device 106 for communication functions and an input and output device 108. It will be understood by those skilled in the art that Figure 1The structure shown is only for illustration and does not limit the structure of the above-mentioned blue laser. Figure 1 More or fewer components than shown, or with Figure 1 Equivalent functions or comparisons shown Figure 1 Shown are different configurations with more functionality.

[0014] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the blue laser cutting intelligent control method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories can be connected to a secure text via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0015] Transmission system 106 is configured to receive or transmit data via a network. A specific example of such a network may include a wireless network provided by a communications provider of the blue laser. In one embodiment, transmission system 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet.

[0016] In this embodiment, a blue laser cutting intelligent control method is provided, which is applied to a blue laser. Figure 2 : is a flow chart of an optional blue laser cutting intelligent control method according to an embodiment of the present application, the process includes the following steps: Step S202: performing time domain analysis on power timing data of a blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuation of the blue laser; Step S204, matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; Step S206: if it is determined that the probability of the defect occurring is greater than a preset probability threshold, obtaining cutting parameters corresponding to the current cutting task from a preset processing parameter database, and obtaining material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; Step S208, determining initial power compensation parameters and kerf quality optimization parameters through a compensation parameter calculation model according to the cutting parameters and the material parameters; Step S210, performing multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; Step S212 , adjusting the output power and beam focus position of the blue laser according to the dynamic power control signal, so as to correct the cutting path of the blue laser in real time.

[0017] Through the above steps, the power timing data of the blue laser is analyzed in real time during the operation of the blue laser to obtain power fluctuation data, wherein the blue laser is used to laser cut the target material by outputting the blue laser, and the power fluctuation data indicates the power fluctuation of the blue laser; the power fluctuation data is matched with the preset defect characteristic model to determine the defect occurrence probability of the blue laser; if the defect occurrence probability is greater than the preset probability threshold, the cutting parameters corresponding to the current cutting task are obtained from the preset processing parameter database, and the material parameters of the target material are obtained from the material parameter database, wherein the current cutting task is used to cut the target material; according to the cutting parameters and the material parameters, the initial power compensation parameters and the cutting seam quality optimization parameters are determined by the compensation parameter calculation model; then the initial power compensation parameters, the cutting seam quality optimization parameters and the power fluctuation data are iteratively optimized with multiple objectives by the adaptive deviation adjustment algorithm to generate a dynamic power control signal; finally, the output power and beam focusing position of the blue laser are adjusted in real time according to the dynamic power control signal to correct the cutting path of the blue laser in real time; by adopting the above scheme, by real-time monitoring of the power fluctuation data of the blue laser and matching it with the preset defect characteristic model, it is possible to predict in advance the defects that may occur in the cutting process, so that when the probability of defect occurrence exceeds the threshold, the output power and beam focusing position of the laser are dynamically adjusted; through this intelligent control method, not only the cutting quality is improved and material waste is reduced, but also the adaptive ability of the system is enhanced, so that it can automatically optimize the cutting parameters according to different materials and process conditions, significantly improving the intelligence level and efficiency of laser cutting; thereby solving the problem in the related technology that the existing laser cutting control method has significant limitations in dealing with the real-time fluctuation and instability of the laser output power, and lacks the ability to optimize adaptive parameters, resulting in reduced cutting quality, material waste and limited intelligence level.

[0018] Optionally, after adjusting the output power and beam focusing position of the blue light laser according to the dynamic power control signal to correct the cutting path of the blue light laser in real time, the method further includes: performing feature extraction on the cutting path by a feature extraction device to obtain the cutting quality characteristics of the cutting path, and performing a matching degree analysis on the cutting quality characteristics by the preset defect feature model to determine whether the cutting path passes the quality verification; if it is determined that the cutting path fails the quality verification, updating the weight coefficient of the compensation parameter calculation model according to the cutting quality characteristics and the material parameters to obtain an updated compensation parameter calculation model; and re-correcting the cutting path by using the updated compensation parameter calculation model.

[0019] Optionally, an embodiment of the present application provides an optional cutting path quality verification method, the method comprising the following steps: S401, obtaining kerf width data from cutting path data; S402, measuring the slit width using an image processing tool, calculating uniformity features using a uniformity feature extraction tool, and obtaining uniformity feature data; S403, extracting ablation depth parameters using a depth measurement tool based on the uniformity characteristic data and the surface ablation data, calculating flatness parameters using a flatness analysis tool, and determining defect characteristic data; S404, determining whether the defect feature data matches a preset threshold; S405: If the defect feature data does not match the preset threshold, the defect feature data is compared using a matching analysis tool, a matching score is calculated, and a matching analysis result is determined; S406 , based on the matching analysis result, a quality verification tool is used to comprehensively evaluate the kerf width, ablation depth, and flatness parameters to obtain a quality verification judgment result.

[0020] Specifically, during the laser cutting quality inspection process, the system can collect the slit edge image through a high-resolution linear array CCD, extract the edge using the Canny operator, fit the contour lines on both sides of the slit based on the Hough transform, and calculate the standard deviation of the slit width.

[0021] For example, when the kerf width fluctuations exceeding ±10μm on a certain path are detected, the system triggers an adaptive threshold segmentation algorithm, combining the material's reflectivity (e.g., 304 stainless steel has a reflectivity of 60%) to correct the measurement error, ensuring that the width uniformity determination error is less than 3μm. For surface ablation depth detection, the system integrates a confocal displacement sensor, which scans the cut surface at a 50kHz sampling frequency. After eliminating noise through Gaussian filtering, the least squares method is used to fit the ablation pit profile curve. When the depth deviation exceeds 15% of the nominal value (e.g., the maximum allowable ablation depth for a 1.2mm plate is 180μm), it is automatically marked as a defective area. The flatness analysis module uses phase measurement deflectometry, utilizing a 12MP industrial camera to capture surface light fringe distortion. The phase distribution is then calculated using a fast Fourier transform. When the local curvature radius is less than 500mm, the flatness is determined to be excessive. The defect signature model utilizes a support vector machine classifier. Input parameters include the coefficient of variation of kerf width (threshold 0.08), ablation depth gradient (threshold 25μm / mm), and the root mean square value of flatness (threshold 40μm). A quality alarm is generated when all three indicators exceed their thresholds. For carbon steel, the system also incorporates oxidation color analysis, extracting kerf edge color values ​​using the HSV color space. An overburn defect is identified when the H component exceeds the range of 160-200.

[0022] Quality characteristic data is obtained from the cutting path data and combined with thermal conductivity data, absorptivity data, and surface reflectivity data. Comprehensive characteristic data is generated using a data fusion tool to obtain comprehensive characteristic data. Based on this comprehensive characteristic data (equivalent to the aforementioned cutting quality characteristics), a logistic regression algorithm is used to calculate weight coefficients. If the comprehensive characteristic data does not match the preset quality standards (i.e., the cutting path fails quality verification), the weight coefficients are adjusted using the parameter optimization tool to determine updated weight coefficients. Using the updated weight coefficients, the compensation parameter calculation tool is used to recalculate the compensation parameters in combination with the thermal conductivity data and absorptivity data to obtain optimized compensation parameters. Based on the optimized compensation parameters, the quality verification tool is used to re-evaluate the quality characteristics of the cutting path. If the quality characteristics meet the preset quality standards, the quality verification is considered passed.

[0023] Specifically, if quality verification fails during the laser cutting process, the system dynamically updates the weight coefficients of the compensation parameter calculation model based on quality characteristics such as kerf width uniformity, surface ablation depth, and flatness, combined with the material's thermal conductivity (e.g., aluminum alloy has a thermal conductivity of 237 W / m·K), absorptivity (e.g., copper has an absorptivity of 0.3), and surface reflectivity (e.g., titanium alloy has a reflectivity of 70%). The system uses a gradient descent algorithm to optimize the weight coefficients, with an initial learning rate of 0.01 and 100 iterations to ensure model convergence. The updated compensation parameters are then used to adjust the laser power and cutting speed via a PID controller. The laser power adjustment range is 500 W to 3000 W, and the cutting speed adjustment range is 10 mm / s to 50 mm / s. The system monitors the temperature distribution during the cutting process in real time and simulates the heat conduction process using finite element analysis. If the local temperature gradient exceeds 50°C / mm, the laser focus position is automatically adjusted with an accuracy of ±0.1 mm. The system also uses a spectrum analyzer to monitor the plasma emission spectrum during the cutting process. If an abnormal intensity is detected at a specific wavelength (e.g., 532nm), an adaptive filtering algorithm is triggered with a filter bandwidth set to 10nm to ensure spectral data accuracy. Ultimately, the system uses a multi-objective optimization algorithm (such as NSGA-II) to balance cutting quality and efficiency. Optimization targets include kerf width uniformity (target value ±5μm), surface ablation depth (target value ±10μm), and flatness (target value ±20μm), until the quality characteristics of the cutting path meet preset standards.

[0024] Optionally, a time domain analysis is performed on the power timing data of the blue light laser in real time to obtain power fluctuation data, including: acquiring the power timing data from the blue light laser in real time through a high-speed power meter, and performing frequency domain conversion on the power timing data through fast Fourier transform to obtain frequency domain distribution data; performing signal processing on the frequency domain distribution data through a bandpass filter to extract the power fluctuation signal of the target frequency band to obtain a fluctuation signal set; determining the fluctuation amplitude corresponding to the fluctuation signal set, and performing time domain analysis on the fluctuation signal set through a period estimation algorithm to calculate the period and phase of the fluctuation signal set, and determining the fluctuation frequency and phase offset corresponding to the fluctuation signal set based on the period and the phase to obtain the power fluctuation data, wherein the power fluctuation data includes: the fluctuation amplitude, the fluctuation frequency and the phase offset.

[0025] Optionally, determine the fluctuation amplitude corresponding to the fluctuation signal set, and perform time domain analysis on the fluctuation signal set through a period estimation algorithm, including: when the fluctuation amplitude is less than or equal to a preset threshold, perform time domain analysis on the fluctuation signal set through the period estimation algorithm; when the fluctuation amplitude is greater than the preset threshold, perform data preprocessing on the fluctuation signal set through adaptive mean filtering to obtain a smoothed fluctuation signal set, and perform time domain analysis on the fluctuation signal set belonging to the smoothed fluctuation signal set through the period estimation algorithm.

[0026] Real-time power time series data is acquired from the laser using a high-speed power meter. Fast Fourier transform (FFT) is used to convert this time series data into the frequency domain, generating frequency domain distribution data. Based on this frequency domain distribution data, a bandpass filter is used for signal processing to extract power fluctuation signals within a specific frequency band, generating a set of fluctuation signals. If the amplitude of this set of fluctuation signals exceeds a preset threshold, the set of fluctuation signals is preprocessed using adaptive mean filtering to obtain a smoothed fluctuation signal and determine the fluctuation amplitude. A period estimation algorithm is then used to perform time domain analysis on this smoothed fluctuation signal, calculating its period and phase to determine the fluctuation frequency and phase offset.

[0027] In one possible implementation, a high-speed power meter is used to obtain real-time power timing data from a laser. The principle is to capture the dynamic changes of the laser output power through high-frequency sampling.

[0028] For example, assuming the laser operates in continuous-wave mode, a power meter records power data at a 100kHz sampling rate, generating a set of time-series data reflecting power variations over time. This high-frequency sampling can capture microsecond-level power fluctuations, providing a reliable foundation for subsequent frequency-domain analysis. After data acquisition, a fast Fourier transform converts the time-series data into frequency-domain distribution data, revealing the frequency components of the power fluctuations.

[0029] For example, after transformation, it is found that the main frequency components are concentrated in the range of 1kHz to 5kHz, indicating that the laser may be disturbed by a specific frequency.

[0030] Specifically, a bandpass filter is used to extract the power fluctuation signal in a specific frequency band.

[0031] Preferably, a bandpass filter with a center frequency of 3kHz and a bandwidth of 500Hz is designed to filter out irrelevant low-frequency drift and high-frequency noise, retaining only the target frequency band signal. After processing, a set of fluctuation signals is obtained, reflecting the dynamic characteristics of the laser in this frequency band.

[0032] It should be noted that if the amplitude of the fluctuation signal set exceeds a preset threshold, such as 0.5 mW, subsequent processing is triggered.

[0033] In one embodiment, it is detected that the amplitude of a certain signal reaches 0.8 mW, indicating that there is a significant disturbance and further analysis is required.

[0034] For example, adaptive mean filtering is used to smooth fluctuating signal collections and reduce the impact of random noise. In implementation, a five-point sliding window is used to dynamically adjust the filtering parameters, ensuring that noise is reduced while preserving signal characteristics. After smoothing, the fluctuation amplitude is reduced from 0.8mW to 0.6mW, making the signal clearer and facilitating subsequent analysis. This smoothing process improves signal quality and helps accurately identify fluctuation characteristics.

[0035] In one embodiment, the period estimation algorithm performs time domain analysis on the smoothed fluctuation signal to calculate the period and phase.

[0036] For example, autocorrelation analysis revealed a signal period of approximately 0.33 ms, corresponding to a frequency of approximately 3 kHz, which coincides with the center frequency of the bandpass filter. Phase offset analysis revealed a signal lag of approximately 30 degrees relative to the reference signal, indicating a potential delay effect from an external disturbance source. This period and phase information can help locate the source of disturbances, such as mechanical vibration or power supply fluctuations.

[0037] It can be understood that the combined application of the above methods can accurately characterize the power stability of the laser.

[0038] For example, the extracted fluctuation frequency and phase can be used to optimize laser control systems and reduce the impact of external interference. Smoothing and filtering improve data reliability and ensure the accuracy of analysis results. These technical approaches collectively improve the accuracy of laser performance monitoring and provide technical support for high-precision applications such as optical communications and laser processing.

[0039] Optionally, matching the power fluctuation data with a preset defect feature model of the blue laser to determine the defect occurrence probability of the blue laser includes: determining a plurality of power fluctuation data corresponding to a plurality of first sampling points of the power time series data, wherein the plurality of first sampling points correspond to a plurality of time windows of the power time series data, and the power fluctuation data include: fluctuation amplitude, fluctuation frequency, and phase offset; matching the plurality of power fluctuation data with a plurality of first defect feature models respectively, determining a plurality of second sampling points among the plurality of first sampling points that successfully match the first defect feature model, and determining a plurality of second defect feature models among the plurality of first defect feature models that successfully match the plurality of second sampling points, wherein the preset defect feature model includes the plurality of first defect feature models, the plurality of first defect feature models correspond one-to-one to a plurality of cutting defect types, and the plurality of second sampling points correspond one-to-one to the plurality of second defect feature models; determining preset defect thresholds corresponding to the plurality of second defect feature models respectively, and performing a corresponding operation on the fluctuation amplitude A. i (t) and amplitude defect threshold T A，i , the fluctuation frequency ω i (t) and frequency defect threshold T ω，i , the phase offset ϕ i (t) and the phase defect threshold Tϕ,i are normalized to obtain the normalized fluctuation amplitude and normalized amplitude defect threshold , the normalized fluctuation frequency and the normalized frequency defect threshold , the normalized phase offset and the normalized phase defect threshold According to the weight of the fluctuation amplitude, the fluctuation frequency and the phase offset, the formula Calculate the dynamic correlation coefficient γ i (t), where is the amplitude weight, is the frequency weight, is the phase weight, is the tolerance coefficient corresponding to the fluctuation frequency; through the formula Calculate the defect occurrence probability corresponding to each cutting defect type under the power time series data, where P(defect i ) is the probability of occurrence of the i-th type of cutting defect among the multiple cutting defect types, N is the number of the multiple first sampling points, δ ik It is used to indicate whether the kth first sampling point among the multiple first sampling points has the i-th type of cutting defect, and N and k are both positive integers.

[0040] The fluctuation amplitude, frequency and phase offset in the power fluctuation data are extracted and matched with the corresponding ranges in the preset defect feature model one by one; for the defect feature model that matches successfully, the corresponding preset defect threshold is determined, and the fluctuation amplitude, fluctuation frequency, phase offset and the corresponding preset threshold are normalized to the same dimension (0-1 range), and according to the formula Calculate the dynamic correlation coefficient between the actual fluctuation amplitude and the preset defect threshold, where γ(t) represents the dynamic correlation coefficient, is the amplitude weight, is the frequency weight, is the phase weight, is the tolerance coefficient corresponding to the fluctuation frequency, and the weights must satisfy that the sum of the three is equal to 1; according to the dynamic correlation coefficient, combined with the matching degree, according to the formula Calculate the defect occurrence probability of each cutting defect type under the current power fluctuation data, P(defect i ) represents the probability of occurrence of the i-th type of cutting defect, N represents the total number of power fluctuation data sampling points (i.e., the number of the above-mentioned multiple first sampling points), δ ik Indicates whether the i-th type of cutting defect occurs at the k-th sampling point (1 indicates occurrence, 0 indicates non-occurrence).

[0041] For example, in analyzing laser power fluctuation data, the fluctuation amplitude, frequency, and phase offset are extracted and matched against a pre-set defect signature model to identify potential defects in laser cutting. The fluctuation amplitude reflects the intensity of the power variation, the frequency indicates the period of the disturbance, and the phase offset reveals deviations in the signal timing. For example, if the laser is used to cut thin metal sheets, power fluctuations may result in rough cut surfaces or burns. By matching the fluctuations with the defect signature model, it can be determined whether the fluctuations correspond to specific cutting defects.

[0042] For example, the model defines that fluctuations with an amplitude exceeding 1 mW, a frequency between 2 and 4 kHz, and a phase shift of 30 to 60 degrees are associated with cutting burns.

[0043] In one possible implementation, after extracting power fluctuation data, the model parameters are compared one by one. Suppose real-time data shows a fluctuation amplitude of 1.2 mW, a frequency of 3 kHz, and a phase shift of 45 degrees. Comparing this data with the model reveals a high degree of agreement with the burn defect characteristics.

[0044] It should be noted that the matching process must take into account parameter tolerances, such as a ±0.5kHz frequency deviation, to ensure robustness. After a successful match, the preset threshold for the corresponding defect is extracted, such as the amplitude threshold for a burn defect is 1mW.

[0045] Specifically, the dynamic correlation coefficient between the actual fluctuation amplitude and the preset threshold is calculated to reflect the degree of correlation between the two.

[0046] Preferably, a normalization method is used to quantify the difference between the actual amplitude 1.2 mW and the threshold 1 mW, and a coefficient is generated in combination with the matching degree of frequency and phase.

[0047] For example, the amplitude exceeds the threshold by 20%, the frequency deviation is less than 5%, and the phase deviation is within the tolerance. The comprehensive calculation results in a correlation coefficient of 0.85, indicating a strong correlation.

[0048] In one embodiment, the probability of a defect occurring is calculated based on the dynamic correlation coefficient and the degree of matching. Assuming the weight factor for a burn defect is 0.7, determined based on historical data, combined with a coefficient of 0.85 and a degree of matching of 90%, the probability of a burn defect is calculated to be 63%. Meanwhile, considering other defects such as incomplete cutting, which are characterized by low-frequency fluctuations and small amplitudes, the probability after matching is only 15%.

[0049] Understandably, a high probability of defects suggests that it is necessary to prioritize adjusting laser parameters, such as reducing power or optimizing modulation frequency.

[0050] For example, for burn defects, the source of fluctuations can be further analyzed. Assuming that the frequency of 3kHz is related to the vibration of the cooling system, after adjusting the cooling frequency, the fluctuation amplitude is reduced to 0.8mW, and the probability is reduced to 20%. This multi-faceted analysis, from parameter matching to probability calculation, and then to defect tracing, forms a complete logical chain to ensure that the analysis is accurate and operational. In another embodiment, the expansion solution can introduce an adaptive model to dynamically update the defect feature library. Assuming that the new data shows that the 1.5mW fluctuation is related to the new material, the model automatically adjusts the threshold to improve the matching accuracy. This method ensures the comprehensiveness and flexibility of the solution through core matching and expansion updates.

[0051] It should be noted that the defect model proposed in this application is a pre-defined mathematical model used to describe the relationship between different defect types and laser power fluctuation characteristics. It is constructed based on historical data, experimental results, and material properties to predict the type and probability of cutting defects that may occur under specific power fluctuation characteristics.

[0052] The defect feature model usually includes the following: Defect Type: Describes the type of cutting defects that may occur, such as burns, uneven cuts, incomplete cuts, etc.

[0053] Characteristic range: The power fluctuation characteristic range corresponding to each defect type, including fluctuation amplitude, frequency, and phase offset.

[0054] Preset defect threshold: The critical value used to determine whether power fluctuations may cause defects.

[0055] Through this defect characteristic model, the present application can match the power fluctuation data monitored in real time with the defect characteristic model, and the system can predict the types of defects that may occur under the current cutting conditions and their probabilities; thereby dynamically adjusting the output power and cutting parameters of the laser according to the prediction results to reduce or avoid the occurrence of defects.

[0056] Optionally, according to the cutting parameters and the material parameters, initial power compensation parameters and slit quality optimization parameters are determined through a compensation parameter calculation model, including: determining the energy input efficiency of the blue light laser according to the absorptivity and thermal conductivity in the material parameters, wherein the absorptivity is used to indicate the absorption rate of the target material to the blue light laser, the energy input efficiency is used to indicate the efficiency of converting the blue light laser into cutting energy, and the thermal conductivity is used to indicate the thermal conductivity of the target material; determining the thermal diffusion compensation amount of the blue light laser according to the thermal conductivity and the cutting depth, cutting width and cutting speed in the cutting parameters, wherein the thermal diffusion compensation amount is used to compensate for the energy loss caused by thermal conduction of the target material; and determining the initial power compensation parameter according to the energy input efficiency and the thermal diffusion compensation amount.

[0057] The cutting speed v, depth d, cutting width w and focus offset parameter F o Input this into the compensation parameter calculation model. Using the preset calculation formula in the compensation parameter calculation model, the energy input efficiency and thermal diffusion compensation amount are calculated by combining the thermal conductivity k and the absorptivity a. The power compensation parameter is calculated using the preset calculation formula by combining the energy input efficiency with the thermal diffusion compensation amount.

[0058] For example, during the laser cutting process, the cutting speed v, depth d, cutting width w and focus offset parameter F are set to o When inputting into the compensation parameter calculation model, the accuracy of the parameters must be ensured to support subsequent calculations. The cutting speed refers to the movement rate of the laser head along the material surface, the depth indicates the vertical distance the cutting enters the material, the width is the lateral dimension of the slit, and the focus offset describes the positional deviation of the laser focus relative to the material surface. These parameters jointly determine the energy distribution and cutting effect. Assume that in a certain stainless steel sheet cutting task, the cutting speed is 2m / s, the depth is 0.5mm, the width is 0.2mm, and the focus offset is 0.1mm. These values ​​are extracted from the database through the task identifier to ensure consistency with the actual processing requirements. After entering the model, the system performs energy distribution analysis based on these parameters, laying the foundation for subsequent compensation calculations.

[0059] In one possible implementation, the compensation parameter calculation model uses a preset formula to combine thermal conductivity k and absorptivity a to calculate energy input efficiency and thermal diffusion compensation. Thermal conductivity reflects the material's ability to conduct heat, while absorptivity indicates the proportion of laser energy absorbed by the material. Energy input efficiency measures the effectiveness of converting laser energy into cutting energy, while thermal diffusion compensation corrects for energy loss due to material thermal conductivity.

[0060] Specifically, the energy input efficiency η can be calculated by the following formula: , where w is the cutting width (unit: m), d is the cutting depth (unit: m), and f is the focus offset (unit: m). α is the empirical attenuation coefficient (unit: m), which is related to the scattering characteristics of the material to the laser and is usually calibrated experimentally (for example, α≈0.001 m for stainless steel).

[0061] The thermal diffusion compensation amount ΔP can be calculated using the following formula: , v is the cutting speed, β is the correction coefficient (dimensionless), which is related to the melting point of the material (for example, the higher the melting point, the smaller β, the empirical value β = 0.1∼0.5).

[0062] For example, for SS-304 stainless steel, the thermal conductivity is approximately 15 W / m·K, and the absorptivity is 35%. Model analysis shows that low absorptivity can lead to insufficient energy utilization, while high thermal conductivity causes rapid heat diffusion, affecting kerf quality. The system therefore calculates the need to increase energy input efficiency and adjusts power allocation by compensating for thermal diffusion.

[0063] Specifically, energy input efficiency is calculated based on the matching of absorptivity and laser power. Assuming an initial laser power of 2000W, the actual effective energy is low due to absorptivity of only 35%. The model analyzes this and recommends increasing the power to compensate for energy loss. It also uses thermal conductivity to assess the heat diffusion range to ensure that the kerf edge does not overheat. Thermal diffusion compensation is determined by simulating heat distribution to determine whether additional adjustments to cutting speed or focus shift are necessary.

[0064] For example, if heat diffusion causes edge burning, the model may recommend reducing the speed to 1.8 m / s to reduce heat accumulation.

[0065] Preferably, the calculation of the power compensation parameter integrates the energy input efficiency and the thermal diffusion compensation amount, aiming to optimize the laser output.

[0066] In one embodiment, based on the model output, the system recommends adjusting the power from 2000W to 2200W to balance the effects of insufficient absorption and heat diffusion. Simultaneously, the system checks whether fine-tuning to 0.05mm is necessary to focus the energy distribution, incorporating the focus offset parameter. This adjustment ensures a smooth kerf without an excessive heat-affected zone.

[0067] It is understandable that the above process relies on the synergistic analysis between parameters.

[0068] For example, the coordination of cutting speed and focus offset must be aligned with the material's thermal conductivity to avoid uneven energy distribution caused by excessive speed or focus deviation. For example, if uneven kerf width is detected during a task, the model can analyze the width parameter and absorptivity to infer that excessive focus offset may be the cause and provide optimization recommendations. This multi-parameter synergy improves the targeted nature of the calculations.

[0069] In one embodiment, the extended solution may introduce an adaptive adjustment module to dynamically update compensation parameters based on real-time cutting feedback.

[0070] For example, if a decrease in the kerf quality is detected, the system can recalculate the power compensation and apply it immediately. This flexibility ensures process stability and is particularly suitable for high-precision tasks.

[0071] Optionally, based on the cutting parameters and the material parameters, the initial power compensation parameters and the slit quality optimization parameters are determined through a compensation parameter calculation model, including: determining the laser energy utilization rate based on the absorptivity and surface reflectivity in the material parameters, wherein the laser energy utilization rate is used to indicate the proportion of laser energy absorbed by the target material and used for cutting; determining the beam focusing optimization coefficient based on the surface reflectivity and the focus offset parameter in the cutting parameters, wherein the beam focusing optimization coefficient is used to indicate the degree of energy concentration of the blue light laser at the focus; determining the slit quality optimization parameters based on the laser energy utilization rate, the beam focusing optimization coefficient and the cutting width in the cutting parameters.

[0072] The cutting width, focus offset parameter, absorptivity, and surface reflectivity are obtained. The laser energy utilization rate is calculated using a linear regression algorithm using a preset compensation parameter calculation model to obtain a laser energy utilization rate value. Based on the laser energy utilization rate value, the focus offset parameter and the beam quality are combined to obtain a beam focusing optimization coefficient using a beam focusing optimization formula. If the laser energy utilization rate value is lower than a preset threshold, the input weights of the compensation parameter calculation model are adjusted based on the laser power and material properties, and the laser energy utilization rate is recalculated to obtain an adjusted laser energy utilization rate value. The adjusted laser energy utilization rate value and the beam focusing optimization coefficient are then combined with the slit quality optimization formula to obtain a slit quality optimization parameter.

[0073] For example, in the laser cutting of stainless steel sheets, determining the cutting width, focus offset parameters, absorptivity, and surface reflectivity is fundamental to optimizing laser energy utilization. Cutting width refers to the lateral dimension of the kerf, focus offset describes the deviation of the laser focus from the material surface, absorptivity reflects the proportion of laser energy absorbed by the material, and surface reflectivity indicates the proportion of laser energy reflected by the material surface. These parameters directly influence energy distribution.

[0074] For example, SS-304 stainless steel has an absorptivity of 35%, a surface reflectivity of 60%, a cutting width of 0.2mm, and a focus offset of 0.1mm. Using a linear regression algorithm, the system analyzes historical data for these parameters and estimates the laser energy utilization, arriving at an initial value of 70%.

[0075] In one possible implementation, if the laser energy utilization rate is lower than a preset threshold of 80%, the input weight of the compensation parameter calculation model needs to be adjusted.

[0076] Specifically, considering the laser power of 2000W and the material properties, such as the high surface reflectivity of stainless steel, the system increased the absorptivity weighting, reduced the reflectivity effect, and recalculated the adjusted energy utilization rate to 75%. This adjustment ensures that energy distribution is more closely aligned with the material properties.

[0077] It should be noted that the calculation of the beam focusing optimization coefficient is based on the focus offset and beam quality. Beam quality describes the focusing ability of the laser beam. A focus offset of 0.1mm may cause energy dispersion.

[0078] For example, under high-quality beam quality, the system calculates the optimization coefficient as 0.85 through the beam focusing optimization formula combined with the surface reflectivity of 60%.

[0079] Preferably, if the focus shift is reduced to 0.05 mm, the coefficient can be increased to 0.9, indicating that the energy is more concentrated.

[0080] In one embodiment, the slit quality optimization parameter combines the adjusted energy utilization and beam focusing optimization coefficients.

[0081] For example, if the energy utilization rate is 75%, the cutting width is 0.2mm, and the optimization coefficient is 0.85, the kerf quality optimization formula yields a quality parameter of 0.8, reflecting the kerf smoothness and edge quality. Based on this, the system determines whether further adjustments to focus offset or power are needed.

[0082] It can be understood that the above process emphasizes parameter coordination.

[0083] For example, high surface reflectivity can lead to energy loss, which needs to be compensated by reducing focus offset or increasing power. In one expansion solution, the system incorporates a real-time monitoring module. If the kerf width is uneven, the system automatically adjusts the focus offset to 0.03mm, increasing energy utilization to 78% and optimizing kerf quality. This adaptive mechanism improves processing stability.

[0084] Specifically, for highly reflective materials, priority is given to optimizing beam quality and focal position to ensure energy concentration.

[0085] For example, if the cutting width increases to 0.25mm, the system analyzes the absorptivity and reflectivity and recommends fine-tuning the power to 2100W to maintain kerf consistency. This multi-parameter collaborative analysis improves processing accuracy and efficiency.

[0086] In an optional embodiment, the present application provides an optional blue laser cutting intelligent control method, the control method comprising the following steps: S301, acquiring the power time series data output by the laser in real time, and extracting the power fluctuation data including the fluctuation amplitude, frequency and phase offset through time domain analysis; S302, matching the power fluctuation data with a preset defect characteristic model, and calculating the probability of defect occurrence based on a dynamic correlation between the fluctuation amplitude and a preset defect threshold; S303, when the probability of the defect occurring exceeds a preset probability threshold, the current cutting speed, depth, cutting width, and focus offset parameters are retrieved from a preset processing parameter database, and the thermal conductivity, absorptivity, surface reflectivity, and melting point parameters of the target material are retrieved from a material parameter database; When the probability of a defect occurring exceeds a preset probability threshold, a database retrieval instruction is triggered; according to the retrieval instruction, the preset processing parameter database is first accessed, and in the database, based on the unique identifier of the current cutting task or the associated key information, the cutting speed, cutting depth, cutting width and focus offset parameters corresponding to the current cutting task are accurately located and retrieved. These parameters are pre-set and stored based on a large amount of historical cutting experience and current processing requirements; at the same time, the material parameter database is accessed, and also based on the unique identifier of the current cutting material or the corresponding key feature information, the thermal conductivity, absorptivity, surface reflectivity and melting point parameters corresponding to the target material are found in the material parameter database. These parameters are obtained through material performance testing or from material suppliers and pre-entered into the database, thereby completing the retrieval of all required parameters.

[0087] For example, during laser cutting, when the probability of a defect exceeds a preset threshold, a mechanism triggering a database callout can quickly respond to potential issues. The defect probability is typically calculated based on matching power fluctuations with defect characteristics. For example, in a sheet metal cutting task, if the probability of a burn defect reaches 65%, exceeding the preset threshold of 60%, the system automatically triggers a callout. The core of this mechanism is to ensure that when a high-risk defect is detected, relevant parameters are promptly retrieved to support subsequent adjustments and maintain cutting quality.

[0088] In one possible implementation, accessing the preset processing parameter database relies on a unique task identifier. The database stores core parameters related to cutting tasks, such as cutting speed, cutting depth, cutting width, and focus offset. For example, if the current task is identified as TC-001, the corresponding parameters are cutting speed 2m / s, cutting depth 0.5mm, cutting width 0.2mm, and focus offset 0.1mm. These parameters are pre-set based on historical cutting experience and processing requirements, and can be quickly located using the identifier when accessing them.

[0089] It should be noted that the design of the task identification needs to ensure uniqueness, for example, combining material type and processing batch to avoid parameter confusion.

[0090] Specifically, access to the material parameter database is equally critical. For the sheet metal currently being cut, assuming the material is SS-304, the database stores its thermal conductivity of 15 W / m·K, absorptivity of 35%, surface reflectivity of 60%, and melting point of 1370°C. These parameters are obtained through material testing or provided by the supplier to ensure accuracy.

[0091] In one embodiment, the system directly extracts parameters based on material identification and combines them with processing parameters to form a complete data set, providing a basis for subsequent analysis.

[0092] Preferably, the database supports dynamic updating, for example, when new materials are added, their parameters are entered in a timely manner.

[0093] It is understandable that the speed and accuracy of parameter retrieval are crucial to defect response.

[0094] For example, burn defects can be associated with high absorption rates and inappropriate cutting speeds. Using the retrieved parameters, the system can analyze whether it is necessary to reduce the speed or adjust the focus offset.

[0095] In one embodiment, thermal conductivity and melting point are combined to determine whether the current speed of 2 m / s causes heat accumulation, and then the speed is optimized to 1.8 m / s. This multi-parameter collaborative analysis ensures that the adjustment plan is targeted.

[0096] For example, regarding the focus offset parameter, excessive offset can lead to uneven laser energy distribution, increasing the risk of burns. If the retrieved data shows a focus offset of 0.1mm, while historical experience suggests 0.05mm is more suitable for SS-304 material, the system can make adjustment recommendations accordingly. This logical chain from parameter retrieval to analysis ensures the solution's operability. In another embodiment, the expanded solution can introduce a parameter recommendation module to automatically generate optimized parameter combinations based on the retrieved data and defect type, further improving efficiency.

[0097] S304, calculating energy input efficiency and thermal diffusion compensation amount using a compensation parameter calculation model based on the cutting speed, depth, cutting width, and focus offset parameters, combined with the thermal conductivity and absorptivity, to generate initial power compensation parameters; S305, based on the cutting width and focus offset parameters, combined with the absorptivity and surface reflectivity, calculating the laser energy utilization rate and the beam focusing optimization coefficient by using the compensation parameter calculation model to generate a slit quality optimization parameter; S306, using an adaptive deviation adjustment algorithm to perform multi-objective iterative optimization on the initial power compensation parameters, the slit quality optimization parameters, and the real-time power fluctuation characteristics to generate a dynamic power control signal; Real-time power data containing initial power and compensation parameters is received, and the fluctuation characteristics of the real-time power data are processed using an adaptive deviation adjustment algorithm to obtain adjusted power fluctuation data. Based on the adjusted power fluctuation data and the seam quality optimization parameters, a logical judgment method is used. If the real-time power is lower than a preset threshold, the compensation parameter is adjusted to determine a new compensation parameter value. The new compensation parameter value and the fluctuation characteristics are received, and an iterative optimization formula is used to calculate a dynamic power value. Based on the dynamic power value and the adaptive requirements, a signal generation tool is used to perform multi-objective adjustment on the power signal to obtain a dynamic power control signal.

[0098] Specifically, during the laser cutting process, the system uses an adaptive deviation adjustment algorithm to perform multi-objective iterative optimization on the initial power compensation parameters, cutting quality optimization parameters and real-time power fluctuation characteristics to generate a dynamic power control signal.

[0099] For example, the initial power compensation parameter is set to 2000W. The system analyzes the real-time power fluctuation characteristics based on the Gaussian distribution model and finds that the power fluctuation range is ±50W and the standard deviation is 15W. Combined with the seam quality optimization parameter 0.8, the system uses the particle swarm optimization algorithm to iteratively optimize the power compensation parameters. The objective function is to minimize power fluctuations and maximize seam quality. After 10 iterations, the system adjusts the power compensation parameter to 2050W, the power fluctuation range is reduced to ±30W, and the standard deviation is reduced to 10W. At the same time, the system uses the Kalman filter algorithm to predict real-time power fluctuations. Combined with historical data, it predicts the power fluctuation trend in the next 5 seconds and dynamically adjusts the power output.

[0100] For example, when predictions indicate that future power fluctuations could increase to ±40W, the system fine-tunes the power compensation parameters to 2070W in advance to ensure power stability. Furthermore, the system incorporates a multi-objective optimization model that comprehensively considers kerf quality, power utilization, and processing efficiency, generating the optimal power control signal using a genetic algorithm.

[0101] For example, when cutting 1.5mm thick carbon steel, the system calculated the optimal power at 2100W using a multi-objective optimization model, improving the kerf quality parameter to 0.85 and achieving a power utilization rate of 82%. This dynamic power control mechanism effectively improves machining accuracy and stability.

[0102] S307, driving the laser execution unit according to the dynamic power control signal, adjusting the output power and the beam focus position, and generating a real-time corrected cutting path; Dynamic power and control signals are acquired and transmitted to the laser unit via a signal driver, resulting in adjusted output power data. Based on the adjusted output power data and the beam focus position, the beam adjustment device calibrates the focus position and determines the calibrated focus position parameters. Using the calibrated focus position parameters and the path correction data, the path generation tool performs real-time corrections to the cutting path, generating corrected cutting path data. If the corrected cutting path data falls below a preset threshold, the power adjustment tool performs iterative optimization in conjunction with the output power data to obtain an optimized power control signal.

[0103] Specifically, when the system drives the laser execution unit based on the dynamic power control signal, it uses the Bayesian filtering algorithm to make real-time corrections to the beam focusing position and adjusts the spot diameter in combination with the material thickness change and thermal deformation coefficient.

[0104] For example, when cutting a 3mm stainless steel plate, the initial spot diameter is 0.2mm. The system detects through the heat conduction model that the local temperature rise of the material reaches 120°C, resulting in an increase of 0.15% in the thermal expansion coefficient. At this time, the gradient descent method is used to calculate the optimal focus offset, and the Z-axis position is lowered by 0.05mm to compensate for thermal deformation. At the same time, the path planning module reconstructs the cutting trajectory through the B-spline curve fitting algorithm. When the sensor detects a plate assembly deviation of 0.1mm, the system updates the interpolation point coordinates at a cycle of 20ms and generates a compensation path in the XY plane, so that the root mean square error between the actual cutting trajectory and the theoretical path is controlled within 5μm. To address the problem of overburning at corners, the system establishes a correlation model between the pulse duty cycle and the corner speed. When a 60° sharp angle is identified, the pulse frequency is automatically increased from 1000Hz to 1500Hz. The uniformity of the temperature field distribution is verified through finite element analysis to ensure that the energy density deviation in the corner area does not exceed 8%. In the multi-layer material cutting scenario, the system dynamically adjusts the power ramp rate based on the interlayer peeling signal strength fed back by the acoustic emission sensor. For example, when a 2.5MHz high-frequency component is detected, the power rise time is compressed from 50ms to 30ms, and combined with wavelet transform analysis, real-time judgment of the interlayer separation status is achieved.

[0105] S308, extracting kerf width uniformity, surface ablation depth, and flatness features from the cutting path, and determining whether quality verification has passed by analyzing the matching degree of the defect feature model; S309: If the quality verification fails, the weight coefficient of the compensation parameter calculation model is updated based on the quality characteristics, thermal conductivity, absorptivity and surface reflectivity, and the steps S304 to S308 are returned to be executed until the quality characteristics of the cutting path meet the preset standards.

[0106] The embodiment of the present application discloses a method for adaptive control of laser cutting power, which obtains the laser output power timing data in real time, extracts the power fluctuation characteristics and matches them with the preset defect model to calculate the defect probability. When the probability exceeds the threshold, the energy input efficiency, thermal diffusion compensation amount, etc. are calculated in combination with the cutting parameters and material properties to generate the initial power compensation parameters and the slit quality optimization parameters. The adaptive algorithm is used to perform multi-objective iterative optimization of the parameters to generate a dynamic power control signal, adjust the laser output power and focus position, and realize real-time correction of the cutting path. By extracting the slit characteristics and verifying the quality, the compensation model is continuously updated until the preset standard is reached. Through this application, the quality and efficiency of laser cutting can be effectively improved, and it can adapt to different materials and process conditions, and has strong applicability and promotion value.

[0107] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the existing technology, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0108] Figure 3 This is a structural block diagram of a blue laser cutting intelligent control system according to an embodiment of the present application; Figure 3 Shown, including: a time domain analysis module 31 for performing time domain analysis on power timing data of a blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuations of the blue laser; a matching module 32 for matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; an acquisition module 33 for acquiring, when determining that the probability of the defect occurring is greater than a preset probability threshold, cutting parameters corresponding to a current cutting task from a preset processing parameter database, and acquiring material parameters of the target material from a material parameter database, wherein the current cutting task is for cutting the target material; A determination module 34 is configured to determine an initial power compensation parameter and a kerf quality optimization parameter according to the cutting parameter and the material parameter through a compensation parameter calculation model; an optimization module 35 for performing multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data by using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; The adjustment module 36 is configured to adjust the output power and beam focus position of the blue laser according to the dynamic power control signal, so as to correct the cutting path of the blue laser in real time.

[0109] By means of the above-mentioned device, the power timing data of the blue laser is subjected to time series analysis in real time during the operation of the blue laser to obtain power fluctuation data, wherein the blue laser is used to perform laser cutting on the target material by outputting the blue laser, and the power fluctuation data indicates the power fluctuation of the blue laser; the power fluctuation data is matched with a preset defect characteristic model to determine the defect occurrence probability of the blue laser; if the defect occurrence probability is greater than a preset probability threshold, the cutting parameters corresponding to the current cutting task are obtained from a preset processing parameter database, and the material parameters of the target material are obtained from a material parameter database, wherein the current cutting task is used to cut the target material; according to the cutting parameters and the material parameters, the initial power compensation parameters and the slit quality optimization parameters are determined by the compensation parameter calculation model; then, the initial power compensation parameters, the slit quality optimization parameters and the power fluctuation data are subjected to multi-objective iterative optimization by an adaptive deviation adjustment algorithm to generate a dynamic power control signal; finally, the output power and beam focusing position of the blue laser are adjusted in real time according to the dynamic power control signal to correct the cutting path of the blue laser in real time; by adopting the above scheme, by real-time monitoring of the power fluctuation data of the blue laser and matching it with the preset defect characteristic model, it is possible to predict in advance the defects that may occur in the cutting process, so that when the probability of defect occurrence exceeds the threshold, the output power and beam focusing position of the laser are dynamically adjusted; through this intelligent control method, not only the cutting quality is improved and material waste is reduced, but also the adaptive ability of the system is enhanced, so that it can automatically optimize the cutting parameters according to different materials and process conditions, significantly improving the intelligence level and efficiency of laser cutting; thereby solving the problem in the related technology that the existing laser cutting control method has significant limitations in dealing with the real-time fluctuation and instability of the laser output power, and lacks the ability to optimize adaptive parameters, resulting in reduced cutting quality, material waste and limited intelligence level.

[0110] In an exemplary embodiment, the adjustment module 36 is also used to extract features from the cutting path through a feature extraction device to obtain cutting quality features of the cutting path, and perform a matching analysis on the cutting quality features through the preset defect feature model to determine whether the cutting path passes the quality verification; if it is determined that the cutting path fails the quality verification, the weight coefficient of the compensation parameter calculation model is updated according to the cutting quality features and the material parameters to obtain an updated compensation parameter calculation model; and the cutting path is re-corrected through the updated compensation parameter calculation model.

[0111] In an exemplary embodiment, the time domain analysis module 31 is also used to obtain the power timing data from the blue light laser in real time through a high-speed power meter, and perform frequency domain conversion on the power timing data through fast Fourier transform to obtain frequency domain distribution data; perform signal processing on the frequency domain distribution data through a bandpass filter to extract the power fluctuation signal of the target frequency band to obtain a fluctuation signal set; determine the fluctuation amplitude corresponding to the fluctuation signal set, and perform time domain analysis on the fluctuation signal set through a period estimation algorithm to calculate the period and phase of the fluctuation signal set, and determine the fluctuation frequency and phase offset corresponding to the fluctuation signal set based on the period and the phase to obtain the power fluctuation data, wherein the power fluctuation data includes: the fluctuation amplitude, the fluctuation frequency and the phase offset.

[0112] Optionally, the time domain analysis module 31 is also used to perform time domain analysis on the fluctuation signal set through the period estimation algorithm when the fluctuation amplitude is less than or equal to the preset threshold; and to perform data preprocessing on the fluctuation signal set through adaptive mean filtering to obtain a smoothed fluctuation signal set when the fluctuation amplitude is greater than the preset threshold, and to perform time domain analysis on the fluctuation signal set belonging to the smoothed fluctuation signal set through the period estimation algorithm.

[0113] In an exemplary embodiment, the matching module 32 is further configured to determine a plurality of power fluctuation data corresponding to a plurality of first sampling points of the power time series data, wherein the plurality of first sampling points correspond to a plurality of time windows of the power time series data, and the power fluctuation data include: fluctuation amplitude, fluctuation frequency, and phase offset; respectively match the plurality of power fluctuation data with a plurality of first defect feature models, determine a plurality of second sampling points among the plurality of first sampling points that successfully match the first defect feature model, and determine a plurality of second defect feature models among the plurality of first defect feature models that successfully match the plurality of second sampling points, wherein the preset defect feature model includes the plurality of first defect feature models, the plurality of first defect feature models correspond one-to-one to a plurality of cutting defect types, and the plurality of second sampling points correspond one-to-one to the plurality of second defect feature models; respectively determine the preset defect thresholds corresponding to the plurality of second defect feature models, and respectively perform the following operations on the fluctuation amplitude A. i (t) and amplitude defect threshold T A，i , the fluctuation frequency ω i (t) and frequency defect threshold T ω，i , the phase offset ϕ i (t) and the phase defect threshold Tϕ,i are normalized to obtain the normalized fluctuation amplitude and normalized amplitude defect threshold , the normalized fluctuation frequency and the normalized frequency defect threshold , the normalized phase offset and the normalized phase defect threshold According to the weight of the fluctuation amplitude, the fluctuation frequency and the phase offset, the formula Calculate the dynamic correlation coefficient γ i (t), where is the amplitude weight, is the frequency weight, is the phase weight, is the tolerance coefficient corresponding to the fluctuation frequency; through the formula Calculate the defect occurrence probability corresponding to each cutting defect type under the power time series data, where P(defect i ) is the probability of occurrence of the i-th type of cutting defect among the multiple cutting defect types, N is the number of the multiple first sampling points, δ ik It is used to indicate whether the kth first sampling point among the multiple first sampling points has the i-th type of cutting defect, and N and k are both positive integers.

[0114] Optionally, the determination module 34 is used to determine the energy input efficiency of the blue light laser based on the absorptivity and thermal conductivity in the material parameters, wherein the absorptivity is used to indicate the absorption rate of the blue light laser by the target material, the energy input efficiency is used to indicate the efficiency of converting the blue light laser into cutting energy, and the thermal conductivity is used to indicate the thermal conductivity of the target material; determine the thermal diffusion compensation amount of the blue light laser based on the thermal conductivity and the cutting depth, cutting width and cutting speed in the cutting parameters, wherein the thermal diffusion compensation amount is used to compensate for the energy loss caused by thermal conduction of the target material; and determine the initial power compensation parameter based on the energy input efficiency and the thermal diffusion compensation amount.

[0115] In an exemplary embodiment, the above-mentioned determination module 64 is also used to determine the laser energy utilization rate based on the absorption rate and surface reflectivity in the material parameters, wherein the laser energy utilization rate is used to indicate the proportion of laser energy absorbed by the target material and used for cutting; determine the beam focusing optimization coefficient based on the surface reflectivity and the focus offset parameter in the cutting parameters, wherein the beam focusing optimization coefficient is used to indicate the degree of energy concentration of the blue light laser at the focus; determine the slit quality optimization parameter based on the laser energy utilization rate, the beam focusing optimization coefficient and the cutting width in the cutting parameters.

[0116] An embodiment of the present application further provides a storage medium, which includes a stored program, wherein the program executes any of the above methods when it is run.

[0117] Optionally, in this embodiment, the storage medium may be configured to store program codes for executing the following steps: S1, performing time domain analysis on power timing data of a blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuation of the blue laser; S2, matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; S3, when it is determined that the probability of the defect occurring is greater than a preset probability threshold, obtaining cutting parameters corresponding to the current cutting task from a preset processing parameter database, and obtaining material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; S4, determining an initial power compensation parameter and a kerf quality optimization parameter using a compensation parameter calculation model according to the cutting parameter and the material parameter; S5, performing multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; S6, adjusting the output power and beam focus position of the blue laser according to the dynamic power control signal, so as to correct the cutting path of the blue laser in real time.

[0118] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0119] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0120] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0121] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program: S1, performing time domain analysis on power timing data of a blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuation of the blue laser; S2, matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; S3, when it is determined that the probability of the defect occurring is greater than a preset probability threshold, obtaining cutting parameters corresponding to the current cutting task from a preset processing parameter database, and obtaining material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; S4, determining an initial power compensation parameter and a kerf quality optimization parameter using a compensation parameter calculation model according to the cutting parameter and the material parameter; S5, performing multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; S6, adjusting the output power and beam focus position of the blue laser according to the dynamic power control signal, so as to correct the cutting path of the blue laser in real time.

[0122] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0123] Optionally, specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0124] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, optionally, they can be implemented using program code executable by the computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be performed in a different order than herein, or they can be made into separate integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0125] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A blue laser cutting intelligent control method, characterized in that: include: Performing time domain analysis on power timing data of a blue laser in real time to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuation of the blue laser; Matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; When it is determined that the probability of the defect occurring is greater than a preset probability threshold, obtaining cutting parameters corresponding to the current cutting task from a preset processing parameter database, and obtaining material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; Determining initial power compensation parameters and kerf quality optimization parameters through a compensation parameter calculation model according to the cutting parameters and the material parameters; Performing multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data by using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; The output power and beam focus position of the blue laser are adjusted according to the dynamic power control signal to correct the cutting path of the blue laser in real time.

2. The blue laser cutting intelligent control method according to claim 1, characterized in that: After adjusting the output power and beam focus position of the blue laser according to the dynamic power control signal to correct the cutting path of the blue laser in real time, the method further includes: Extracting features from the cutting path using a feature extraction device to obtain cutting quality features of the cutting path, and performing a matching degree analysis on the cutting quality features using the preset defect feature model to determine whether the cutting path passes quality verification; When it is determined that the cutting path fails the quality verification, updating the weight coefficient of the compensation parameter calculation model according to the cutting quality characteristics and the material parameters to obtain an updated compensation parameter calculation model; The cutting path is re-corrected using the updated compensation parameter calculation model.

3. The blue laser cutting intelligent control method according to claim 1, characterized in that: Perform real-time time domain analysis on the power timing data of the blue laser to obtain power fluctuation data, including: Acquire the power time series data from the blue laser in real time using a high-speed power meter, and perform frequency domain conversion on the power time series data using fast Fourier transform to obtain frequency domain distribution data; Performing signal processing on the frequency domain distribution data through a bandpass filter to extract the power fluctuation signal of the target frequency band to obtain a fluctuation signal set; Determine the fluctuation amplitude corresponding to the fluctuation signal set, and perform time domain analysis on the fluctuation signal set through a period estimation algorithm to calculate the period and phase of the fluctuation signal set, and determine the fluctuation frequency and phase offset corresponding to the fluctuation signal set based on the period and the phase to obtain the power fluctuation data, wherein the power fluctuation data includes: the fluctuation amplitude, the fluctuation frequency and the phase offset.

4. The blue laser cutting intelligent control method according to claim 3, characterized in that: Determining the fluctuation amplitude corresponding to the fluctuation signal set and performing time domain analysis on the fluctuation signal set using a period estimation algorithm, including: When the fluctuation amplitude is less than or equal to a preset threshold, performing time domain analysis on the fluctuation signal set by using the period estimation algorithm; When the fluctuation amplitude is greater than a preset threshold, the fluctuation signal set is preprocessed by adaptive mean filtering to obtain a smoothed fluctuation signal set, and the smoothed fluctuation signal set is analyzed in the time domain by the period estimation algorithm.

5. The blue laser cutting intelligent control method according to claim 1, characterized in that: Matching the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser includes: Determine a plurality of power fluctuation data corresponding to a plurality of first sampling points of the power time series data, wherein the plurality of first sampling points correspond to a plurality of time windows of the power time series data, and the power fluctuation data include: fluctuation amplitude, fluctuation frequency, and phase offset; Matching the plurality of power fluctuation data with a plurality of first defect feature models respectively, determining a plurality of second sampling points among the plurality of first sampling points that successfully match the first defect feature models, and determining a plurality of second defect feature models among the plurality of first defect feature models that successfully match the plurality of second sampling points, wherein the preset defect feature model includes the plurality of first defect feature models, the plurality of first defect feature models correspond one-to-one to a plurality of cutting defect types, and the plurality of second sampling points correspond one-to-one to the plurality of second defect feature models; Determine the preset defect thresholds corresponding to the plurality of second defect feature models respectively, and respectively i (t) and amplitude defect threshold T A，i , the fluctuation frequency ω i (t) and frequency defect threshold T ω，i , the phase offset ϕ i (t) and the phase defect threshold Tϕ,i are normalized to obtain the normalized fluctuation amplitude and normalized amplitude defect threshold , the normalized fluctuation frequency and the normalized frequency defect threshold , the normalized phase offset and the normalized phase defect threshold ; According to the weights of the fluctuation amplitude, the fluctuation frequency and the phase offset, the formula Calculate the dynamic correlation coefficient γ i (t), where is the amplitude weight, is the frequency weight, is the phase weight, is the tolerance coefficient corresponding to the fluctuation frequency; By formula Calculate the defect occurrence probability corresponding to each cutting defect type under the power time series data, where P(defect i ) is the probability of occurrence of the i-th type of cutting defect among the multiple cutting defect types, N is the number of the multiple first sampling points, δ ik It is used to indicate whether the kth first sampling point among the multiple first sampling points has the i-th type of cutting defect, and N and k are both positive integers.

6. The blue laser cutting intelligent control method according to claim 1, characterized in that: According to the cutting parameters and the material parameters, determining the initial power compensation parameters and the kerf quality optimization parameters through a compensation parameter calculation model includes: determining the energy input efficiency of the blue laser according to the absorptivity and thermal conductivity of the material parameters, wherein the absorptivity is used to indicate the absorption rate of the blue laser by the target material, the energy input efficiency is used to indicate the efficiency of converting the blue laser into cutting energy, and the thermal conductivity is used to indicate the thermal conductivity of the target material; determining a thermal diffusion compensation amount of the blue laser according to the thermal conductivity and the cutting depth, cutting width, and cutting speed among the cutting parameters, wherein the thermal diffusion compensation amount is used to compensate for energy loss caused by thermal conduction of the target material; The initial power compensation parameter is determined according to the energy input efficiency and the thermal diffusion compensation amount.

7. The blue laser cutting intelligent control method according to claim 1, characterized in that: According to the cutting parameters and the material parameters, determining the initial power compensation parameters and the kerf quality optimization parameters through a compensation parameter calculation model includes: determining a laser energy utilization rate according to the absorptivity and the surface reflectivity in the material parameters, wherein the laser energy utilization rate is used to indicate a proportion of the laser energy absorbed by the target material and used for cutting; Determining a beam focusing optimization coefficient according to the surface reflectivity and a focus offset parameter in the cutting parameters, wherein the beam focusing optimization coefficient is used to indicate the energy concentration degree of the blue laser at the focus; The slit quality optimization parameter is determined according to the laser energy utilization rate, the beam focusing optimization coefficient and the cutting width in the cutting parameters.

8. A blue laser cutting intelligent control system, characterized in that: include: a time domain analysis module for performing real-time time domain analysis on power timing data of a blue laser to obtain power fluctuation data, wherein the blue laser is used to cut a target material by outputting a blue laser, and the power fluctuation data is used to indicate power fluctuations of the blue laser; a matching module, configured to match the power fluctuation data with a preset defect characteristic model of the blue laser to determine a defect occurrence probability of the blue laser; an acquisition module, configured to, when determining that the probability of occurrence of the defect is greater than a preset probability threshold, acquire cutting parameters corresponding to a current cutting task from a preset processing parameter database, and acquire material parameters of the target material from a material parameter database, wherein the current cutting task is used to cut the target material; a determination module, configured to determine, according to the cutting parameters and the material parameters, an initial power compensation parameter and a slit quality optimization parameter through a compensation parameter calculation model; an optimization module, configured to perform multi-objective iterative optimization on the initial power compensation parameter, the slit quality optimization parameter, and the power fluctuation data by using an adaptive deviation adjustment algorithm to generate a dynamic power control signal; The adjustment module is used to adjust the output power and beam focus position of the blue laser according to the dynamic power control signal, so as to correct the cutting path of the blue laser in real time.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 7 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 7 through the computer program.

Images