Automatic cloth slitting carbon dioxide laser tube device

Abstract

This invention relates to the field of laser cutting equipment technology and provides an automatic fabric slitting CO2 laser tube device with an adaptive adjustment system. The system includes an adaptive cutting quality prediction module, which constructs an adaptive cutting quality prediction model based on the previous cycle's comprehensive cutting quality index, fabric condition comprehensive index, equipment health index, current laser power setting, and previously set cutting speed, and outputs the predicted cutting quality value. It also includes an intelligent pulse frequency dynamic adjustment module, which constructs an intelligent pulse frequency dynamic adjustment model based on the predicted cutting quality value, the target cutting quality benchmark value, and the current pulse frequency setting, and outputs the target pulse frequency setting value. This application integrates multiple modules, including comprehensive fabric condition assessment, equipment health assessment, online cutting quality detection and prediction, and intelligent pulse frequency dynamic adjustment, to form a highly adaptive and intelligent laser cutting control system.

Description

Technical Field

[0001] This invention belongs to the field of laser cutting equipment technology, and in particular relates to an automatic fabric cutting carbon dioxide laser tube cutting device. Background Technology

[0002] In the fields of textiles, apparel, and flexible materials processing, carbon dioxide laser technology has become a key means of automatic fabric cutting due to its high efficiency and precision.

[0003] Traditional fabric laser cutting equipment typically includes basic components such as a cutting table, guide frame, laser cutting device that can move along the width direction, and belt conveyor, supplemented by sensor systems to monitor fabric flatness, tension, and equipment parameters such as laser power and pulse frequency. However, these devices mostly employ static or segmented control strategies, making it difficult to adapt to the combined effects of fabric characteristic fluctuations, equipment performance drift, and complex coupling between process parameters during dynamic conveying.

[0004] Specifically, the existing system's assessment of fabric conditions is limited to data from a single sensor, such as relying solely on tension sensors or visual inspection of flatness. It fails to establish a real-time comprehensive assessment model that integrates multiple factors, including fabric flatness and tension, resulting in an inability to accurately capture dynamic changes in fabric conditions and causing lag or deviation in control response. Furthermore, the health status monitoring of key equipment parameters, such as laser output power and actual pulse frequency, is disconnected from the cutting process. The system ignores the cumulative effect of the decay or shift of these parameters over time on cutting quality and lacks a real-time process correction mechanism based on equipment health.

[0005] The aforementioned defects collectively cause fluctuations in cutting quality, manifested in problems such as excessive carbonization, increased burr density, and uneven kerf width, which severely restrict the improvement of precision and automation level in fabric laser cutting. Summary of the Invention

[0006] The purpose of this invention is to provide an automatic fabric cutting carbon dioxide laser tube device to solve the above-mentioned problems.

[0007] This invention is implemented as follows: an automatic fabric slitting CO2 laser tube device includes: a cutting table, a guide frame mounted on the cutting table, a laser cutting device movable along the width of the cutting table mounted on the guide frame, a belt conveyor for conveying the fabric mounted on the cutting table, and a row of guide rollers mounted above the belt conveyor; it also includes an adaptive adjustment system, comprising:

[0008] The fabric condition comprehensive evaluation module constructs a comprehensive evaluation model for fabric conditions based on the fabric flatness index and real-time fabric tension, and outputs the comprehensive fabric condition index.

[0009] The equipment health assessment module constructs an equipment health assessment model based on the measured output power of the laser cutting device, the actual laser pulse frequency, the actual acceleration of the axis motion, and the offset of the focused spot position, and outputs the equipment health index.

[0010] The online detection and quantification module for cutting quality outputs a comprehensive cutting quality index for the previous period by constructing an online quantitative evaluation model for cutting quality.

[0011] The adaptive cutting quality prediction module constructs an adaptive cutting quality prediction model based on the previous cycle's comprehensive cutting quality index, comprehensive fabric condition index, equipment health index, current laser power setting, and previous cutting speed setting, and outputs the predicted cutting quality value.

[0012] The intelligent pulse frequency dynamic adjustment module constructs an intelligent pulse frequency dynamic adjustment model based on the predicted cutting quality value, the target cutting quality benchmark value, and the current pulse frequency setting value. It outputs the target pulse frequency setting value and executes frequency adjustment control through the controller.

[0013] Further technical solutions, in the comprehensive evaluation model of fabric working conditions:

[0014] The comprehensive index of fabric working conditions is obtained by multiplying the normalized fabric smoothness index and fabric tension index by the corresponding positive influence coefficients and summing them, and then using the sum as a variable to input an S-shaped exponential function for calculation and output.

[0015] The fabric smoothness index and fabric tension index are obtained by normalizing the current smoothness and current fabric tension by substituting them into the maximum-minimum normalization formula.

[0016] Further technical solutions, in the equipment health assessment model:

[0017] The equipment health index is obtained by multiplying the normalized output power index, pulse frequency index, acceleration index, and (1 minus the focused spot position offset index) by the corresponding preset positive influence coefficients and then summing them up, where the sum of each influence coefficient is 1.

[0018] The output power index, pulse frequency index, acceleration index, and focused spot position offset index are obtained by substituting the measured output power, actual pulse frequency, actual axis motion acceleration, and focused spot position offset of the laser cutting device into the maximum-minimum normalization formula for normalization.

[0019] A further technical solution involves calculating the comprehensive cutting quality index of the previous cycle as follows: based on the uniformity coefficient of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length after the previous cycle, an online quantitative evaluation model for cutting quality is constructed, and the comprehensive cutting quality index of the previous cycle is calculated and output.

[0020] Further technical solutions include an online quantitative evaluation model for cutting quality:

[0021] The previous cycle's cutting quality comprehensive index is obtained by multiplying the kerf width uniformity index by an exponential decay function with the cutting edge carbonization index and the burr density per unit length as variables, where both influence coefficients are positive.

[0022] The uniformity index of kerf width, the carbonization index of cutting edge, and the burr density index per unit length are obtained by substituting the uniformity coefficient of kerf width, the carbonization index of cutting edge, and the burr density per unit length after the previous cutting cycle into the maximum-minimum normalization formula for normalization.

[0023] A further technical solution, in the adaptive cutting quality prediction model:

[0024] The predicted cutting quality value is obtained by multiplying the weighted indexes of the previous cycle's cutting quality comprehensive index, fabric condition comprehensive index, equipment health index, and (the ratio of the current set laser power index to the current set cutting speed index) together.

[0025] All weighting coefficients are positive numbers. The current set laser power index and the current set cutting speed index are obtained by substituting the current set cutting speed and the current set laser power into the maximum-minimum value normalization formula for normalization.

[0026] A further technical solution, in the intelligent pulse frequency dynamic adjustment model:

[0027] The predicted cutting quality value is compared with the target cutting quality benchmark value, and an exponential adjustment is performed based on the current pulse frequency setting value to obtain the normalized value of the target pulse frequency. This value is then mapped to the preset upper and lower frequency limits to output the actual target pulse frequency.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] This application integrates multiple modules, including comprehensive fabric condition assessment, equipment health assessment, online cutting quality detection and prediction, and intelligent pulse frequency dynamic adjustment, to form a highly adaptive and intelligent laser cutting control system. This system can comprehensively perceive, intelligently decide, and precisely execute, significantly improving the stability, accuracy, and automation level of fabric laser cutting, providing a more advanced solution for the flexible materials processing industry.

[0030] The adaptive cutting quality prediction module of this application effectively solves the problems of inaccuracy and lack of dynamic adaptability in traditional prediction methods. This scheme constructs a multi-factor, adaptive prediction model by comprehensively considering the previous cycle's cutting quality, the overall condition of the fabric, equipment health, and the coupling relationship between laser power and cutting speed. This allows the prediction results to more accurately reflect the quality fluctuations that may occur during the actual cutting process, thus providing a reliable and accurate basis for subsequent intelligent pulse frequency dynamic adjustment. Specifically, the introduction of the previous cycle's comprehensive cutting quality index enables feedback learning from historical cutting results, avoiding the accumulation of errors; combining the comprehensive fabric condition index and the equipment health index allows the prediction model to respond in real time to changes in fabric characteristics and equipment performance drift, enhancing the real-time nature and comprehensiveness of the prediction; and the introduction of the laser power and cutting speed coupling term accurately captures the interaction between these two key process parameters, further improving the accuracy of the prediction. Through this comprehensive and dynamic prediction mechanism, the equipment can anticipate potential cutting quality problems and adjust process parameters in a timely manner, significantly improving the cutting quality stability, accuracy, and intelligence level of the automatic fabric slitting CO2 laser tube equipment, reducing the scrap rate, and increasing production efficiency.

[0031] This application effectively solves the problems of insufficient precision or efficiency in traditional adjustment mechanisms, and the inefficiency caused by reliance on manual experience leading to fluctuations in cutting quality. The intelligent pulse frequency dynamic adjustment model, by introducing specific mathematical relationships, achieves precise quantitative adjustment of the pulse frequency. Based on the predicted cutting quality value and real-time comparison with the target cutting quality benchmark value, the system can proactively and forward-lookingly adjust the pulse frequency of the laser cutting device 3, rather than responding laggingly to quality problems that have already occurred. This dynamic adjustment mechanism significantly improves the real-time performance and accuracy of the cutting process, avoiding instability in cutting quality caused by changes in fabric conditions or equipment state drift. Furthermore, by inversely normalizing the normalized target frequency to the actual working range, and considering the frequency adjustment sensitivity and small constants to prevent division by zero, the stability and safety of the adjustment process are ensured. Overall, this solution enables the automatic fabric slitting CO2 laser tube equipment to achieve a higher level of automation and intelligence, effectively reducing quality defects such as over-carbonization, increased burrs, and uneven cuts, thereby improving product qualification rate and production efficiency. Attached Figure Description

[0032] Figure 1 A schematic diagram of a carbon dioxide laser tube device for automatic fabric cutting;

[0033] Figure 2 This is a schematic diagram of the modules of an adaptive adjustment system.

[0034] In the attached diagram: 1. Cutting table; 2. Guide frame; 3. Laser cutting device; 4. Belt conveyor; 5. Guide roller. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0036] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0037] like Figures 1-2 As shown, an automatic fabric slitting CO2 laser tube device according to an embodiment of the present invention includes: a cutting table 1, a guide frame 2 on the cutting table 1, a laser cutting device 3 movable along the width direction of the cutting table on the guide frame 2, a belt conveyor 4 for conveying the fabric on the cutting table 1, and a row of guide rollers 5 above the belt conveyor 4; it also includes an adaptive adjustment system, which includes:

[0038] The fabric condition comprehensive evaluation module constructs a comprehensive evaluation model for fabric conditions based on the fabric flatness index and real-time fabric tension, and outputs the comprehensive fabric condition index.

[0039] The equipment health assessment module constructs an equipment health assessment model based on the measured output power of the laser cutting device 3, the actual laser pulse frequency, the actual acceleration of the axis motion, and the offset of the focused spot position, and outputs the equipment health index.

[0040] The online detection and quantification module for cutting quality outputs a comprehensive cutting quality index for the previous period by constructing an online quantitative evaluation model for cutting quality.

[0041] The adaptive cutting quality prediction module constructs an adaptive cutting quality prediction model based on the previous cycle's comprehensive cutting quality index, comprehensive fabric condition index, equipment health index, current laser power setting, and previous cutting speed setting, and outputs the predicted cutting quality value.

[0042] The intelligent pulse frequency dynamic adjustment module constructs an intelligent pulse frequency dynamic adjustment model based on the predicted cutting quality value, the target cutting quality benchmark value, and the current pulse frequency setting value. It outputs the target pulse frequency setting value and executes frequency adjustment control through the controller.

[0043] In this embodiment, the cutting table 1 provides a stable support surface for the placement and cutting of the fabric. The guide frame 2 is securely mounted on the cutting table 1, serving to provide a precise moving track for the laser cutting device 3. The laser cutting device 3 is configured to move along the width of the cutting table 1 on the guide frame 2, thereby covering the entire fabric cutting area. A belt conveyor 4 is also mounted on the cutting table 1, its main function being to achieve continuous and automated fabric transport, ensuring smooth cutting operations. To further ensure the flatness of the fabric during transport, a row of guide rollers 5 is positioned above the belt conveyor 4 to apply appropriate pressure and guide the fabric. For example, in practical applications, the laser cutting device 3 can integrate a carbon dioxide laser, a focusing lens group, and a gas-assisted cutting nozzle to achieve effective cutting of fabrics of different thicknesses.

[0044] The core of this application lies in the introduction of an adaptive adjustment system. This system achieves intelligent optimization of the cutting process through the collaborative work of multiple functional modules. Firstly, the adaptive adjustment system includes a comprehensive fabric condition assessment module. This module's main function is to acquire real-time information on the fabric's flatness and tension, and based on this information, construct a comprehensive fabric condition assessment model, ultimately outputting a comprehensive fabric condition index. Secondly, the adaptive adjustment system also includes an equipment health assessment module. This module is responsible for monitoring key operating parameters of the laser cutting device 3, and based on these parameters, constructing an equipment health assessment model and outputting an equipment health index. Thirdly, the adaptive adjustment system includes an online cutting quality detection and quantification module. This module constructs an online quantitative assessment model for cutting quality and outputs the comprehensive cutting quality index for the previous cycle. Furthermore, the adaptive adjustment system also includes an adaptive cutting quality prediction module. This module constructs an adaptive cutting quality prediction model based on the comprehensive cutting quality index of the previous cycle, the comprehensive fabric condition index, the equipment health index, the currently set laser power, and the previously set cutting speed, and outputs the predicted cutting quality value.

[0045] Finally, the adaptive adjustment system also includes an intelligent pulse frequency dynamic adjustment module. This module constructs an intelligent pulse frequency dynamic adjustment model based on the predicted cutting quality value, the target cutting quality reference value, and the current pulse frequency setpoint. It outputs the target pulse frequency setpoint and executes frequency adjustment control through the controller. For example, this adjustment model can be a simple proportional controller. When the predicted cutting quality value is lower than the target cutting quality reference value, the adjustment model calculates an increment in the pulse frequency and adds it to the current pulse frequency setpoint to obtain the target pulse frequency setpoint. Conversely, when the predicted value is higher than the target value, the pulse frequency is reduced. This target pulse frequency setpoint is then sent to the controller, which converts the setpoint into an actual electrical signal to drive the laser of the laser cutting device 3 to adjust its pulse frequency, thereby achieving real-time closed-loop control of the cutting quality.

[0046] In a preferred embodiment of the present invention, the comprehensive evaluation model for fabric working conditions includes:

[0047] ;

[0048] in This is the flatness influence coefficient. The tension influence coefficient is... , The fabric smoothness index. The fabric tension index. This refers to the comprehensive index of fabric working conditions.

[0049] , The calculation method is as follows: Substitute the current flatness and current fabric tension into the maximum-minimum normalization formula for normalization, and then obtain the fabric flatness index and fabric tension index.

[0050] In this embodiment, the comprehensive fabric condition evaluation model is a mathematical model that comprehensively considers multiple physical parameters affecting fabric cutting quality, such as fabric flatness and fabric tension, and outputs a unified index to facilitate decision-making by the subsequent adaptive adjustment system. This model can be constructed and optimized based on empirical data, expert knowledge, or machine learning algorithms.

[0051] Fabric flatness is measured by acquiring images from vision sensors installed above the conveyor path, and then analyzing the surface undulations using image processing algorithms. Fabric tension is measured in real time by tension sensors installed at both ends of the conveyor belt. This raw data is then input into a comprehensive fabric condition evaluation model, which can be a simple weighted average model that linearly combines the flatness and tension data to obtain a preliminary comprehensive fabric condition index.

[0052] Among them, the comprehensive index of fabric working conditions The calculation formula uses the Sigmoid function form, which is characterized by its ability to map input values ​​to a finite interval and its nonlinear properties, thus better simulating the complex relationships between actual physical quantities. This function form makes the comprehensive index of fabric working conditions... It can smoothly reflect the fabric smoothness index. and fabric tension index The changes avoid the abrupt changes or inaccuracies that might arise from linear models.

[0053] Flatness Influence Coefficient and tension influence coefficient Used to adjust the fabric smoothness index And the fabric tension index T in the comprehensive index of fabric working conditions The relative weights used in the calculations. They are all positive values, indicating that both flatness and tension have a positive impact on fabric conditions. , The model can be obtained through offline experiments, expert experience settings, or by learning from historical data through machine learning methods to ensure that it can accurately reflect the actual working conditions under different fabric types and cutting conditions.

[0054] Fabric smoothness index and fabric tension index These are normalized values ​​representing fabric smoothness and fabric tension, respectively. They are obtained by normalizing the original measurement data (current smoothness, current fabric tension) to eliminate dimensional differences between different physical quantities, making them comparable in the model. The max-min normalization formula is a commonly used data preprocessing method. Its basic principle is to linearly transform the original data to a specified interval. This normalization method effectively unifies data with different dimensions to the same scale, thus preventing certain features from dominating the model due to excessively large numerical ranges, ensuring that the model fairly considers the influence of all input features.

[0055] The automatic fabric slitting CO2 laser tube device of this application addresses the lack of a specific mathematical model to comprehensively evaluate fabric conditions, particularly flatness and tension, in traditional equipment. By incorporating the aforementioned comprehensive fabric condition evaluation model into its fabric condition evaluation module, the core of this model lies in using a Sigmoid function to calculate the comprehensive fabric condition index. During model operation, the fabric's smoothness and tension data are first acquired in real time via sensors. This raw data is then processed using a maximum-minimum normalization formula to obtain a dimensionless fabric smoothness index. and fabric tension index This normalization process ensures the numerical comparability of different physical quantities and avoids model bias towards a particular parameter due to differences in dimensions. Next, the normalized... and Combined with the preset flatness influence coefficient and tension influence coefficient (All are positive values, used to adjust their respective weights in the model), which are then substituted into the Sigmoid function. The Sigmoid function can... and The linear combination is nonlinearly mapped to the interval between 0 and 1 to generate a comprehensive index of fabric working conditions. This nonlinear mapping allows the model to capture more precisely the complex changes in fabric conditions. For example, once flatness or tension reaches a certain threshold, its impact on the overall condition may no longer be linear. In this way, the fabric condition comprehensive evaluation module can output a continuous and comprehensive fabric condition comprehensive index. This index accurately reflects the overall condition of the fabric during the cutting process. This comprehensive fabric condition index... It can then be used as input to the adaptive cutting quality prediction module, working together with the equipment health index, the comprehensive cutting quality index, etc., to achieve accurate prediction of cutting quality and intelligent dynamic adjustment of subsequent process parameters, significantly improving the intelligence level of the entire adaptive adjustment system and the stability of cutting quality.

[0056] In a preferred embodiment of the present invention, the equipment health assessment model includes:

[0057] ;

[0058] in The power influence coefficient, For frequency influence coefficient, For acceleration influence coefficient, This is the influence coefficient of light spot offset. , , , as well as All are greater than 0; The output power index, The pulse frequency index, For acceleration index, This is the index for the position offset of the focused light spot. Equipment health index;

[0059] , , as well as The calculation method is as follows: the measured output power, actual pulse frequency, actual acceleration of axis motion, and focus spot position offset of the laser cutting device 3 are successively substituted into the maximum-minimum value normalization formula for normalization processing, and then the output power index, pulse frequency index, acceleration index, and focus spot position offset index are obtained in sequence.

[0060] In this embodiment, the equipment health assessment model aims to comprehensively reflect the overall health status of the automatic fabric slitting CO2 laser tube equipment during operation through quantitative methods. The concept involves integrating key parameters affecting equipment performance to form a unified health index, facilitating subsequent process parameter adjustments and maintenance decisions. The formula integrates multiple normalized equipment performance indicators into a comprehensive equipment health index through a linear weighted combination. .

[0061] The actual output power of the laser cutting device 3 can be measured in real time using a power meter integrated inside the laser. The actual laser pulse frequency can be obtained from the feedback signal of the laser drive circuit. The actual acceleration of the axis motion can be measured using an accelerometer mounted on the motion axis of the laser cutting device 3. The focused spot position offset can be obtained by setting a simple optical detector in the cutting area to periodically detect the deviation between the spot center position and the theoretical position. These parameters are input into the equipment health assessment model, which can be a threshold-based system. When any parameter exceeds a preset safety range, the equipment health index will decrease accordingly.

[0062] Power Influence Coefficient Frequency influence coefficient Acceleration influence coefficient and the influence coefficient of light spot offset These represent the relative importance or influence of each equipment performance indicator on the equipment health assessment. Their purpose is to adjust the weight of different parameters in the health assessment based on actual process requirements and equipment characteristics. These coefficients can be determined through expert experience, i.e., by experienced engineers based on equipment operating characteristics and failure modes; or through data-driven methods, such as using historical operating data and equipment failure records, and optimizing them using statistical methods such as regression analysis and principal component analysis. This constraint... , , , as well as All values ​​greater than 0 ensure that the sum of all influence coefficients is 1, and that each coefficient is positive. This guarantees the equipment health index. The calculation results are comparable, and each parameter has a positive impact on the health assessment. Besides limiting the sum to 1, the coefficients can also be set to any positive value, and then the final... The values ​​should be normalized; or decision-making methods such as the analytic hierarchy process (AHP) should be used to determine the relative weights of these coefficients to ensure their rationality.

[0063] Output power index Pulse frequency index Acceleration index Focused spot position offset index and equipment health index These are normalized equipment performance indicators and the final comprehensive health indicators. Their function is to unify raw measurement data of different dimensions and ranges onto a comparable scale, thereby enabling effective weighted combination. In addition to exponential form, these indicators can also be expressed in percentage or rank form to adapt to different data representation needs. , , as well as The calculation method involves converting the raw measurement data of the laser cutting device 3, such as the measured output power, actual pulse frequency, actual acceleration of the axis motion, and the offset of the focused spot position, into a dimensionless exponent using a maximum-minimum normalization formula. Its purpose is to eliminate dimensional and numerical range differences between different physical quantities, allowing them to be compared and weighted on a unified scale.

[0064] Max-min normalization is a commonly used data preprocessing method. Its basic principle is to linearly map the original data to the range [0,1] (or a specified range). Its purpose is to ensure that all input parameters have the same numerical range before weighted summation, thus guaranteeing the fairness and accuracy of the evaluation. Besides this linear normalization, non-linear normalization methods, such as logarithmic transformations or sigmoid function transformations, can also be used to handle data with skewed distributions or outliers.

[0065] This application proposes a solution that constructs a device health assessment model to achieve a comprehensive and quantitative evaluation of the operating status of an automatic fabric slitting CO2 laser tube device. The model first collects key operating parameters in real time, including the measured output power, actual pulse frequency, actual acceleration of the axis motion, and the offset of the focused spot position of the laser cutting device 3. These raw parameters, due to their different physical dimensions and numerical ranges, cannot be directly compared and synthesized. Therefore, this solution uses a maximum-minimum normalization formula to process these measured data, uniformly converting them into an output power index. Pulse frequency index Acceleration index and the index of focused spot position offset These indices all fall within a standardized numerical range, thus eliminating dimensional differences and making different parameters comparable. Based on this, this scheme uses these normalized indices as input and applies a linear weighted combination formula... To calculate the overall health index of the equipment .in, , , , These are the power influence coefficient, frequency influence coefficient, acceleration influence coefficient, and spot offset influence coefficient, each representing the relative importance of the corresponding parameter to the overall equipment health. These coefficients are all set to be greater than 0 and sum to 1, ensuring that each key parameter has a positive impact on the health assessment, and that the assessment results have a uniform scale and interpretability. In this way, the equipment health assessment model can integrate multiple independent equipment operating parameters into a single, comprehensive quantitative index, objectively reflecting the current operating status of the equipment. This equipment health index... The output provides crucial input information for the subsequent adaptive cutting quality prediction module. In the automatic fabric slitting CO2 laser tube equipment, the equipment's health status directly affects the cutting quality. For example, attenuation of laser output power, instability of pulse frequency, abnormal acceleration caused by wear of the motion axis, and deviation of the focused spot position can all lead to poor cutting results, such as uneven cuts, increased burrs, and severe carbonization. The equipment health assessment model in this solution can promptly and accurately capture these potential equipment problems and quantify them into a health index. This index is then used for adaptive cutting quality prediction, enabling the system to anticipate potential cutting quality problems under the current equipment condition. This provides a basis for the intelligent pulse frequency dynamic adjustment module, allowing for advance adjustment and optimization of cutting process parameters, avoiding a decline in cutting quality due to poor equipment condition. This mechanism, which combines equipment health with cutting quality prediction and process parameter adjustment, gives the entire automatic fabric slitting CO2 laser tube equipment stronger adaptability and stability.

[0066] As a preferred embodiment of the present invention, the comprehensive cutting quality index of the previous cycle is calculated as follows: based on the uniformity coefficient of the kerf width after the previous cycle, the carbonization index of the cutting edge, and the burr density per unit length, an online quantitative evaluation model for cutting quality is constructed, and the comprehensive cutting quality index of the previous cycle is calculated and output.

[0067] In this embodiment, the uniformity coefficient of the kerf width after the previous cutting cycle is used to quantify the degree of fluctuation of the kerf width within a certain length range after the fabric is cut by the laser cutting device 3. This reflects the stability of the cutting process and the control capability of the cutting parameters on the kerf width. This coefficient provides an objective and quantifiable indicator to evaluate the consistency of the kerf width in the cutting quality, and is a key basis for judging whether the cutting is uniform. It can be implemented by using a vision inspection system to acquire images of the kerf after cutting, then using image processing algorithms to measure the width of the kerf at multiple points, and calculating the standard deviation or coefficient of variation of these widths to obtain the kerf width uniformity coefficient; alternatively, it can be used with a laser displacement sensor or a linear CCD camera to continuously scan the kerf along the cutting path, acquire the contour data of the kerf edge, and then calculate the fluctuation of the kerf width, converting it into a uniformity coefficient.

[0068] The edge carbonization index is used to quantify the degree of carbonization of fabric edges caused by laser thermal effects. Carbonization is a common quality defect in laser cutting, manifested as blackening, hardening, or brittleness of the cut edges. This index provides a quantitative indicator to assess the degree of thermal damage to the cut edges and is an important basis for evaluating the rationality of cutting process parameters. It can be implemented by using a high-resolution vision system combined with color recognition or grayscale analysis algorithms to analyze the color and texture of the cut edges and calculate the carbonization index based on the area, depth, or color intensity of the carbonized region; alternatively, it can be achieved through spectral analysis techniques to detect changes in the chemical composition of the cut edge material, particularly the content of carbonization products, thereby indirectly or directly quantifying the degree of carbonization.

[0069] Burr density per unit length is used to quantify the number or length of incompletely cut fibers or melt (i.e., burrs) appearing per unit length at the edge of a fabric cut. Burrs are a significant factor affecting cutting quality and subsequent processing. This density provides a quantitative indicator to assess the neatness and smoothness of the cut edge, and is a key parameter for determining whether the cut is thorough and the edge is clean. It can be achieved by using a high-magnification vision inspection system to acquire images of the cut edge, using edge detection and feature extraction algorithms to identify and count the number of burrs per unit length, or by measuring the total length of the burrs to calculate the burr density; alternatively, it can be achieved by using contact or non-contact sensors (such as laser profilometers) to scan the microscopic morphology of the cut edge, identify and quantify protrusions that extend beyond the normal edge contour, thereby assessing the burr density.

[0070] The online quantitative evaluation model for cutting quality is a mathematical or algorithmic framework used to comprehensively process the aforementioned cutting quality indicators (kerf width uniformity coefficient, cutting edge carbonization index, and burr density per unit length) and transform them into a single, quantifiable "previous cycle cutting quality comprehensive index." The model's function is to integrate multiple independent quality indicators into a unified evaluation value, enabling the overall cutting quality to be reflected in real time and objectively, providing a basis for subsequent adaptive adjustments. Its implementation can employ weighted average methods, fuzzy comprehensive evaluation methods, or neural network models. After normalizing each indicator, different weights are assigned based on their importance to cutting quality, followed by weighted summation or mapping through a nonlinear function to obtain the comprehensive index. Alternatively, it can be based on expert experience or historical data, constructing decision trees, support vector machines, or deep learning models to learn the complex relationship between various indicators and the final cutting quality, achieving intelligent quantitative evaluation of cutting quality.

[0071] The calculation and output of the previous cycle's comprehensive cutting quality index refers to processing various quality indicators of the currently completed cutting cycle (i.e., the "previous cycle") using the aforementioned online quantitative evaluation model for cutting quality, and generating a single numerical value representing the cutting quality of that cycle. This can be achieved by setting up an independent quality evaluation processor. This processor receives raw data from modules such as visual inspection and sensors, performs preprocessing and feature extraction, then calls the online quantitative evaluation model for cutting quality calculation, and sends the results to other control modules.

[0072] In this application, to address the lag and inaccuracy issues in cutting quality assessment of traditional fabric laser cutting equipment, a refined online quantitative assessment mechanism for cutting quality is proposed. The core of this mechanism lies in its departure from relying on single or vague quality judgments. Instead, it quantifies multiple key cutting defect indicators and integrates them into a comprehensive model. Specifically, after each cutting cycle, the system acquires in real-time the uniformity coefficient of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length. These indicators comprehensively reflect the cutting quality from three dimensions: geometric accuracy of the kerf, degree of thermal damage, and edge cleanliness. Subsequently, these quantitative indicators are input into a pre-constructed online quantitative assessment model for cutting quality. This model uses scientific algorithms to fuse and process this multi-dimensional information, ultimately calculating and outputting a single comprehensive index representing the overall cutting quality of the previous cycle.

[0073] The online cutting quality detection and quantification module provides accurate and timely input to the adaptive cutting quality prediction module, enabling the entire adaptive adjustment system to provide effective feedback and optimization based on actual cutting results. For example, when the comprehensive cutting quality index indicates a decline in cutting quality, the adaptive cutting quality prediction module can promptly predict future cutting quality trends and, in conjunction with the intelligent pulse frequency dynamic adjustment module, adjust the pulse frequency of the laser cutting device 3 to correct cutting defects and ensure the stability and consistency of cutting quality. This close linkage mechanism allows the equipment to shift from passive response to active prediction and adjustment, significantly improving the intelligence level and cutting performance of the automatic fabric slitting CO2 laser tube equipment.

[0074] In a preferred embodiment of the present invention, the online quantitative evaluation model for cutting quality includes:

[0075] ;

[0076] in The carbonization influence coefficient at the cutting edge. The coefficient representing the influence of burr density is... , The uniformity index of kerf width. The carbonization index of the cutting edge. The burr density index per unit length. This represents the comprehensive index of cutting quality in the previous cycle.

[0077] , as well as The calculation method is as follows: Substitute the uniformity coefficient of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length after the previous cutting cycle into the maximum-minimum normalization formula for normalization. Then, the uniformity index of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length are obtained in sequence.

[0078] In this embodiment, the online quantitative evaluation model for cutting quality aims to quantitatively evaluate the cutting quality of fabric during laser cutting in real time or near real time. This model can be deployed in the device's controller, for example, via an embedded system or industrial PC, or as part of a cloud service, receiving data from sensors and performing calculations.

[0079] formula This application is used to calculate the comprehensive index of cutting quality in the previous cycle. The core mathematical expression of the kerf width uniformity index is expressed through multiplication and exponential functions. Carbonization index of cutting edge and the burr density index per unit length These three key quality indicators are combined. Among them, As a positive indicator, it directly affects The cardinality; and and As a negative indicator, it is expressed through the exponential decay term. To reduce The value reflects the negative impact of carbonization and burrs on cutting quality. This structure can effectively capture the nonlinear relationship between different quality factors and provide a continuous and sensitive quality assessment result.

[0080] The carbonization influence coefficient at the cutting edge. These are the burr density influence coefficients. These two coefficients are used to adjust the negative impact of edge carbonization and burr density per unit length on the overall cutting quality index. Controlling the carbon index The weight, Controlling the burr density index The weights are all positive, ensuring that the exponential decay term decreases as carbonization and burrs increase, thus... The reduction aligns with the actual situation of quality degradation. These coefficients can be set through expert experience or trained and optimized using machine learning algorithms (such as regression analysis and neural networks) on a large amount of historical segmented data to improve the model output. It closely matches the actual perceived cutting quality.

[0081] The uniformity index of kerf width. The carbonization index of the cutting edge. These are the burr density indices per unit length, and they constitute the overall cutting quality index. The basic quantitative indicator. Cut width uniformity index. This reflects the stability and consistency of the cutting line width, and is highly... The value indicates minimal variation in kerf width and good cutting uniformity. Cutting edge carbonization index. It quantifies the degree of scorching and discoloration of the material at the cut edge, and high The value indicates severe carbonization and poor cutting quality. Burr density index per unit length. It measures the number of burrs or incompletely cut fibers at the cut edge, high The value indicates that there are many rough edges and the cutting quality is poor. These indices are usually obtained by a visual inspection system, such as real-time image acquisition of the cutting area by a high-speed camera, and combined with image processing algorithms (such as edge detection, grayscale analysis, and texture analysis) to extract the corresponding feature values.

[0082] , as well as The calculation method is as follows: Substitute the uniformity coefficient of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length after the previous cutting cycle into the maximum-minimum normalization formula for normalization. This yields the kerf width uniformity index, the cutting edge carbonization index, and the burr density per unit length index. Normalization is the process of converting the original measurement data to a specific range (e.g., [0,1]). The maximum-minimum normalization formula is typically X... norm = (XX) min ) / (X max -X min ), where X is the original data, X min and X max These are the minimum and maximum values ​​of the data within a historical or preset range, respectively. This processing method eliminates the differences in physical dimensions and numerical ranges, allowing different indicators to be compared and calculated on the same scale. It avoids the influence of one indicator dominating or overshadowing other indicators due to its excessively large or small numerical range, thereby improving the stability and accuracy of the model.

[0083] To address the inaccuracy of online quantification of cutting quality in existing technologies, this application proposes a precise online quantification evaluation model for cutting quality. This model first performs maximum-minimum normalization on three key indicators obtained after the previous cutting cycle: the kerf width uniformity coefficient, the cutting edge carbonization index, and the burr density per unit length. This results in a dimensionless kerf width uniformity index. Carbonization index of cutting edge and the burr density index per unit length This normalization process ensures that indices with different physical dimensions can be compared and calculated on a unified scale, effectively avoiding the undue influence of a single index on the overall evaluation results due to differences in dimensions or numerical ranges. Based on this, the model adopts... The mathematical form is used to calculate the comprehensive index of cutting quality in the previous cycle. Among them, the uniformity index of kerf width. As a positive factor, it directly reflects the quality of the cut; the higher the value, the better the cut quality. The carbonization index of the cutting edge... and the burr density index per unit length Then, as a negative factor, it passes through the exponential decay term. This is to demonstrate their negative impact on cutting quality. and As influence coefficients, they respectively adjust the weights of carbonization and burrs on the overall quality. Their positive value ensures that as the degree of carbonization or burrs increases, The value will decrease accordingly. This modeling approach, combining multiplication and exponentiation, not only captures the nonlinear coupling relationships between various quality factors but also provides a continuous evaluation result that is sensitive to quality changes. Through this mechanism, the model can comprehensively and accurately quantify the cutting quality of the previous cycle, providing reliable input data for the subsequent adaptive cutting quality prediction module. This precise online quantification capability enables the entire adaptive adjustment system to make decisions based on real and detailed quality feedback, thereby effectively improving the cutting stability, accuracy, and intelligence level of the automatic fabric slitting CO2 laser tube equipment, and solving the problem of lag or inaccuracy in control response caused by inaccurate quality assessment in traditional methods.

[0084] In a preferred embodiment of the present invention, the adaptive cutting quality prediction model includes:

[0085] ;

[0086] in This is the weighting coefficient for the cutting quality of the previous cycle. This is the fabric working condition weighting coefficient. For equipment health weighting coefficient, These are the coupling weight coefficients. , This is the comprehensive index of cutting quality in the previous cycle. This is a comprehensive index of fabric working conditions. For equipment health index, Set the current cutting speed index. To set the laser power index for the current situation, It is a very small positive constant. This is a predicted value for cutting quality.

[0087] , The calculation method is as follows: Substitute the current set cutting speed and the current set laser power into the maximum-minimum value normalization formula for normalization, and then obtain the current set cutting speed index and the current set laser power index.

[0088] In this embodiment, the adaptive cutting quality prediction model is a mathematical model used to predict cutting quality. Its concept is to utilize multiple input parameters and, through specific mathematical relationships, output a numerical value representing the future cutting quality. This model can be deployed in the central controller of the device, with its mathematical logic implemented using a programming language, or it can run as a standalone software module in a host computer system, interacting with the device controller via a network interface. The weighting coefficients in the model... , , , These coefficients are used to adjust the degree of influence of each input factor on the final prediction result. They are positive to ensure that each factor contributes to the prediction value. Their values ​​can be determined through offline training or expert experience.

[0089] Previous cycle cutting quality comprehensive index This reflects the actual performance of the equipment in its most recent cut, and can be obtained in real time through an online cutting quality detection and quantification module. Fabric Condition Comprehensive Index It is a comprehensive quantitative representation of the current state of the fabric, reflecting factors such as fabric flatness and tension. This value can be obtained in real-time through a comprehensive fabric condition assessment module. Equipment Health Index This is a quantitative representation of the current health status of the laser cutting device 3, reflecting factors such as laser output power, pulse frequency, axis motion acceleration, and focused spot position offset. It can be obtained in real-time through the equipment health assessment module. The currently set cutting speed index... And the currently set laser power index These are the normalized values ​​of the currently set cutting speed and laser power. They map the actual physical quantities to a unified scale for comparison and calculation within the model. Minimal positive constant. Its purpose is to prevent calculation At that time, A mathematical error caused by a value close to or equal to zero, resulting in a denominator of zero, can be a pre-defined, fixed small value. (Cut quality prediction value) It is the final output of the model, representing the expected cutting quality under the current fabric conditions, equipment health, set laser power, and cutting speed.

[0090] The adaptive cutting quality prediction module of this application achieves accurate prediction of fabric cutting quality by constructing a comprehensive adaptive cutting quality prediction model. The core of this model lies in the organic integration of multiple key influencing factors to generate reliable cutting quality prediction values. Specifically, the model first receives the comprehensive cutting quality index from the previous period. This index reflects the actual effect of the equipment's most recent cut, providing historical feedback for prediction. Meanwhile, the comprehensive fabric condition index... It reflects the dynamic state of the fabric, such as its smoothness and tension, in real time, while the equipment health index... The system monitors the performance degradation or drift of the laser cutting device 3 in real time. The introduction of these indices allows the model to comprehensively capture both the external environment and the internal state of the equipment that affect cutting quality. Based on this, the model further considers the currently set process parameters, namely the currently set laser power index. And the current set cutting speed index By introducing The model cleverly captures the coupling effect between laser power and cutting speed, where a minimal positive constant is used. The inclusion of this factor effectively avoids the calculation error of a denominator of zero, ensuring the stability of the calculation. The model uses exponential weight coefficients. These coefficients allow the system to flexibly adjust the relative importance of different factors in the prediction based on actual needs or through learning algorithms, thereby enhancing the model's adaptability and accuracy. All input parameters, including the currently set cutting speed and laser power, undergo max-min normalization, converting them into dimensionless exponential form. This ensures that parameters with different physical dimensions can be effectively calculated on a uniform scale, avoiding prediction bias caused by dimensional differences. Through this multi-factor, dynamically weighted collaborative calculation mechanism, the adaptive cutting quality prediction model can output a highly accurate predicted cutting quality value. This predicted value not only integrates historical performance, real-time operating conditions, and equipment status, but also takes into account the interaction of current process parameters, thus providing an accurate and reliable input for the subsequent intelligent pulse frequency dynamic adjustment module.

[0091] In a preferred embodiment of the present invention, the intelligent pulse frequency dynamic adjustment model includes:

[0092] ;

[0093] ;

[0094] in To adjust the sensitivity for frequency, , The target cutting quality benchmark value, This is a predicted value for cutting quality. To prevent small constants from being divided by zero, The normalized value of the target pulse frequency. , Set an upper limit for the current pulse frequency. Set a lower limit for the current pulse frequency. The actual target pulse frequency; This is the normalized value of the current pulse frequency, which is obtained by substituting the current set pulse frequency into the maximum-minimum normalization formula for normalization.

[0095] In this embodiment, the target pulse frequency normalized value This represents the normalized expected pulse frequency, typically ranging from 0 to 1, used for calculation and comparison on a uniform scale. This normalized value is one of the core outputs of the intelligent pulse frequency dynamic adjustment model, directly reflecting the direction and magnitude of frequency adjustment expected by the system based on current operating conditions and quality predictions.

[0096] Current pulse frequency normalization value This refers to the pulse frequency set by the current device after maximum-minimum normalization. This normalization process maps frequency values ​​of different dimensions or ranges to a unified [0,1] interval, facilitating model calculations. For example, it can be obtained by subtracting the minimum set frequency from the current set frequency and then dividing by the difference between the maximum and minimum set frequencies.

[0097] Target cutting quality benchmark This refers to the ideal cutting quality level expected to be achieved during the operation of an automatic fabric slitting CO2 laser tube device. This benchmark value can be preset according to the specific fabric type, cutting requirements, and product standards. For example, it can be set to a high score such as 0.9 or 0.95, representing a high-quality cutting effect. This value serves as the target for adjusting the model, guiding the direction of pulse frequency adjustment.

[0098] Cutting quality prediction value This is output by the adaptive cutting quality prediction module, representing the expected cutting quality under the current fabric conditions, equipment health, and set process parameters. This predicted value serves as the basis for the model's dynamic adjustments, and its accuracy directly affects the adjustment effect.

[0099] Small constants to prevent division by zero It is a very small positive number, its purpose being to avoid the denominator being zero during calculations, thus ensuring the numerical stability of the model. For example, it can be set to a value of 0.001 or smaller.

[0100] Frequency adjustment sensitivity It is a parameter greater than 0, used to control the response speed and amplitude of the pulse frequency to the difference between the predicted and target values ​​of the cutting quality. When When the value is larger, the frequency adjustment will be more aggressive, and the response to quality deviations will be faster; when When the value is smaller, the adjustment will be more gradual. This parameter can be set empirically based on the actual application scenario and system stability requirements, or it can be learned through optimization algorithms.

[0101] Actual target pulse frequency This refers to the pulse frequency setting value obtained after normalization calculation and then inverse normalization, which can be directly applied to the laser cutting device 3. This value is the final frequency command executed by the system, ensuring that the laser cutting device 3 cuts at the optimal frequency.

[0102] Current pulse frequency setting upper limit and lower limit value These represent the maximum and minimum pulse frequencies that the laser cutting device 3 is allowed to set during operation. These limits are typically determined by the physical characteristics of the laser, the processing characteristics of the fabric, and safety regulations, and are used to ensure that the equipment operates within safe and efficient limits.

[0103] The solution proposed in this application achieves precise and real-time adjustment of the pulse frequency of the laser cutting device by constructing an intelligent pulse frequency dynamic adjustment model. The core of this model lies in utilizing predicted cutting quality values. Cutting quality benchmark value The difference between them, combined with the current pulse frequency normalized value The normalized value of the target pulse frequency is calculated using a power-law relationship. Specifically, when the predicted cutting quality value Below the target cutting quality benchmark At that time, the model will calculate a relatively high value. The aim is to improve cutting quality by increasing the pulse frequency; conversely, if... Higher than or close to ,but The settings will be adjusted accordingly to maintain or optimize the cutting effect. This includes adjusting the frequency sensitivity. The small constant that determines the strength of this adjustment and prevents division by zero This ensures the stability of the calculation process. The normalized target pulse frequency is obtained. Subsequently, the model further sets an upper limit value using the current pulse frequency through linear interpolation. and lower limit value Inversely normalize it to the actual target pulse frequency. This conversion process ensures the final output pulse frequency value. The system always operates within the safe operating range allowed by the equipment, avoiding equipment damage or deterioration in cutting quality due to excessively high or low operating frequencies. This solution, closely integrated with the aforementioned adaptive cutting quality prediction module, forms a highly efficient closed-loop control mechanism. The adaptive cutting quality prediction module comprehensively considers fabric conditions, equipment health, and current process parameters, outputting a forward-looking prediction of cutting quality. The intelligent pulse frequency dynamic adjustment module uses this predicted value as input to actively adjust the pulse frequency, rather than passively responding to existing quality problems. This strategy, combining feedforward and feedback, enables the equipment to make predictive adjustments based on future cutting quality trends. This allows it to maintain stable cutting quality even when fabric characteristics fluctuate or equipment status drifts, significantly improving the system's adaptability and intelligence.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fabric automatic slitting carbon dioxide laser tube apparatus, comprising: A cutting table (1), on which a guide frame (2) is provided, and on which a laser cutting device (3) movable along the width direction of the cutting table is provided, and on which the cutting table (1) a belt conveyor (4) for conveying fabric is also provided, and a row of guide rollers (5) is provided above the belt conveyor (4); characterized in that it further includes an adaptive adjustment system, which includes: The fabric condition comprehensive evaluation module constructs a comprehensive evaluation model for fabric conditions based on the fabric flatness index and real-time fabric tension, and outputs the comprehensive fabric condition index. The equipment health assessment module constructs an equipment health assessment model based on the measured output power of the laser cutting device (3), the actual laser pulse frequency, the actual acceleration of the axis motion, and the offset of the focused spot position, and outputs the equipment health index. The online detection and quantification module for cutting quality outputs a comprehensive cutting quality index for the previous period by constructing an online quantitative evaluation model for cutting quality. The adaptive cutting quality prediction module constructs an adaptive cutting quality prediction model based on the previous cycle's comprehensive cutting quality index, comprehensive fabric condition index, equipment health index, current laser power setting, and previous cutting speed setting, and outputs the predicted cutting quality value. The intelligent pulse frequency dynamic adjustment module constructs an intelligent pulse frequency dynamic adjustment model based on the predicted cutting quality value, the target cutting quality benchmark value, and the current pulse frequency setting value. It outputs the target pulse frequency setting value and executes frequency adjustment control through the controller.

2. The automatic fabric slitting carbon dioxide laser tube equipment according to claim 1, characterized in that, In the comprehensive evaluation model of fabric working conditions: The comprehensive index of fabric working conditions is obtained by multiplying the normalized fabric smoothness index and fabric tension index by the corresponding positive influence coefficients and summing them, and then using the sum as a variable to input an S-shaped exponential function for calculation and output. The fabric smoothness index and fabric tension index are obtained by normalizing the current smoothness and current fabric tension by substituting them into the maximum-minimum normalization formula.

3. The automatic fabric slitting carbon dioxide laser tube device according to claim 1, characterized in that, In the equipment health assessment model: The equipment health index is obtained by multiplying the normalized output power index, pulse frequency index, acceleration index, and (1 minus the focused spot position offset index) by the corresponding preset positive influence coefficients and then summing them up, where the sum of each influence coefficient is 1. The output power index, pulse frequency index, acceleration index, and focused spot position offset index are obtained by substituting the measured output power, actual pulse frequency, actual acceleration of axis motion, and focused spot position offset of the laser cutting device (3) into the maximum-minimum value normalization formula for normalization.

4. The automatic fabric slitting carbon dioxide laser tube device according to claim 1, characterized in that, The comprehensive cutting quality index of the previous cycle is calculated as follows: based on the uniformity coefficient of the kerf width, the carbonization index of the cutting edge, and the burr density per unit length after the previous cycle, an online quantitative evaluation model of cutting quality is constructed, and the comprehensive cutting quality index of the previous cycle is calculated and output.

5. The automatic fabric slitting carbon dioxide laser tube device according to claim 4, characterized in that, In the online quantitative evaluation model of cutting quality: The previous cycle's cutting quality comprehensive index is obtained by multiplying the kerf width uniformity index by an exponential decay function with the cutting edge carbonization index and the burr density per unit length as variables, where both influence coefficients are positive. The uniformity index of kerf width, the carbonization index of cutting edge, and the burr density index per unit length are obtained by substituting the uniformity coefficient of kerf width, the carbonization index of cutting edge, and the burr density per unit length after the previous cutting cycle into the maximum-minimum normalization formula for normalization.

6. The automatic fabric slitting carbon dioxide laser tube device according to claim 1, characterized in that, In the adaptive cutting quality prediction model: The predicted cutting quality value is obtained by multiplying the weighted indexes of the previous cycle's cutting quality comprehensive index, fabric condition comprehensive index, equipment health index, and (the ratio of the current set laser power index to the current set cutting speed index) together. All weighting coefficients are positive numbers. The current set laser power index and the current set cutting speed index are obtained by substituting the current set cutting speed and the current set laser power into the maximum-minimum value normalization formula for normalization.

7. The automatic fabric slitting carbon dioxide laser tube device according to claim 1, characterized in that, In the intelligent pulse frequency dynamic adjustment model: The predicted cutting quality value is compared with the target cutting quality benchmark value, and an exponential adjustment is performed based on the current pulse frequency setting value to obtain the normalized value of the target pulse frequency. This value is then mapped to the preset upper and lower frequency limits to output the actual target pulse frequency.

Images