A paper cutting method and system for a cutting machine

Abstract

The application discloses a paper cutting method and system for a cutting machine, relates to the technical field of cutting and automation equipment, and realizes the change of the cutting system from blind execution to visual servoing based on a dynamic reference calibration and tool path correction mechanism of real-time modeling based on three-dimensional point clouds, three-dimensional point cloud data of the side of a paper stack is actively projected and collected by a side contour camera through a laser triangulation method, the three-dimensional spatial topological structure of the surface of the paper stack is reconstructed, the inclination angle of the whole paper stack relative to the ideal workbench reference surface is calculated, the inclination angle is converted into a dynamic perpendicularity compensation instruction, the attitude angle of the tool holder in the plane perpendicular to the tool holder can be adjusted in real time according to the dynamic perpendicularity compensation instruction during the cutting and falling process, the ideal vertical relationship with the surface of the paper is always maintained, and therefore the defects of cutting surface inclination and fuzzing caused by the inclination of the paper stack are fundamentally eliminated, and the cutting precision of each cutting is ensured.

Description

Technical Field

[0001] This invention relates to the field of cutting and automation equipment technology, specifically to a paper cutting method and system for a cutting machine. Background Technology

[0002] Paper cutting is an indispensable key process in printing, packaging, and post-printing processing. Its precision and efficiency directly determine the quality of the final product and the production cost. Traditional and current mainstream programmable paper cutters are essentially programmed in their operation mode. That is, the operator sets or calls the preset cutting program in the control system according to the production order, then places the paper stack on the worktable and starts the equipment to complete the operation according to the predetermined program. Although this mode is a significant improvement over the early pure mechanical control, its level of intelligence is limited and it cannot cope with the dynamic changes in the processing process. In the context of modern intelligent manufacturing that pursues high efficiency, low cost, and zero defects, its limitations are becoming increasingly prominent.

[0003] However, existing technologies have the following four inherent drawbacks: Cutting plan planning relies on human experience. Optimal layout is a complex combinatorial optimization problem, especially in multi-specification, small-batch orders. How to plan the cutting path to maximize material utilization and minimize the number of cuts is highly dependent on the operator's skills and experience. The lack of automated intelligent decision support leads to serious material waste and fluctuations in production efficiency.

[0004] The inability to make a final confirmation of the state of the blade path before the cutting action is performed poses a risk of blade damage, equipment failure, or even safety accidents due to foreign objects or large-area defects in the paper itself.

[0005] Cutting speed, pressure and other parameters are usually fixed values ​​or need to be set manually. They cannot be adaptively and finely adjusted according to variables such as paper basis weight, material, stiffness and environmental humidity, which affects the stability of the cut surface quality and aggravates the wear of the cutter.

[0006] In summary, existing paper cutting technology is essentially an open-loop, static control system, lacking sufficient intelligence, adaptability, and reliability to meet the demands of modern industry for flexible, precise, and unmanned production. Therefore, there is an urgent need for a control method that integrates real-time sensing, intelligent decision-making, and precise execution to fundamentally overcome these shortcomings. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a paper cutting method and system for a cutting machine. It can realize dynamic benchmark calibration and real-time blade path correction of the cutting system through real-time three-dimensional point cloud modeling. Based on side-view contour camera and laser triangulation technology, it actively collects three-dimensional point cloud data of the side of the paper stack, reconstructs its spatial topology, calculates the tilt angle of the paper stack relative to the worktable reference plane, and converts it into dynamic verticality compensation commands. This allows the cutting blade holder to finely adjust its posture in real time according to the commands during the falling process, always maintaining perpendicularity to the paper surface, fundamentally eliminating cutting defects caused by paper stack tilting, and ensuring the cutting accuracy of each cut.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, a paper cutting method for a cutting machine, the specific steps of which are as follows: S100: The visual perception module acquires visual images of the paper stack to be cut, identifies the actual physical outline size and corner position of the paper stack, and constructs a side height profile model of the paper stack that reflects the tilt. At the same time, it acquires paper material information and environmental parameters. Based on the actual physical outline dimensions of the paper stack and the order finished product size list obtained by S100, S200 and the central control module call the intelligent nesting optimization algorithm to generate an optimized cutting path scheme with the dual objectives of minimizing the total number of cuts and maximizing material utilization. The module then uses a virtual simulation module to perform a cut path pre-play of the optimized cutting path scheme. S300 and the central control module map and align the virtual coordinate system of the optimized cutting path scheme generated in S200 with the actual physical corner points of the paper stack identified in S100 to achieve automatic calibration of the cutting benchmark. Based on the tilt calculated by the paper stack side height profile model constructed in S100, dynamic verticality compensation instructions for the cutting blade descent path are generated. S400 and the central control module match the optimal cutting execution parameter set from the adaptive parameter library based on the paper material information and environmental parameters obtained by S100. Before each cut, the predetermined cutting area of ​​the cutting blade is scanned. After confirming the safety of foreign objects and paper status, the motion execution module is driven to perform the current cut in combination with the dynamic perpendicularity compensation command and the matched cutting execution parameters. S500: During the cutting process, the cross-section line scan camera acquires images of the completed cut surface, and the central control module analyzes the cross-section images to evaluate the cutting quality in real time.

[0009] Furthermore, the step of constructing the height profile model of the paper stack side in S100 is as follows: Images containing the sides of the paper stack and the reference surface of the workbench are obtained by using a side-view contour camera. Laser triangulation is used to obtain three-dimensional point cloud data of multiple sides of the paper stack. The height information of the paper stack outline is extracted from the three-dimensional point cloud data to form multiple height change curves, and the workbench reference line is extracted from the workbench reference surface. Based on the calibration parameters of the side-view contour camera, the pixel coordinates on the contour line are transformed to the world coordinate system, resulting in the set of actual height values ​​corresponding to each discrete sampling point along the length of the paper stack. ,in Representing the The height of the paper stack relative to the reference surface of the workbench at each sampling point; Based on the actual height value set Calculate the overall flatness error set of the paper stack; A linear fit is performed on the overall flatness error set to obtain a fitted straight line. The angle between this fitted straight line and the reference plane of the worktable is... This refers to the tilt angle of the paper stack.

[0010] Furthermore, in step S200, the intelligent nesting optimization algorithm employs a genetic algorithm, and its optimization steps are as follows: Encoding: Encode a cutting scheme into a chromosome, where each gene on the chromosome represents a cutting parameter, including the position coordinates of the cutting line, the cutting order, and the cutting direction; Initialize the population: Generate a population based on heuristic rules. The population of the initial cutting scheme ; Fitness assessment: Calculate the fitness value for each pruning scheme in the population. The fitness function ,in The total number of cuts required for the cutting scheme. To improve material utilization in the cutting scheme, and These are preset weighting coefficients used to balance the two optimization objectives of minimizing the number of cuts and maximizing material utilization. Selection, crossover, and mutation: based on fitness values The size of the population determines the pruning scheme for individuals to be selected from the current population to enter the next generation. The selected pruning scheme is then evaluated with probability. Performing crossover operations generates new solutions; in terms of probability Perform mutation operations to change the pruning parameters of the pruning scheme to increase population diversity; Iteration and Output: Repeat fitness evaluation and selection, crossover and mutation until the preset fitness convergence threshold is reached, and output the pruning scheme corresponding to the chromosome with the highest fitness as the optimized pruning path scheme.

[0011] Furthermore, in S200, the step of the virtual simulation module performing a pre-run of the scheme is as follows: Path Coordinate Conversion: Receive the optimized cutting path scheme generated by the intelligent nesting optimization algorithm, and convert the position, length, cutting order and direction of each cutting line in the optimized cutting path scheme in the virtual coordinate system of the raw material paper stack into a virtual cutting path of ordered spatial coordinate points that can be executed by the motion execution module. Physical process simulation: In a virtual environment, based on the coordinate sequence of the virtual cutting path, the simulated cutting blade holder, paper presser, and block shape are driven to perform motion simulation, presenting the cutting action, blade path trajectory, and the coordinated timing of each actuator for each cut. Output and Optimization: The output includes a virtual cutting path containing a complete coordinate sequence, cutting order, and associated process parameters, as well as the estimated time.

[0012] Furthermore, in step S300, when mapping and aligning the virtual coordinate system with the actual physical corner points, a coordinate system from the virtual clipping scheme is established. - To the actual physical coordinate system of the paper stack - The affine transformation relationship will be used to identify the actual corner points of the paper stack in S100. physical coordinates The mapping is to the corresponding virtual corner points in the virtual scheme generated by S200. coordinates The transformation matrix is ​​solved by using multiple corresponding points. This allows for any virtual blade position in the optimized cutting path scheme. Its calibrated coordinates in the actual coordinate system pass The calculations yield the translation, rotation, and scaling compensations for the entire cutting path, combined with the overall tilt of the paper stack calculated in S100. The central control module generates dynamic verticality compensation commands that change over time. The data is sent to the high-precision motion execution module, which dynamically adjusts the posture of the blade holder in the plane perpendicular to it during the descent of the cutting blade. This ensures that the blade remains theoretically perpendicular to the worktable surface as it contacts and penetrates the entire tilted paper stack, thereby obtaining a vertical cutting surface.

[0013] Furthermore, the S500 acquires a grayscale image of the cut surface using the cross-sectional line scan camera, performs filtering and enhancement preprocessing, extracts the grayscale contour line of the cut surface edge from the preprocessed image along the cutting direction, calculates the average roughness and maximum peak-valley height of the grayscale contour line, and compares the calculated average roughness and maximum peak-valley height with preset roughness and peak-valley height thresholds. When both the average roughness and maximum peak-valley height are below the threshold, the cutting quality is deemed acceptable; when either index exceeds the limit, a defect is deemed to exist.

[0014] On the other hand, a paper cutting system for a cutting machine includes: a central control module, a vision perception module, a motion execution module, and an information interaction module; The central control module integrates an intelligent planning unit for executing intelligent nesting optimization algorithms, a virtual simulation unit for performing toolpath pre-simulation, an adaptive parameter library for storing and matching cutting parameters, and an image processing and quality assessment unit for processing images and performing quality analysis. The visual perception module includes a global vision unit for acquiring a top view image of the paper stack to identify the contours and corners, a side view contour measurement unit for acquiring a side view image to construct a height contour model, a near-range safety detection unit for performing a safety scan before the knife is dropped, and a cross-sectional quality scanning unit for acquiring images of the cut surface. The motion execution module is used to receive and execute dynamic verticality compensation instructions and cutting parameters from the central control module to complete the physical cutting action. The information interaction module is used to interact with the system and the external environment. It includes a human-machine interface unit for inputting order parameters and receiving operation instructions, an environmental sensing unit for acquiring ambient temperature and humidity, and a material information reading unit for identifying paper material information.

[0015] Furthermore, the side profile measurement unit is a laser profilometer based on the principle of laser triangulation. Its light plane is set perpendicular to the side of the paper stack to be measured. It is used to acquire the three-dimensional point cloud data of the side edge of the paper stack in a non-contact manner, and send the data to the image processing and quality assessment unit of the central control module through the data bus to construct the height profile model of the side of the paper stack and calculate the tilt.

[0016] Furthermore, the near-field safety detection unit is a linear image sensor, which is triggered by the central control module to quickly scan the area under the blade before each cut, and sends the imaging data to the central control module in real time for analysis, so as to complete the safety confirmation of the foreign object and the paper status.

[0017] Compared with existing technologies, this paper cutting method for a cutting machine has the following advantages: I. This invention collects paper stack outline and tilt data in real time through a visual perception module, and automatically generates a cutting path scheme with the goal of minimizing the number of cuts and maximizing material utilization based on an intelligent nesting optimization algorithm. This realizes the transformation from relying on human experience to fully automatic intelligent decision-making, and significantly improves the intelligence level of the cutting system.

[0018] I. This invention matches the optimal cutting execution parameter set from an adaptive parameter library based on parameters such as paper material and ambient temperature and humidity, and performs a scan confirmation by a close-range safety detection unit before each cut, thereby achieving adaptive adjustment to different paper conditions and external conditions, ensuring that the cutting process is always in the optimal working condition.

[0019] Third, by using a cross-sectional line scanning camera to acquire and analyze the quality of the cut surface in real time, and by combining dynamic verticality compensation commands to adjust the blade holder posture in real time, the quality defects such as cross-sectional tilting and roughening caused by paper stack tilting are effectively avoided, thus realizing quality control in the cutting process and improving the reliability and cutting stability of the system.

[0020] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0022] Figure 1 This is a flowchart illustrating a paper cutting method for a cutting machine according to one embodiment of the present invention. Figure 2 This is a schematic diagram of a paper cutting system for a cutting machine. Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention; Figure 4 This is an operation flowchart of a paper cutting system for a cutting machine. Detailed Implementation

[0023] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a method for cutting paper for a cutting machine,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0025] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0026] To address the shortcomings of existing technologies, this invention first describes the application scenarios for paper cutting. It is primarily applied in printing, packaging, and post-press processing, particularly in modern intelligent cutting scenarios where high cutting accuracy, material utilization, and automation are required. In these scenarios, factors such as the flatness and tilt of the paper stack, material properties, and environmental temperature and humidity significantly affect cutting quality and efficiency. Traditional cutting methods rely on manual experience and fixed procedures, making it difficult to adapt to dynamically changing production conditions. This invention, through real-time visual perception, intelligent path planning, dynamic posture compensation, and adaptive parameter adjustment, aims to achieve fully automated paper cutting with high precision, high efficiency, and high reliability.

[0027] The paper cutting system for a cutting machine provided by this invention comprehensively collects the outline, posture and material information of the paper stack through a visual perception module. The central control module intelligently generates an optimized cutting plan and performs virtual pre-simulation based on this information. Through coordinate mapping and dynamic compensation, it achieves precise alignment between physical execution and virtual planning. The motion execution module completes the cutting action under safety confirmation and adaptive parameter control. Finally, online quality inspection ensures the cutting effect, forming an intelligent system from perception, decision-making, execution to feedback.

[0028] Specifically, such as Figure 1As shown, a paper cutting method for a cutting machine includes the following steps: S100: The visual perception module acquires visual images of the paper stack to be cut, identifies the actual physical outline size and corner position of the paper stack, and constructs a side height profile model of the paper stack that reflects the tilt. At the same time, it acquires paper material information and environmental parameters. Based on the actual physical outline dimensions of the paper stack and the order finished product size list obtained by S100, S200 and the central control module call the intelligent nesting optimization algorithm to generate an optimized cutting path scheme with the dual objectives of minimizing the total number of cuts and maximizing material utilization. The module then uses a virtual simulation module to perform a cut path pre-play of the optimized cutting path scheme. S300 and the central control module map and align the virtual coordinate system of the optimized cutting path scheme generated in S200 with the actual physical corner points of the paper stack identified in S100 to achieve automatic calibration of the cutting benchmark. Based on the tilt calculated by the paper stack side height profile model constructed in S100, dynamic verticality compensation instructions for the cutting blade descent path are generated. S400 and the central control module match the optimal cutting execution parameter set from the adaptive parameter library based on the paper material information and environmental parameters obtained by S100. Before each cut, the predetermined cutting area of ​​the cutting blade is scanned. After confirming the safety of foreign objects and paper status, the motion execution module is driven to perform the current cut in combination with the dynamic perpendicularity compensation command and the matched cutting execution parameters. S500: During the cutting process, the cross-section line scan camera acquires images of the completed cut surface, and the central control module analyzes the cross-section images to evaluate the cutting quality in real time.

[0029] like Figure 2 As shown, a paper cutting system for a cutting machine is applicable to the above-mentioned paper cutting method. The system includes: a central control module, a vision perception module, a motion execution module, and an information interaction module. The central control module integrates an intelligent planning unit for executing intelligent nesting optimization algorithms, a virtual simulation unit for performing toolpath pre-simulation, an adaptive parameter library for storing and matching cutting parameters, and an image processing and quality assessment unit for processing images and performing quality analysis. The visual perception module includes a global vision unit for acquiring a top view image of the paper stack to identify the contours and corners, a side view contour measurement unit for acquiring a side view image to construct a height contour model, a near-range safety detection unit for performing a safety scan before the knife is dropped, and a cross-sectional quality scanning unit for acquiring images of the cut surface. The motion execution module is used to receive and execute dynamic verticality compensation instructions and cutting parameters from the central control module to complete the physical cutting action. The information interaction module is used to interact with the system and the external environment. It includes a human-machine interface unit for inputting order parameters and receiving operation instructions, an environmental sensing unit for acquiring ambient temperature and humidity, and a material information reading unit for identifying paper material information.

[0030] like Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention, such as... Figure 3 As shown, the computer device may include the above-mentioned Figure 2 The illustration shows a paper cutting system for a cutting machine. Optionally, the computer device may include a processor.

[0031] Optionally, the computer device may also include memory and a machine-executable program.

[0032] The processor, memory, and machine-executable program can be connected via a communication bus.

[0033] In its specific implementation, S100 acquires visual images of the paper stack to be cut through a visual perception module, identifies the actual physical contour dimensions and corner positions of the paper stack, and constructs a side height contour model of the paper stack reflecting its tilt. Simultaneously, it acquires paper material information and environmental parameters. The visual perception module includes a global vision unit, a side contour measurement unit, a near-field safety detection unit, and a cross-sectional quality scanning unit, wherein: The global vision unit is fixed above the cutting machine and captures a top-down image of the paper stack vertically downwards. The image is processed by grayscale conversion, filtering, edge detection, etc., including converting the color image to grayscale to reduce data dimensionality and computational complexity, and using digital filtering algorithms to suppress noise interference introduced during image acquisition. Edge detection algorithms are used to extract the boundary features between the paper stack and the background in the image, and contour tracking technology is used to obtain the complete top-down outer contour of the paper stack. Based on this top-down outer contour, a corner detection algorithm is further used to locate the pixel coordinates of the four vertices (corner points) in the image. Combined with the pre-calibrated camera intrinsic and extrinsic parameters, the pixel coordinates of the corner points are transformed to the machine world coordinate system of the cutting machine through a coordinate transformation model. This allows us to obtain the precise length, width, and corner positions of the paper stack in the actual physical space, providing reliable reference data for the subsequent mapping and alignment of the virtual coordinate system and the actual physical coordinate system of the cutting path.

[0034] The side profile measurement unit uses a laser profilometer based on the principle of laser triangulation. The light plane of this unit is perpendicular to the side of the paper stack to be measured. The laser projects a linear structured light onto the side of the paper stack. The side profile camera receives the image of the deformed laser line modulated by the surface profile of the paper stack at a pre-calibrated angle. By performing triangulation on each pixel in the deformed image, the three-dimensional spatial coordinates of multiple discrete points of the paper stack side profile illuminated by the laser line are obtained, thereby generating a three-dimensional point cloud of the side to be measured. By sequentially scanning multiple sides of the paper stack, discrete but ordered three-dimensional point cloud data can be obtained. These three-dimensional point cloud data collectively characterize the topological structure and geometric shape of the overall outer surface of the paper stack in three-dimensional space, that is, the three-dimensional surface information of the entire paper stack.

[0035] The image processing and quality assessment unit of the central control module receives side-view point cloud data and separates the point set representing the reference plane of the workbench from the point cloud. , Given the three-dimensional coordinates of any point on the reference plane in the machine world coordinate system, the least squares plane fitting algorithm is used to fit this point set. A fitting process is performed to solve for the plane equation that best represents the ideal table reference plane. The standard form of this reference plane equation is: ,in The unit normal vector of the fitted plane Components, parameters As a constant term, the equation of the reference plane defines the spatial orientation of the ideal worktable surface in the machine coordinate system.

[0036] Extract the contour point set representing the upper edge of the paper stack from the point cloud data of each side. Calculate each point in the contour point set The directed distance (i.e., height value) to the reference plane. Its calculation formula is Along the stack of paper The contour is sampled at equal intervals along the axial direction to obtain a set of height values ​​corresponding to a series of sampling points. ,in Representing the The height of the paper stack relative to the reference surface of the workbench at each sampling point constitutes the height profile curve of the paper stack side.

[0037] To quantify the overall tilt state of the paper stack, the height set is... Its corresponding length direction coordinate set Linear fitting was performed using univariate linear regression analysis, and the equation of the fitted line was solved using the least squares method. ,in, The slope of the fitted line, The intercept represents the overall trend of the stack height along its length, and the stack's tilt is also considered. To fit the angle between the straight line and the reference horizontal plane, for the fitted straight line... Its slope on the height-length two-dimensional projection plane is... Inclination angle Calculated using the arctangent function .

[0038] Based on the actual physical outline dimensions of the paper stack and the order finished product size list obtained from S100, the S200 central control module calls the intelligent nesting optimization algorithm to generate an optimized cutting path scheme with the dual objectives of minimizing total cuts and maximizing material utilization. The virtual simulation module then performs a pre-simulation of the cut path scheme. The intelligent planning unit of the central control module executes the intelligent nesting optimization algorithm; in this embodiment, a genetic algorithm is preferred, and its optimization steps are as follows: Encoding: The cutting line position coordinates, cutting sequence and cutting direction of the paper stack are used as cutting parameters and encoded as a chromosome. Each chromosome uniquely corresponds to a cutting scheme for the actual physical outline size of the current paper stack and the actual finished product size of the order. Initialize the population: Based on the actual physical outline dimensions of the paper stack and the finished product size requirements of the order, generate a population containing multiple initial cutting schemes that meet production requirements; Fitness assessment: The fitness value of each cutting scheme in the population is calculated, and the fitness function has the dual optimization objectives of minimizing the total number of cuts and maximizing the material utilization rate. Selection, crossover, and mutation: Based on the fitness value, individuals with the best fitness from the current population are selected to enter the next generation. The selected pruning scheme is then processed probabilistically. Performing crossover operations generates new schemes with probability. Perform a mutation operation to adjust the position coordinates of the cutting line, the cutting order, and the direction. Iteration and Output: Repeat the fitness evaluation and selection, crossover and mutation process until the preset fitness convergence threshold is reached, and output the cutting scheme corresponding to the chromosome with the highest fitness as the optimized cutting path scheme to adapt to the current paper stack and order requirements.

[0039] The virtual simulation unit then pre-runs the optimized cutting path scheme, converting each cutting line in the scheme into a sequence of coordinate points traversed by the motion execution module in three-dimensional space, generating a virtual cutting path. In the computer-created virtual cutting environment, the paper stack 3D model, knife holder model, and paper presser model are driven to move according to the coordinate sequence and timing of the virtual cutting path, pre-running and simulating the complete cutting process.

[0040] After the preliminary simulation is successful, the virtual simulation module outputs the final executable cutting instruction package, which includes the precise coordinate sequence, cutting order, speed planning, and triggering times of each auxiliary action, and estimates the total time required to complete the scheme.

[0041] S300 maps and aligns the virtual coordinate system of the optimized cutting path scheme generated in S200 with the actual physical corner points of the paper stack identified in S100, thereby achieving automatic calibration of the cutting reference. Based on the tilt calculated by the side height profile model of the paper stack constructed in S100, it generates dynamic verticality compensation commands for the descent path of the cutting blade. Its optimized cutting path scheme is based on an ideal and standard virtual material coordinate system. The S100 identifies the actual corner points of the paper stack in the machine's physical coordinate system, as specified in the standard. In this context, the paper stacks may undergo translation, rotation, or even slight scaling. The central control module achieves alignment by establishing an affine transformation relationship between two coordinate systems, specifically: Select multiple actual physical corner points identified in S100 Their corresponding ideal corner points in the virtual scheme Multiple sets of corresponding points are constructed, and a transformation matrix is ​​obtained by using these corresponding points and algorithms such as the least squares method. For any blade position in the virtual cutting path Its corresponding calibrated execution coordinates in the actual machine coordinate system It is possible Calculations show that this process automatically compensates for positional deviations and angular rotations caused by the placement of paper stacks, achieving automatic calibration of the cutting reference based on the calculated tilt angle. And its direction, generate a dynamic verticality compensation command. This dynamic verticality compensation command is represented by a compensation curve, whose parameters include: paper stack tilt angle, compensation direction vector, maximum angular velocity, and angular acceleration limit. To ensure smooth and shock-free compensation motion, this dynamic verticality compensation command... During the cutting blade's descent, the servo controller... The attitude angle of the blade holder in the plane perpendicular to it is dynamically adjusted so that the blade remains perpendicular to the paper surface as it contacts and penetrates the entire tilted paper stack, thus eliminating quality defects such as tilted cutting surface and roughening caused by the tilt of the paper stack.

[0042] Based on the paper material information and environmental parameters obtained by S100, the S400 central control module matches the optimal cutting execution parameter set from the adaptive parameter library. Before each cut, it scans the predetermined cutting area of ​​the cutting blade to ensure the safety of foreign objects and paper status. Then, combining the dynamic perpendicularity compensation command and the matched cutting execution parameters, it drives the motion execution module to perform the current cut. The central control module fuses the physical coordinates of the current cut, the matched cutting execution parameter set, and the dynamic perpendicularity compensation command generated by S300, sending this as an executable command package to the motion execution module. The motion execution module parses and executes the executable command package, adjusting the pitch or roll posture of the blade holder in real time via the servo motor driving the blade holder rotation axis. Simultaneously, it performs force-position hybrid control based on the matched cutting speed and pressure parameters to ensure the blade penetrates the paper stack with optimal dynamics.

[0043] The central control module's adaptive parameter library pre-stores a vast array of optimal cutting parameter combinations for different paper basis weights, materials, number of layers, and ambient temperature and humidity. These include cutting speed, blade pressure, blade angle fine-tuning, paper clamp pressure, and lead time. Based on the current paper material and real-time ambient temperature and humidity read by the S100, the most suitable cutting execution parameter set is retrieved from the parameter library. Before each cut is executed, the central control module triggers a proximity safety detection unit. This sensor rapidly performs a line scan image of the narrow area below the predetermined blade trajectory. The image is transmitted back for analysis in real time to detect the presence of foreign objects such as metal debris, tools, large-area tears in the paper, or abnormal protrusions, as well as whether the paper edges are neat and free from severe curling. Only after safety confirmation is passed will the system allow the cutting to proceed; otherwise, an alarm is immediately triggered and the process is paused.

[0044] The motion execution module also receives instruction packets from the central control module. These instruction packets integrate: the current cutter position coordinates after coordinate mapping and alignment, dynamic verticality compensation instructions, and a set of matched adaptive cutting parameters. The motion execution module then completes the actions of moving and positioning, pressing down the paper pressure plate, and compensating for the falling cutting.

[0045] During the cutting process, the S500 uses a cross-sectional line scan camera to acquire images of the completed cut surface. The central control module analyzes the cross-sectional images to evaluate the cutting quality in real time. The cross-sectional quality scanning unit is installed at a suitable position behind or to the side of the cutter holder. Each time a cut is completed, the camera performs a line scan along the cutting direction on the cut surface of the paper. The cross-sectional line scan camera acquires the reflected light signal of the cross-section line by line along the cutting direction, converts the light signal into the corresponding pixel grayscale value, and enhances the difference in brightness between the cross-section and the background through grayscale calibration. Simultaneously, noise suppression processing is performed, and then the discrete pixel sequence is stitched together according to the scanning sequence to finally synthesize a high-contrast grayscale image of the cut surface that fully presents the outline details of the cut surface. The grayscale image of the cut surface is transmitted to the image processing and quality assessment unit of the central control module. The grayscale contour line of the cut edge is extracted along the cutting direction. By analyzing the undulation of this contour line, key quality indicators are calculated, including average roughness (the arithmetic mean of the absolute values ​​of the deviations of each point on the contour line from the average line, reflecting the overall smoothness of the cut surface) and maximum peak-valley height (the vertical distance between the highest and lowest peaks within a single sampling length, reflecting the maximum burr or tear depth of the cut surface). The calculated average roughness and maximum peak-valley height values ​​are compared with preset quality thresholds. If both are below the threshold, the cutting quality of that cut is considered acceptable. If the average roughness is greater than the preset threshold or the maximum peak-valley height is greater than the preset threshold, a defect is identified. The system can record the location and type of the defect in real time and can trigger alarms, speed reduction, or maintenance notifications according to preset strategies.

[0046] like Figure 3 As shown, the specific steps of the paper cutting method for a cutting machine according to the present invention are as follows: (1) System startup and preparation The operator places the stack of paper to be cut on the workbench.

[0047] The finished product size list for the current order can be entered or retrieved through the human-machine interface.

[0048] (2) Visual acquisition and paper stack modeling The camera was used to take an image of the top of the paper stack from above, and the outline of the top of the paper stack and the positions of its four corner points were identified.

[0049] A laser profilometer scans the paper stack from the side to acquire three-dimensional point cloud data from multiple sides.

[0050] The height of each position of the paper stack is calculated based on the point cloud data, and a height profile model is constructed.

[0051] Analyze the height data to calculate the overall tilt angle of the paper stack.

[0052] Simultaneously, the paper material label is read and the ambient temperature and humidity are obtained.

[0053] (3) Intelligent cutting scheme generation The intelligent layout algorithm is automatically invoked based on the actual size of the paper stack and order requirements.

[0054] With the goal of minimizing the total number of cuts and maximizing material utilization, multiple cutting schemes are generated.

[0055] The optimal cutting path scheme is selected through iterative optimization.

[0056] The scheme is rehearsed in a virtual simulation environment to check for any motion interference or timing conflicts.

[0057] (4) Coordinate mapping and dynamic compensation The coordinate system of the virtual cutting scheme is mapped and aligned with the corner points of the actual paper stack.

[0058] Based on the tilt angle of the paper stack, a dynamic verticality compensation command is generated.

[0059] This command will cause the cutting blade to automatically adjust its posture during the descent, keeping it perpendicular to the paper.

[0060] (5) Parameter matching and safety confirmation Based on the paper material and environmental parameters, it automatically matches the optimal parameters such as cutting speed and pressure.

[0061] Before each cut, the linear sensor quickly scans the cut area.

[0062] Cutting can only be carried out after confirming that there are no foreign objects and the paper is free of abnormalities.

[0063] (6) Perform cutting and quality inspection Perform the trimming based on the mapped coordinates, compensation instructions, and matching parameters.

[0064] After each cut is completed, the cross-section line scan camera immediately captures an image of the cut surface.

[0065] Analyze the images and calculate quality indicators such as cross-sectional roughness.

[0066] If the quality is acceptable, proceed to the next cut; if it is unacceptable, trigger an alarm and record the error.

[0067] (7) Task completion and data archiving Once all cropping is complete, the system will indicate that the task is finished.

[0068] Save the process data and quality report of this cutting to the database.

[0069] It can output production reports, including material utilization rate, time taken, quality statistics, etc.

[0070] In summary, through the close integration and synergy of the above steps, this invention constructs a complete intelligent cutting control system encompassing multi-dimensional perception, intelligent optimization decision-making, dynamic and precise compensation, safe and adaptive execution, and online quality feedback. This system collectively ensures the superior performance of the cutting system in terms of high efficiency, high material utilization, high safety, and high stability, providing an effective solution for the intelligent upgrading of the modern printing and packaging industry.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A paper cutting method for a cutting machine, characterized in that, The steps of this method are as follows: S100: The visual perception module acquires visual images of the paper stack to be cut, identifies the actual physical contour size and corner points of the paper stack, and constructs a side height contour model of the paper stack that reflects the tilt of the paper stack, and obtains paper material information and environmental parameters. S200: Based on the actual physical outline dimensions of the paper stack and the actual order finished product size list, the intelligent nesting optimization algorithm is called to generate an optimized cutting path scheme with the dual objectives of minimizing the total number of cuts and maximizing material utilization. The toolpath of the optimized cutting path scheme is then pre-simulated using a virtual simulation module. S300: Map and align the virtual coordinate system of the optimized cutting path scheme generated in S200 with the actual physical corner points of the paper stack. Based on the paper stack tilt reflected by the paper stack side height profile model constructed in S100, generate a dynamic verticality compensation command for the cutting blade descent path. S400. Based on the paper material information and environmental parameters, the optimal cutting execution parameter set is matched from the adaptive parameter library. Before each cut is executed, the predetermined cutting area of ​​the cutting blade is scanned. Combined with the dynamic perpendicularity compensation command and the matched cutting execution parameters, the motion execution module is driven to execute the current cut. S500: During the cutting process, the cross-section line scan camera acquires images of the completed cut surface and analyzes the cross-section images to evaluate the cutting quality in real time.

2. The paper cutting method for a cutting machine according to claim 1, characterized in that, The steps for constructing the side height profile model of the paper stack in S100 are as follows: Images containing the sides of the paper stack and the reference surface of the workbench are obtained by using a side-view contour camera. Laser triangulation is used to obtain three-dimensional point cloud data of multiple sides of the paper stack. The height information of the paper stack outline is extracted from the three-dimensional point cloud data to form multiple height change curves, and the workbench reference line is extracted from the workbench reference surface. Based on the calibration parameters of the side-view contour camera, the pixel coordinates on the contour line are transformed to the world coordinate system to obtain the set of actual height values ​​corresponding to each discrete sampling point along the length of the paper stack. Based on the set of actual height values, calculate the overall flatness error set of the paper stack; A linear fit is performed on the overall flatness error set to obtain a fitted straight line, and the angle between the fitted straight line and the reference surface of the workbench is taken as the inclination of the paper stack.

3. The paper cutting method for a cutting machine according to claim 1, characterized in that, The step of calling the intelligent nesting optimization algorithm to generate an optimized cutting path scheme with the dual objectives of minimizing the total number of cuts and maximizing material utilization includes: In S200, the intelligent nesting optimization algorithm adopts a genetic algorithm, and its optimization steps are as follows: Encoding: The cutting line position coordinates, cutting sequence and cutting direction of the paper stack are used as cutting parameters and encoded as a chromosome. Each chromosome uniquely corresponds to a cutting scheme for the actual physical outline size of the current paper stack and the actual order finished product size list. Initialize the population: Based on the actual physical outline dimensions of the paper stack and the finished product size requirements of the order, generate a population containing multiple initial cutting schemes that meet production requirements; Fitness assessment: The fitness value of each cutting scheme in the population is calculated, and the fitness function has the dual optimization objectives of minimizing the total number of cuts and maximizing the material utilization rate. Selection, crossover, and mutation: Based on the fitness value, individuals with the best fitness from the current population are selected to enter the next generation. The selected pruning scheme is then processed probabilistically. Performing crossover operations generates new schemes with probability. Perform a mutation operation to adjust the position coordinates of the cutting line, the cutting order, and the direction. Iteration and Output: Repeat the fitness evaluation and selection, crossover and mutation process until the preset fitness convergence threshold is reached, and output the cutting scheme corresponding to the chromosome with the highest fitness as the optimized cutting path scheme to adapt to the current paper stack and order requirements.

4. A paper cutting method for a cutting machine according to claim 3, characterized in that, In step S200, the virtual simulation module performs a pre-run of the scheme as follows: Receive the optimized cutting path scheme, parse the position, length, cutting order and direction of each cutting line in the optimized cutting path scheme in the virtual coordinate system of the raw material paper stack, and convert it into a virtual cutting path of ordered spatial coordinate points that can be executed by the motion execution module; In the virtual environment, based on the coordinate sequence of the virtual cutting path, the simulated cutting blade holder, paper presser, and block shape are driven to perform motion simulation, presenting the cutting action, blade path trajectory, and the coordinated timing of each actuator for each cut. The output includes a virtual cutting path containing a complete coordinate sequence, cutting order, and associated process parameters, as well as the estimated cutting time.

5. A paper cutting method for a cutting machine according to claim 1, characterized in that, In S300, the process of mapping and aligning the virtual coordinate system with the actual physical corner points is as follows: By establishing the coordinate system of the virtual clipping scheme - To the actual physical coordinate system of the paper stack - affine transformation relations; The actual corner points of the paper stack identified in S100 physical coordinates The mapping is to the corresponding virtual corner points in the virtual scheme generated by S200. coordinates This yields virtual corner points corresponding to the actual corner points of multiple paper stacks; Solve for the transformation matrix of the virtual corner points corresponding to the actual corner points of the paper stack. This allows for any virtual blade position in the optimized cutting path scheme. Its calibrated coordinates in the actual coordinate system are This allows us to express the translation, rotation, and scaling compensation of the entire cutting path; Based on the overall tilt of the paper stack calculated in S100, the central control module generates a dynamic verticality compensation command that changes over time and sends it to the high-precision motion execution module to dynamically adjust the posture of the blade holder in the plane perpendicular to it during the descent of the cutting blade.

6. A paper cutting method for a cutting machine according to claim 1, characterized in that, The S500 acquires a grayscale image of the cut surface using the cross-sectional line scan camera, and preprocesses the grayscale image, including filtering and enhancement operations. From the preprocessed grayscale image, it extracts the grayscale contour line of the cut surface edge along the cutting direction, calculates the average roughness and maximum peak-valley height of the grayscale contour line, and then calculates the average roughness. and maximum peak and valley height With the preset roughness Peak and valley height thresholds When comparing, and When the cutting quality is deemed acceptable, or If so, it is determined that there is a defect.

7. A paper cutting system for a cutting machine, applicable to the paper cutting method for a cutting machine as described in any one of claims 1-6, characterized in that, The system includes: a central control module, a visual perception module, a motion execution module, and an information interaction module; The central control module integrates an intelligent planning unit for executing intelligent nesting optimization algorithms, a virtual simulation unit for performing toolpath pre-simulation, an adaptive parameter library for storing and matching cutting parameters, and an image processing and quality assessment unit for processing images and performing quality analysis. The visual perception module includes a global vision unit for acquiring a top view image of the paper stack to identify the contours and corners, a side view contour measurement unit for acquiring a side view image to construct a height contour model, a near-range safety detection unit for performing a safety scan before the knife is dropped, and a cross-sectional quality scanning unit for acquiring images of the cut surface. The motion execution module is used to receive and execute dynamic verticality compensation instructions and cutting parameters from the central control module to complete the physical cutting action. The information interaction module is used to interact with the system and the external environment. It includes a human-machine interface unit for inputting order parameters and receiving operation instructions, an environmental sensing unit for acquiring ambient temperature and humidity, and a material information reading unit for identifying paper material information.

8. A paper cutting system for a cutting machine according to claim 7, characterized in that, The side profile measurement unit is a laser profilometer based on the principle of laser triangulation. Its light plane is set perpendicular to the side of the paper stack to be measured. It is used to acquire three-dimensional point cloud data of the side edge of the paper stack in a non-contact manner, and send the data to the image processing and quality assessment unit of the central control module through the data bus to construct the height profile model of the side of the paper stack and calculate the tilt.

9. A paper cutting system for a cutting machine according to claim 7, characterized in that, The near-field safety detection unit is a linear image sensor, which is triggered by the central control module to quickly scan the area under the blade before each cut, and sends the imaging data to the central control module in real time for analysis to complete the safety confirmation of the foreign object and the paper status.

Images