Mylar self-adaptive cutting method
By using an adaptive cutting method with an air suspension platform and laser cutting head, the problems of precision and efficiency in Mylar film cutting were solved, achieving non-contact processing and static electricity elimination, thus improving the precision and automation level of Mylar film cutting.
Patent Information
- Application Number
- CN202511917322.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-18
- Publication Date
- 2026-02-13
AI Technical Summary
Existing Mylar sheet cutting technology faces challenges such as difficulty in precision control, material deformation, electrostatic adsorption, and thermal management, making it unable to adapt to changes in product size and resulting in low processing efficiency.
An adaptive cutting method combining an air suspension platform and a laser cutting head is adopted. The cutting path is generated by acquiring product contour data in real time, the laser focal length is dynamically adjusted, non-contact separation is achieved by utilizing air pressure difference, electrostatic adsorption is eliminated by combining ionized gas flow, and thermodynamic parameters are monitored in real time to optimize the cutting strategy.
It achieves sub-millimeter-level processing precision, avoids material deformation and electrostatic adsorption, improves cutting efficiency and automation level, and ensures the continuity and quality stability of the cutting process.
Smart Images

Figure CN121514716A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the field of precision processing of flexible materials, in particular to a Mylar sheet self-adaptive cutting method. BACKGROUND
[0002] Current Mylar sheet automatic cutting technology faces many technical bottlenecks, which seriously restricts the development of precision electronic component mounting process. In terms of precision control, the traditional cutting equipment adopts the mode of preset program driving mechanical cutter head, which has inherent defects. When the product size changes, the parameters need to be manually recalibrated, and the inevitable gap error of the mechanical transmission system makes it difficult to break through the technical barrier of ±0.5mm cutting precision. More importantly, the existing system lacks real-time edge detection and path compensation capability, and cannot cope with the position offset problem of the product on the work station.
[0003] The challenge of material processing link is particularly prominent. The unique ultra-thin and flexible characteristics of Mylar sheet make it prone to deformation during processing. The traditional pneumatic pressing plate fixing scheme is in a dilemma: insufficient pressure leads to material slip, and excessive pressure causes irreversible plastic deformation. At the same time, the accumulated static charge in the processing process makes the cut sheet stubbornly adsorbed on the surface of the equipment, seriously damaging the production continuity. The wrinkle problem in the conveying process is also troublesome, and the existing mechanical flattening mechanism has limited correction effect on serious wrinkles.
[0004] The third major technical obstacle is the heat management problem. The high temperature generated by traditional mechanical cutting not only causes material melting and adhesion, but also accelerates tool wear. The heat dissipation efficiency and energy consumption ratio of air cooling scheme are difficult to balance, and the material displacement problem during multi-layer cutting still lacks effective solutions. These systematic defects collectively result in low comprehensive efficiency of existing Mylar sheet cutting equipment, which is difficult to meet the stringent requirements of modern electronic manufacturing for precision mounting materials. In view of the above problems, the existing technology needs to be improved. SUMMARY
[0005] (I) Technical problems to be solved To solve the above problems, the application proposes a Mylar sheet self-adaptive cutting method, which aims to solve the problem that the traditional cutting equipment in the prior art relies on preset program to execute fixed path, and cannot adapt to the dynamic change of product size, resulting in frequent shutdown and parameter adjustment.
[0006] (II) Technical scheme The application provides a melamine sheet adaptive cutting method, and the technical scheme is as follows: obtaining contour data of a product to be attached, generating a cutting path based on the contour data, and the contour data containing a product boundary coordinate set; the cutting path is enlarged by a preset edge allowance amount to the product boundary in proportion, the melamine sheet is transported to an air suspension platform, the laminar air cushion sprayed by the multi-hole ceramic plate makes the melamine sheet in a suspended state and keeps the suspended state to be transmitted to a cutting station; a laser cutting head is used to scan and cut along the cutting path, and the laser focal length is dynamically adjusted according to real-time ranging data; after cutting is completed, a negative pressure adsorption array is activated to separate finished products and waste materials.
[0007] Further, the application also provides that obtaining the contour information comprises: capturing a product surface image through an image acquisition device, performing edge feature extraction on the surface image to obtain a boundary coordinate set, and converting the boundary coordinate set into physical coordinate system data.
[0008] Further, the application also provides that generating the target cutting path comprises: determining an edge allowance compensation parameter according to a material attribute database, performing equidistant offset processing on the contour information, and generating a transition structure with continuous curvature at a path turning point, wherein the offset amount is determined by the edge allowance compensation parameter.
[0009] Further, the application also provides that the control of the non-contact suspension platform comprises: applying a decreasing air pressure distribution along the conveying direction, and synchronously applying ionized gas flow to inhibit electrostatic adsorption effect.
[0010] Further, the application also provides that the dynamic focusing cutting comprises: monitoring the distance change between the melamine sheet surface and the laser cutting head in real time, dynamically adjusting the laser beam focusing parameter according to the distance change amount, and maintaining the spot size of the laser action point constant.
[0011] Further, the application also provides that the separation based on the air pressure difference control comprises: applying directional negative pressure adsorption in the cutting completion area; applying positive air flow pushing in the non-cutting area; and realizing physical separation through the air pressure action difference.
[0012] Further, the application also provides that the application further comprises: collecting thermodynamic parameters in a cutting process in real time, adjusting laser processing parameters when the thermodynamic parameters exceed a preset threshold, feeding processing data to an optimization model to update a cutting strategy.
[0013] Further, the application also provides a melamine sheet adaptive cutting device, comprising: a visual positioning unit configured to obtain and process product contour data; a path planning unit configured to generate a cutting path with material compensation; an air film conveying module configured to provide a controllable suspension force field; a laser processing unit configured to perform a focus adaptive cutting operation; and a sorting unit configured to realize material separation through an air pressure gradient.
[0014] Further, the application also proposes that the visual positioning unit comprises an anti-interference optical assembly for eliminating surface reflection noise, and a real-time coordinate conversion module for establishing a mapping relationship between image coordinates and physical coordinates.
[0015] Further, the application also proposes that the air film conveying module further comprises an air pressure gradient controller configured to generate a spatially variable air pressure distribution, and an electrostatic elimination assembly configured to maintain material surface charge balance.
[0016] (Three) beneficial effects Compared with the prior art, the beneficial effects of the application are: In the application, the material is kept flat by air suspension transmission, dynamic focusing control ensures cutting precision, and air pressure differential separation avoids electrostatic adsorption, realizing the cooperative control of sub-millimeter level machining precision and zero physical contact, and having the technical effects of improving cutting precision, avoiding material deformation and eliminating electrostatic adsorption. BRIEF DESCRIPTION OF DRAWINGS
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.
[0018] Figure 1 It is a whole logic schematic diagram of the Mylar sheet adaptive cutting method. Figure 2 It is a three-dimensional structure schematic diagram of the Mylar sheet pasting equipment. Figure 3 It is an adaptive cutting logic framework schematic diagram of the Mylar sheet pasting equipment.
[0019] 10, visual positioning unit, 20, path planning unit, 30, air film conveying module, 40, laser processing unit, 50, sorting unit. DETAILED DESCRIPTION
[0020] The technical solutions in the present application will be clearly and completely described below with reference to the drawings in the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application. It should be noted that: similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used for differentiation, and cannot be understood as indicating or implying relative importance.
[0021] In the prior art, the Mylar sheet cutting process has long been faced with the dual challenges of precision and efficiency. Traditional cutting equipment relies on preset programs to execute fixed paths, which cannot adapt to dynamic changes in product size, resulting in frequent downtime to adjust parameters. Mechanical clamping devices are prone to deformation when fixing materials, and the size deviation caused by springback after cutting. The separation process relies on manual intervention, and the degree of automation is low. In the scenario of mounting precise components in consumer electronics, the product iteration speed is accelerating, and the existing technology is difficult to meet the production needs of rapid switching of multiple varieties.
[0022] To solve the above problems, researchers found that material deformation is caused by contact fixing method, and tried to use non-contact support scheme but faced the problem of suspension stability. It is observed that laser cutting has non-contact characteristics, but the traditional fixed focal length mode cannot adapt to the fluctuation of materials. By analyzing the complementarity of air suspension platform and laser ranging, a cooperative control mechanism of dynamic focusing and suspension transmission is proposed. For the efficiency problem of finished product separation, it is found that air pressure difference can replace mechanical ejector pin to form a non-destructive sorting scheme.
[0023] Therefore, the present application proposes a Mylar sheet adaptive cutting method, comprising: S100, acquiring contour data of a product to be mounted, generating a cutting path based on the contour data, and the contour data containing a product boundary coordinate set; S200, the cutting path is enlarged by a preset edge amount to the product boundary, the Mylar sheet is transported to the air suspension platform, and the laminar air cushion sprayed by the multi-hole ceramic plate makes the Mylar sheet in a suspended state and keeps the suspended state to be transmitted to the cutting station; S300, a laser cutting head is used to scan and cut along the cutting path, and the laser focal length is dynamically adjusted according to the real-time ranging data; S400: After cutting, activate the negative pressure adsorption array to separate the finished product from the waste.
[0024] Contour data refers to the geometric information of a product's shape extracted through image processing techniques. Specifically, this can be achieved using a high-resolution industrial camera combined with an edge detection algorithm to generate cutting paths that match the actual dimensions of the product. The edge detection algorithm employs the Canny operator, setting a low threshold of 50 and a high threshold of 150. In the preprocessing stage, a 3×3 Gaussian filter is applied to filter image noise, where the standard deviation of the Gaussian filter is... After edge extraction, morphological closing operations are used to fill the edge gaps, with the structuring element being a 3×3 rectangle. The final output is a continuous set of product boundary coordinates. The equidistant offset algorithm determines the offset direction based on the concavity / convexity of the product contour, achieved by calculating the curvature sign of the contour—counterclockwise offset for positive curvature (convex surface), and clockwise offset for negative curvature (concave surface). If path intersections occur during the offset process, the criterion is that the distance between the two path segments is <0.05mm. The intersection segment is deleted, and a third-order Bézier curve is used to refit the transition path. The control points of the Bézier curve are selected from two contour feature points on each side of the intersection point to ensure the curvature continuity of the transition segment with the original path.
[0025] Cutting path generation refers to the proportional expansion based on the product boundary, which can be achieved using a vector offset algorithm to ensure precise edge trimming allowance on the cut Mylar sheets. The air suspension platform utilizes gas pressure to create non-contact support, typically achieved using a porous ceramic plate combined with laminar gas jets, eliminating material deformation caused by mechanical clamping. Dynamic laser focal length adjustment adjusts the focusing position in real-time based on changes in material surface height, achieved through a laser rangefinder and servo motor linkage, maintaining a constant cutting spot size. The negative pressure adsorption array is a separation device composed of multiple independently controlled air pressure units, typically achieved using a vacuum generator and solenoid valves, selectively adsorbing and sorting finished products.
[0026] Product contour data is acquired in real time by a vision system and converted into physical coordinates, automatically compensating for edge allowance when generating the cutting path. After the Mylar sheet enters the air suspension platform, the laminar air cushion forms a uniform support force field, avoiding localized stress concentration caused by traditional pressure plates. During laser cutting, a ranging device continuously monitors the material surface height and adjusts the focusing lens position through a closed-loop control system. After cutting, the negative pressure adsorption array generates adsorption force only in the finished product area, while a reverse airflow is applied to the waste area to achieve physical separation.
[0027] Traditional mechanical cutting relies on rigid clamping, which can cause material deformation. This solution uses an air-suspended platform to achieve non-contact transmission, combined with dynamically focused laser cutting to eliminate contact stress. Existing equipment requires manual adjustment of the cutting path; this solution automatically generates a compensation path based on real-time contour data, adapting to changes in product size. Traditional separation processes using mechanical ejectors can easily damage materials; this solution achieves non-destructive sorting through air pressure differential control.
[0028] This application effectively solves the problem of dimensional inaccuracies caused by path adjustment lag, avoids material deformation caused by mechanical contact, and achieves stable control of cutting accuracy. The synergistic effect of non-contact transmission and dynamic focusing ensures the processing integrity of ultra-thin materials, and the air pressure difference separation mechanism significantly improves sorting efficiency and automation level.
[0029] This application further proposes to obtain contour information by capturing product surface images through an image acquisition device, extracting edge features from the surface images, obtaining a set of boundary coordinates, and converting the set of boundary coordinates into physical coordinate system data.
[0030] Image acquisition devices are used to capture optical information about a product's surface. Specifically, this can be achieved using a high-resolution industrial camera paired with a ring-polarized light source, ensuring image clarity by suppressing surface reflections. Edge feature extraction refers to the process of identifying the actual boundaries of the product from the image. This can be achieved using computer vision algorithms combined with gradient detection operators, precisely determining contour inflection points through pixel-level analysis. Boundary coordinate sets refer to the digital positional information generated from edge feature extraction. This can be stored as a two-dimensional coordinate point sequence format, providing the foundational data for subsequent path generation. Physical coordinate system data refers to the coordinate system corresponding to the spatial position of the cutting equipment. This can be achieved using coordinate mapping algorithms combined with lens distortion correction parameters, eliminating the cumulative errors between the optical system and mechanical positioning.
[0031] After the product is placed at the processing station, the image acquisition device captures surface images at a preset frequency, and a ring-polarized light source eliminates the impact of material reflection on image quality. The edge feature extraction module performs pixel-by-pixel analysis of the image, identifies the actual contour boundaries of the product through gradient changes, and generates a coordinate set containing all contour inflection points. The coordinate transformation module converts the boundary point data in the image coordinate system, combined with lens distortion parameters and mechanical positioning reference points, into three-dimensional coordinate data in the physical coordinate system of the cutting equipment. This transformation process is achieved through a real-time coordinate mapping algorithm. When the product shifts or rotates at the station, the system automatically generates a transformation matrix to correct the coordinate data, ensuring that the cutting path accurately matches the actual product contour.
[0032] Traditional cutting equipment relies on preset programs to execute fixed cutting paths, which cannot perceive actual position changes of products, resulting in mismatch between cutting areas and product contours. The scheme establishes a closed-loop control mechanism of actual product contours and cutting paths through dynamic image acquisition and real-time coordinate conversion, eliminating cutting errors caused by product offset or rotation. Existing technologies use mechanical positioning devices to limit product position, and the scheme allows products to have position deviations at the workstations, automatically compensates for the offset amount through coordinate conversion, and reduces the dependence on mechanical positioning accuracy.
[0033] The application realizes real-time dynamic capture and accurate coordinate conversion of product contours, solving the cutting mismatch problem caused by the absence of edge detection in traditional cutting equipment. When the product position deviates or rotates during processing, the system automatically updates the cutting path coordinate data to maintain the matching accuracy of the Mylar cutting contour and the actual boundary of the product. The scheme avoids downtime adjustment operations caused by product position changes, improving the continuous operation capability and processing efficiency of the production line.
[0034] The application further proposes generating a target cutting path, which includes determining a edge compensation parameter according to a material attribute database, performing equidistant offset processing on contour information, and generating a curvature-continuous transition structure at the path turning point. The edge compensation parameter is determined by the edge compensation parameter.
[0035] The material attribute database refers to a database that stores physical property parameters of different materials, which can be implemented by using an AI engine to build a dynamically updated relational mapping table. By associating the corresponding relationship between material ductility, resilience coefficient and edge compensation amount, matching offset parameters are provided for different materials. The edge compensation parameter is a offset control parameter dynamically adjusted according to material properties, which can be implemented by using a parameter mapping table combined with a real-time feedback mechanism. By matching material thickness and elastic modulus to determine the compensation coefficient. The equidistant offset processing refers to a geometric operation based on the original contour to generate a parallel external expansion path, which can be implemented by using a vector direction expansion algorithm combined with path smoothing processing to accurately control the compensation amount by maintaining the contour topological relationship. The curvature-continuous transition structure refers to a smooth connection form without abrupt points at the path turning point, which can be implemented by using a Bezier curve fitting algorithm or a spline curve interpolation algorithm to eliminate stress concentration caused by geometric discontinuous points.
[0036] The material attribute database dynamically matches the corresponding edge compensation coefficient by real-time calling the physical parameters of the current processing material, so as to ensure that the offset is automatically adjusted according to the material characteristics. The equidistance offset processing expands the original contour in the vector direction based on the compensation parameter, to generate a cutting path matched with the material ductility. At the path turning point, the curvature continuous algorithm is used to reconstruct the corner area, to replace the right angle or acute angle with a circular arc or a gradual curve, so as to form a smooth transition without abrupt points. The process automatically processes any complex contour through a geometric algorithm, eliminates the stress peak value caused by the path mutation, and keeps the edge amount consistent.
[0037] The traditional cutting path generation adopts fixed compensation parameters, which cannot adapt to the ductility difference of different materials, resulting in edge amount deviation. The present scheme solves the material adaptability problem by realizing dynamic matching of compensation parameters through the material attribute database. The existing technology adopts a straight line splicing method at the path turning point, which forms a sharp corner and causes stress concentration. The present scheme eliminates the geometric abrupt points through the curvature continuous transition structure, and effectively disperses the cutting stress.
[0038] The present application realizes the self-adaptive adjustment of the cutting path to the material attribute, solves the problem of inaccurate edge amount, and eliminates the stress concentration phenomenon at the path turning point through geometric optimization, avoids material tearing and energy accumulation, and improves the precision of the cutting contour and the quality of the cut.
[0039] The present application further proposes to apply a decreasing air pressure distribution along the conveying direction, and synchronously apply an ionized gas flow to inhibit the electrostatic adsorption effect.
[0040] The decreasing air pressure distribution refers to gradually reducing the air flow pressure field along the material conveying path, which can be realized by using a multi-stage air pressure chamber combined with a proportional regulating valve. The front high-pressure gas flow lifts the material away from the contact surface, and the rear low-pressure gas flow forms a buffer layer to avoid material deviation caused by air flow impact. This distribution mode utilizes the Bernoulli effect to form a self-balancing air cushion, to maintain the tension balance of the material in the suspended state. The ionized gas flow refers to a neutralized gas flow carrying positive and negative charges, which can be realized by using a high-voltage ionizing rod combined with a directional nozzle array. The ionized gas flow covers the material surface by releasing equal amounts of positive and negative ions, neutralizes the static charge generated by friction, and eliminates the physical conditions of electrostatic adsorption. The gas flow and the air pressure distribution act synchronously to ensure that the charge neutralization process does not affect the suspension stability.
[0041] When the material enters the suspension platform, the front high-pressure airflow generates sufficient lift to make it separate from the mechanical contact surface. As the conveying process progresses, the air pressure gradient gradually decreases to form a laminar transition zone, and the material naturally flattens under the action of the pressure difference. In this process, the ionized airflow continuously covers the surface of the material, for example, by ionizing nozzles arranged at intervals, releasing positive and negative ions on the upper and lower surfaces of the material simultaneously, neutralizing the accumulated static charge. The air pressure distribution and ion neutralization form a spatiotemporal coupling, with the front section mainly controlled by air pressure to achieve initial positioning, and the middle and rear sections mainly maintained by charge neutralization to maintain stable transmission. The two are dynamically matched through a timing control module.
[0042] Traditional fixed platen applies pressure through mechanical contact, which can easily cause material deformation and cannot eliminate static adsorption. The conventional air floating platform uses a uniform pressure design, which can easily cause airflow turbulence. The present scheme replaces mechanical pressure holding with a gradient air pressure field, combined with real-time charge neutralization, to simultaneously solve the problems of material positioning and static interference in a non-contact state, avoiding the technical contradiction between the fixed and anti-static measures in traditional schemes.
[0043] The present application realizes the self-stabilization control of the Mylar sheet during the suspension transmission process, effectively suppresses the deviation of the material caused by airflow impact, eliminates the phenomenon of product and waste adhesion caused by static adsorption, ensures the accurate positioning of the cutting station, and avoids the material deformation caused by mechanical contact.
[0044] The present application further proposes a dynamic focusing cutting method, which includes real-time monitoring of the distance change between the surface of the Mylar sheet and the laser cutting head, and dynamically adjusting the focusing parameters of the laser beam according to the distance change amount to maintain the constant spot size of the laser action point.
[0045] Real-time monitoring of the distance change between the surface of the Mylar sheet and the laser cutting head refers to continuously obtaining the material surface height data through a non-contact distance measuring device. Specifically, a laser distance measuring sensor integrated in the cutting head can be used to achieve this. Its role is to capture the micron-level fluctuations of the material surface during the air suspension transmission process. Dynamically adjusting the focusing parameters of the laser beam refers to adjusting the focal length compensation amount of the optical system according to the real-time height data. Specifically, this can be achieved by driving the servo motor of the Z-axis beam expander. Its role is to control the spot size change within the allowable tolerance range. Maintaining the constant spot size refers to keeping the relative distance between the focal point position and the material surface stable through a closed-loop control algorithm. Specifically, this can be achieved by using a PID controller and an optical parameter feedback loop to work together. Its role is to eliminate the uneven energy density distribution caused by focal length deviation.
[0046] The laser ranging sensor collects the material surface height data at a millisecond level during the cutting process. When the surface undulation caused by the air suspension platform or the positioning deviation of the tooling is detected, the servo motor immediately drives the beam expander to move along the optical axis direction to compensate for the height difference. The optical feedback loop continuously compares the actual spot size with the preset target value, and forms a double closed-loop control by adjusting the laser power and scanning speed, so that the energy density of the laser action point remains stable. This process completely replaces the intermittent action of the traditional mechanical focusing mechanism, and realizes continuous dynamic compensation during cutting.
[0047] The traditional laser cutting equipment adopts fixed focal length mode or mechanical focusing mechanism. The former cannot adapt to the surface fluctuation of the material, leading to the deterioration of the cutting quality, and the latter has the defects of slow response speed and serious wear. The scheme combines non-contact real-time monitoring and dynamic adjustment of optical parameters, effectively overcomes the height instability caused by suspension transmission while maintaining the advantage of zero contact processing, and realizes the order of magnitude improvement of cutting precision.
[0048] The application solves the problem of laser energy focusing misalignment caused by material surface fluctuation in the suspension transmission process, ensures the consistency of the cutting width in different cutting areas, eliminates the molten adhesion defects caused by the change of spot size, and makes non-contact precision machining of ultra-thin flexible materials possible.
[0049] The application further proposes a separation method based on air pressure difference control, including applying directional negative pressure adsorption in the completed cutting area and applying positive air flow in the non-cutting area to realize physical separation through the air pressure difference.
[0050] The directional negative pressure adsorption refers to forming an adsorption force field lower than the ambient air pressure in the finished product area by a negative pressure generator. Specifically, a vacuum pump can be used in combination with a porous ceramic plate array to realize this. The porous structure allows one-way gas flow and forms a stable adsorption area under the finished product after cutting. This feature effectively fixes the position of the finished product by selective adsorption and prevents displacement after cutting.
[0051] The positive air flow pushing refers to generating directional air flow in the waste area by a compressed air system. Specifically, an adjustable nozzle array can be used to realize this. The jet angle of the nozzle array is consistent with the direction of the waste separation. This feature overcomes the residual adhesion between the material and the platform by applying air flow in the same direction as the waste separation direction.
[0052] The air pressure difference refers to forming a spatially distributed and differentiated pressure field on the platform surface by an air pressure gradient controller. Specifically, a gas path system with independent control in different areas can be used to realize this. This feature forms a separation driving force by building a pressure gradient between the cutting area and the non-cutting area.
[0053] After the cutting process is completed, the negative pressure adsorption system is activated in the corresponding area of the finished product, and the finished product is stably adsorbed on the platform surface through the porous ceramic plate. At the same time, the gas flow nozzle array located in the waste area sprays gas flow in the predetermined direction, pushing the waste to separate from the contact surface. In this process, the ionized gas flow integrated into the air film conveying module continuously neutralizes the surface charge of the material, eliminating the effect of static adsorption. The pressure difference between the cutting area and the non-cutting area forms a pressure gradient, causing the finished product and the waste to move in different directions under the action of mechanics, thereby realizing physical separation.
[0054] The traditional separation method relies on mechanical ejector pins or scrapers to physically contact the waste, which can easily cause deformation or damage to the ultra-thin material. The present solution uses non-contact gas pressure control to achieve separation through the action of the gas pressure field, avoiding material damage caused by mechanical contact. In addition, the traditional process does not consider the problem of static adsorption, and manual intervention is required to remove the retained waste. The present solution actively eliminates static electricity through ionized gas flow, ensuring the continuity of the separation process.
[0055] The present application solves the problem of automatic separation of finished products and waste caused by static adsorption after Mylar sheet cutting, and realizes non-contact physical separation. The finished product remains stable under the action of negative pressure, and the waste effectively separates under the action of gas flow, while ionization treatment eliminates static interference, ensuring the reliable progress of the separation process. This method avoids the material damage caused by traditional mechanical separation, and improves the quality stability of ultra-thin flexible material processing.
[0056] The present application further proposes to collect the thermodynamic parameters in real time during the cutting process, and when the thermodynamic parameters exceed the preset threshold, adjust the laser processing parameters, and feed the processing data to the optimization model to update the cutting strategy.
[0057] Thermodynamic parameters refer to physical quantities reflecting the temperature distribution and heat conduction state of the processing area, which can be realized by using an infrared thermal imager to scan the cutting area at a frequency of five frames per second to generate a temperature field cloud map, which is used to locate local high-temperature areas and quantify the degree of heat accumulation. The preset threshold refers to the critical condition for triggering parameter adjustment, which can be realized by setting a dynamic threshold interval in the control system according to the melting point characteristics of the material, which is used to prevent irreversible thermal damage to the material. Laser processing parameters refer to adjustable variables that affect heat input, which can be realized by adjusting the laser power, scanning speed and pulse frequency in combination, which is used to maintain the stability of the energy density at the action point. The optimization model refers to a process parameter self-learning algorithm, which can be realized by using a reinforcement learning framework to build a mapping relationship between thermodynamic parameters and cutting quality, which is used to continuously improve the adaptability of the strategy. The state space of the reinforcement learning optimization model is defined as wherein , , , the action space is defined as wherein , , the reward function is set as wherein the cut flatness score is calculated according to the cut edge roughness, when the score is 100, decrease by 10 for every 0.01 pm increase; the material safety temperature, such as PET Mylar sheet The model is iteratively optimized using the Q-learning algorithm, updating the Q-table every 20 pieces of Mylar processed to ensure that the laser parameters can adapt to different material thicknesses and cutting requirements.
[0058] The cutting strategy refers to the processing scheme that includes path planning and energy control. Specifically, an adaptive path optimization algorithm based on real-time temperature field distribution can be used to achieve this, which balances processing efficiency and thermal management needs.
[0059] During laser cutting, an infrared thermal imager continuously monitors the temperature distribution changes in the processing area. When the temperature value of a specific coordinate point is detected to exceed the material safety threshold, the control system immediately reduces the laser power in that area and increases the scanning speed, and simultaneously adjusts the cutting path to avoid high heat accumulation areas. The adjusted processing parameters and corresponding cutting quality data are transmitted to the optimization model for training, generating an updated cutting strategy and writing it to the control unit, achieving dynamic optimization iteration of processing parameters.
[0060] The traditional scheme relies on fixed cooling cycles or empirical parameter settings, and cannot respond to thermal fluctuations in the processing process in real time. However, this scheme establishes a closed-loop control of temperature field feedback and parameter adjustment, actively corrects the processing parameters before thermal damage occurs, and combines machine learning to achieve continuous self-optimization of process parameters.
[0061] The present application effectively prevents the phenomenon of material melting and sticking caused by local overheating, maintains the stability of the spot size of the laser action point, avoids the problem of cut quality degradation caused by heat accumulation in traditional processes, and simultaneously enables the system to adapt to processing requirements under different materials and environmental conditions through a self-learning mechanism.
[0062] The present application further proposes a Mylar adaptive cutting device comprising a visual positioning unit 10, a path planning unit 20, an air film conveying module 30, a laser processing unit 40, and a sorting unit 50.
[0063] The visual positioning unit 10 refers to a module for obtaining product contour information through an optical device. Specifically, it can be implemented by combining an anti-interference optical assembly with a real-time coordinate conversion module to eliminate surface reflection noise and establish a mapping relationship between image coordinates and physical coordinates, thereby solving the problem that traditional devices cannot capture product size changes in real time. The path planning unit 20 refers to a calculation module for generating a cutting path. Specifically, it can be implemented by combining a material attribute database with an offset algorithm to generate a transition structure with continuous curvature at path transitions through edge compensation processing of contour data, thereby avoiding size deviations caused by material springback. The air film conveying module 30 refers to a device for providing non-contact suspension transmission. Specifically, it can be implemented by combining a porous ceramic plate with an air pressure gradient controller to maintain the material in a suspended state through a laminar air cushion and simultaneously apply an ionized air flow to eliminate electrostatic adsorption effects, thereby solving the problem of material deformation caused by traditional pressure plate fixation. The laser processing unit 40 refers to an actuator for performing dynamic focus cutting. Specifically, it can be implemented by combining a real-time distance measurement sensor with a dynamic focusing system to monitor changes in the distance between the material surface and the laser head and adjust the laser beam focusing parameters to maintain a constant spot size, thereby overcoming the positioning errors of mechanical cutters. The sorting unit 50 refers to a negative pressure control device for material separation. Specifically, it can be implemented by combining a directional negative pressure adsorption array with a positive air flow generator to establish an air pressure difference between the cutting area and the non-cutting area to form a physical separation force field, thereby replacing the manual intervention separation method.
[0064] The visual positioning unit 10 captures product surface images through an anti-interference optical assembly and generates contour data in a physical coordinate system after processing by a coordinate conversion module. The path planning unit 20 performs offset processing on the contour data based on material compensation parameters to generate a cutting path containing an edge amount and a transition structure with continuous curvature at path transitions to avoid stress concentration. The air film conveying module 30 suspends the Mylar sheet by emitting a laminar air cushion from a porous ceramic plate and simultaneously applies a decreasing air pressure distribution along the conveying direction by an air pressure gradient controller, in combination with an ionized air flow to eliminate electrostatic adsorption. The laser processing unit 40 monitors changes in the material surface height through a real-time distance measurement sensor and dynamically adjusts the laser focal length to maintain a constant spot size. After cutting is complete, the sorting unit 50 applies directional negative pressure adsorption to the finished product in the cutting area and positive air flow to the waste material in the non-cutting area to achieve automated separation through the air pressure difference.
[0065] Traditional cutting devices rely on mechanical pressure plates to fix materials, which causes deformation. However, the present scheme achieves non-contact suspension conveying through the air film conveying module 30. Existing devices use pre-programmed cutting paths, which cannot adapt to size changes. However, the present scheme achieves dynamic path generation through the cooperation of the visual positioning unit 10 and the path planning unit 20. Conventional separation methods require manual intervention to address electrostatic adsorption problems. However, the present scheme achieves automated separation through the establishment of an air pressure gradient by the sorting unit 50.
[0066] This application effectively suppresses material deformation during Mylar film cutting, avoiding dimensional deviations caused by traditional mechanical fixing methods; it ensures the consistency of cutting energy at different positions through a dynamic focusing mechanism, improving contour matching accuracy; it solves the problem of process interruption caused by material adhesion by adopting non-contact levitation transmission combined with electrostatic elimination measures; and it ensures the stability of continuous production by achieving automatic separation of finished products and waste materials through air pressure difference control.
[0067] This application further proposes a visual positioning unit 10 including an anti-interference optical component for eliminating surface reflection noise and a real-time coordinate transformation module for establishing a mapping relationship between image coordinates and physical coordinates.
[0068] Anti-interference optical components refer to devices that suppress surface reflection interference from materials through optical filtering. Specifically, this can be achieved through the synergistic effect of polarizing filters and multispectral compensation light sources. Matching the polarization direction with the incident light angle eliminates specular reflection noise, while multi-band light source compensation balances the differences in diffuse reflection from the material surface. This component selectively filters interfering light waves while retaining effective edge feature information, thus solving the problem of image distortion caused by reflection in traditional equipment.
[0069] The real-time coordinate transformation module is a computational unit that establishes the correspondence between image pixels and physical spatial positions. Specifically, it can be implemented using a dynamic calibration algorithm combined with a distortion correction model. The coordinate transformation matrix is updated by real-time acquisition of feature point data from the calibration board. The dynamic calibration algorithm uses a 12×12 checkerboard calibration board with a grid spacing of 2mm and 144 corner points. Each calibration cycle acquires 20 sets of calibration board images in different poses. The camera's intrinsic and extrinsic parameters are calculated using the Zhang Zhengyou calibration method. The calibration error is evaluated using reprojection error, which is required to be ≤0.02mm. The recalibration process is automatically triggered after processing 50 Mylar films continuously, or when the laser cutting head position undergoes mechanical adjustment.
[0070] This module eliminates lens distortion errors and mechanical installation deviations, ensuring that the two-dimensional image data captured by the vision system is accurately mapped to the three-dimensional physical coordinate system, thus solving the problem of cutting path offset caused by coordinate misalignment in traditional systems.
[0071] During the image acquisition phase, the anti-interference optical component preprocesses the incident light. A polarizing filter filters out stray light reflected in specific polarization directions, and a multispectral compensation light source dynamically adjusts the illumination wavelength based on the material's reflection characteristics, ensuring a high-contrast contour of the product boundary during imaging. The real-time coordinate transformation module continuously receives mechanical motion parameters and image feature point data during the data processing phase. It calculates the coordinate transformation parameters under the current operating conditions using a dynamic calibration algorithm, converting the edge coordinates in the image coordinate system into the physical motion coordinates of the cutting equipment in real time. When both work together, the anti-interference component ensures that the original image data accurately reflects the actual contour of the product, while the coordinate transformation module eliminates inherent system errors, forming a closed-loop correction mechanism from image acquisition to mechanical execution.
[0072] Traditional vision positioning systems lack dedicated optical filtering devices, relying on ordinary industrial cameras to directly acquire images, making them susceptible to edge detection distortion due to surface reflections. Conventional coordinate transformations use fixed calibration parameters, failing to compensate for mechanical deformation and thermal expansion errors generated during equipment operation. This solution significantly improves the image signal-to-noise ratio through a combination of polarization filtering and multispectral compensation optical processing. Simultaneously, it employs a dynamic calibration algorithm to correct coordinate mapping relationships in real time, overcoming the limitation of static calibration parameters in adapting to changing operating conditions.
[0073] This application effectively eliminates image artifacts caused by reflections on the Mylar film surface, ensuring accurate extraction of product contour features. At the same time, it achieves high-precision dynamic matching between the image coordinate system and the physical coordinate system of the device. It jointly ensures the positioning accuracy of the cutting path from two dimensions: the data acquisition source and spatial transformation, avoiding the problem of mismatch between the cutting contour and the actual size of the product due to vision system errors.
[0074] This application further proposes that the air film delivery module 30 also includes an air pressure gradient controller and an electrostatic elimination component.
[0075] A pressure gradient controller is a device capable of generating a spatially variable pressure distribution. Specifically, it can be achieved using a multi-channel independently controlled valve array combined with a pressure sensor for closed-loop regulation. This dynamically adjusts the airflow intensity in different regions to create a gradient pressure field. The controller adjusts the output pressure of each orifice in real time based on the positional changes of the Mylar plate during transmission. During the acceleration phase, it enhances the front-end pressure to counteract inertial offset, and during the constant-speed phase, it maintains uniform suspension in the middle.
[0076] An electrostatic eliminator is a device that actively regulates the surface charge of a material. Specifically, it can be implemented by working in conjunction with an ionization rod array and a charge monitoring module. This neutralizes the accumulated charge on the surface of the Mylar sheet by releasing positive and negative ion streams. The device incorporates multiple charge detection points along the air-float transport path, dynamically adjusting the output polarity and intensity of the ionization rods based on real-time monitoring data.
[0077] When the Mylar sheet enters the air-film conveying module 30, the air pressure gradient controller generates a corresponding air pressure distribution pattern based on the preset motion trajectory. During the initial acceleration phase, the output pressure of the air vents at the front of the transmission direction increases to form a high-pressure zone, preventing the material from shifting backward due to inertia. During the high-speed transmission phase, the air vents in the middle section maintain a reference pressure to ensure stable suspension height, while the pressure of the air vents at the end decreases to prevent airflow recirculation and vibration. Before the material enters the cutting station, the electrostatic elimination component releases an ion flow with the opposite polarity to the surface charge of the material via an ionization rod, maintaining the surface potential of the material within a preset threshold range. The charge monitoring module continuously collects electrostatic data along the transmission path; when it detects that the charge density in a local area exceeds the standard, it immediately triggers the ionization rod at the corresponding location to enhance the neutralization intensity.
[0078] Traditional air flotation platforms, employing a fixed air pressure mode, are prone to material displacement or vibration during variable speed phases. This solution, however, achieves adaptive suspension control of the motion state through dynamic gradient air pressure distribution. Existing electrostatic elimination technologies mostly rely on passive grounding to conduct away charges, failing to respond in real-time to changes in charge distribution. This solution, by combining active ionization neutralization with multi-point monitoring, significantly improves the timeliness and uniformity of electrostatic elimination.
[0079] This application solves the problem of insufficient suspension stability of Mylar flakes caused by uneven air pressure distribution during air flotation. By dynamically adjusting the gradient air pressure field, the material maintains a stable suspension state at different stages of motion. At the same time, it eliminates the adsorption effect caused by electrostatic accumulation on the material surface and maintains the surface charge balance through real-time charge monitoring and active neutralization, ensuring reliable separation of finished products and waste after cutting.
[0080] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A method for adaptive cutting of Mylar film, characterized in that, include: Obtain the outline data of the product to be mounted, and generate a cutting path based on the outline data, wherein the outline data includes a set of product boundary coordinates; The cutting path enlarges the product boundary proportionally according to the preset edge wrapping amount, and transports the Mylar sheet to the air suspension platform. The laminar air cushion sprayed from the porous ceramic plate makes the Mylar sheet suspend and maintains the suspended state for transmission to the cutting station. A laser cutting head is used to scan and cut along the cutting path, while the laser focal length is dynamically adjusted according to real-time ranging data. After cutting, the negative pressure adsorption array is activated to separate the finished product from the waste.
2. The Mylar film adaptive cutting method according to claim 1, characterized in that, The acquisition of contour information includes: An image of the product surface is captured using an image acquisition device. Edge features are extracted from the surface image to obtain a set of boundary coordinates. The boundary coordinate set is converted into physical coordinate system data.
3. The Mylar film adaptive cutting method according to claim 1 or 2, characterized in that, The generated target cutting path includes: Based on the material property database, edge compensation parameters are determined, and the contour information is subjected to equidistant offset processing. The offset amount is determined by the edge compensation parameters, and a transition structure with continuous curvature is generated at the path turning point.
4. The Mylar film adaptive cutting method according to claim 1, characterized in that, The control of the non-contact suspension platform includes: A decreasing gas pressure distribution is applied along the conveying direction, and an ionized gas flow is applied simultaneously to suppress the electrostatic adsorption effect.
5. The Mylar film adaptive cutting method according to claim 1, characterized in that, The dynamic focusing cutting includes: The distance between the Mylar wafer surface and the laser cutting head is monitored in real time, and the laser beam focusing parameters are dynamically adjusted according to the distance change to maintain a constant spot size at the laser point of action.
6. The Mylar film adaptive cutting method according to claim 1, characterized in that, The separation based on pressure difference control includes: Apply directional negative pressure adsorption to the cut area; Apply positive airflow to the non-cutting area; Physical separation is achieved through the difference in air pressure.
7. The Mylar film adaptive cutting method according to any one of claims 1-6, characterized in that, Also includes: Thermodynamic parameters during the cutting process are collected in real time. When the thermodynamic parameters exceed the preset threshold, the laser processing parameters are adjusted, and the processing data is fed back to the optimization model to update the cutting strategy.
8. A Mylar film adaptive cutting device, characterized in that, include: A visual positioning unit is configured to acquire and process product outline data; The path planning unit is configured to generate cutting paths with material compensation. The air-film transport module is configured to provide a controllable levitation force field; The laser processing unit is configured to perform focus-adaptive cutting operations; The sorting unit is configured to separate materials using an air pressure gradient.
9. The apparatus as claimed in claim 8, characterized in that, The visual positioning unit includes: Anti-interference optical components are used to eliminate surface reflection noise, and a real-time coordinate transformation module is used to establish the mapping relationship between image coordinates and physical coordinates.
10. The apparatus as claimed in claim 8 or 9, characterized in that, The air-film delivery module also includes: A pressure gradient controller, configured to generate a spatially variable pressure distribution; An electrostatic elimination component configured to maintain the charge balance on the material surface.