Method for time optimization of laser processing of objects

Abstract

The invention relates to a method for time optimization of laser machining of an object (1), in particular a computer-implemented method, in which a laser light (2) can be guided along a predetermined machining path (4) on the object by means of a current scanning system (3); wherein a position-dependent energy input curve E (s) (15) for entering the object along the predetermined machining path (4) is determined by means of the laser; wherein a position-dependent first power curve L1 (s) (23) of the laser light (2) along the machining path (4) is determined, and wherein a position-dependent ideal speed curve vmax (s) is calculated on the basis of the position-dependent first power curve L1 (s) (23) and the position-dependent energy input curve E (s) (15); wherein a time-dependent path course s (t) is calculated in such a way that an achievable speed curve v (s) (17) of the laser light (2) along the machining path (4) is limited by a position-dependent ideal speed curve vmax (s) and adheres to at least one dynamic limit of the current scanning system (3); and wherein a second time-dependent power curve L2 (s (t)) (24) is calculated, and wherein the first position-dependent power curve L1 (s) (23) is reduced in such a way that the achievable speed curve v (s) (17) of the machining path (4) is smaller than the position (s) of the ideal speed curve vmax (s) (16), and wherein a position-dependent energy input curve E (s) (15) is obtained.

Description

Technical Field

[0001] The invention relates to a method for time-optimizing laser processing of an object, in particular a computer-implemented method, wherein a laser can be guided along a predetermined processing path on the object by means of a galvanometer scanning system; wherein a position-dependent energy input curve E(s) into the object along the predetermined processing path is determined by the laser. Furthermore, in the method a first position-dependent power curve L1(s) of the laser along the processing path is determined, and a position-dependent ideal velocity curve v is calculated based on the first position-dependent power curve L1(s) and the position-dependent energy input curve E(s). max In addition, the time-dependent path progress s(t) is calculated so that the achievable velocity profile v(s) of the laser along the processing path is subject to the position-dependent ideal velocity profile v max (s) and comply with at least one dynamic limit of the galvanometer scanning system along the machining path. A time-dependent second power curve L2(s(t)) is calculated, wherein the position-dependent first power curve L1(s) is reduced so that the achievable speed curve v(s) along the machining path is less than the position-dependent ideal speed curve v max At position (s) of the axis (s), a position-dependent energy input curve E(s) is realized. Background Art

[0002] In laser processing or laser treatment, the highest possible processing throughput or the shortest possible processing time is usually pursued. For this purpose, the highest possible path speed can be selected according to the available power of the laser and the performance of the laser process and specified as the laser speed of the laser process.

[0003] Likewise, for jumps between different path segments of a machining plan (with laser switching off), a higher speed (jump speed) can be set to shorten the overall machining time.

[0004] Real-time control of the galvanometer scanning system typically provides the method with automatic functions that implement adjustment corrections based on control data or sensor values ​​at the execution time of laser material processing (e.g., SCANLAB RTC6 automatic laser control / spot distance control functions).

[0005] These functions can usually be set by the user. In this case, known dependencies of the process for one or more process parameters can be stored in characteristic curves or parameter tables, on the basis of which the laser power is set correctly as a function of the one or more process parameters. For example, the resulting grayscale values ​​of the laser-marked pixels of the image can be related to the laser power to be used via a nonlinear function (e.g. gamma correction). The laser power can also be corrected, for example, as a function of the scanning position and / or the defocusing of the laser. In particular, the required laser power can also be nonlinearly dependent on the path speed of the scanner.

[0006] Typically, position-regulated galvanometer scanners do not follow a predetermined path exactly, but (due to the performance limitations of the position controller) exhibit temporal drag distortions. Drag distortions are usually compensated by laser-controlled time delays ("laser delays"), which allow path deviations (drag errors) to be corrected to a large extent, at least along a straight path at a constant path speed.

[0007] A method for laser processing of an object is known from WO 2020 / 025771 A1. Summary of the invention

[0008] The object of the present invention is to carry out laser machining or laser processing more rapidly.

[0009] This object is achieved by a method, a computer program, a computer-readable storage medium and a laser processing device having the features of the independent claims.

[0010] A method for time-optimized laser processing of an object, in particular a computer-implemented method, is proposed, wherein a laser can be guided along a predetermined processing path on the object by means of a galvanometer scanning system. When guiding the laser, the laser beam is guided and moved on the object, so that the object is processed. The object can be, for example, a workpiece for cutting, welding and / or engraving. Furthermore, the workpiece can also be processed in such a way that areas on the surface are ablated. In addition or alternatively, the object can also be a body part if the laser processing is used in the medical field. Furthermore, the object can also be a textile piece to be marked. In this case, pigments on the textile piece can be removed by the laser, so that the corresponding locations are bleached.

[0011] In the method, a position-dependent energy input curve E(s) is determined by the laser along a predetermined machining path into the object. The position-dependent energy input curve E(s) indicates how high the energy input is at which positions along the machining path. If, for example, material is to be ablated on a workpiece, the amount of ablated material depends at least on the energy input. The position-dependent energy input curve E(s) is therefore a preset value curve or is calculated from a preset value curve. The preset value curve can indicate how the machined object is constructed after machining.

[0012] In the method, a position-dependent first power curve L1(s) of the laser along the machining path is determined, and a position-dependent ideal velocity curve v is calculated based on the position-dependent first power curve L1(s) and the position-dependent energy input curve E(s). max (s). The position-dependent first power curve L1(s) can be determined to be as high as possible in order to obtain a high energy input in a short time. For example, the position-dependent first power curve L1(s) can be the maximum laser power or at least close to the maximum laser power. The position-dependent ideal velocity curve v max (s) indicates the maximum speed of the laser along the machining path, at which speed the position-dependent energy input curve E(s) is exactly reached in the case of the first position-dependent power curve L1(s).

[0013] Furthermore, in this method the time-dependent path progression s(t) is calculated so that the achievable velocity profile v(s) of the laser along the processing path is subject to the position-dependent ideal velocity profile v max (s) and comply with at least one dynamic limit of the galvanometer scanning system along the machining path (4). Although the position-dependent ideal velocity curve v max (s) represents an optimal value, which cannot be achieved due to the dynamic limits of the galvanometer scanning system. For this purpose, the time-dependent path course s(t) is calculated so that the achievable velocity profile v(s) contains the dynamic limits. This profile is then subjected to the position-dependent ideal velocity profile v max (s), that is, the position-dependent achievable speed curve v(s) does not exceed the position-dependent ideal speed curve v at any position. max (s).

[0014] Furthermore, the processing path on the object can be calculated based on the time-dependent path progression s(t), wherein the processing path can be represented as p(s(t)). In this case, p(s(t)) can, for example, include two components, i.e. be two-dimensional. Alternatively, p(s(t)) can also include three components, i.e. be three-dimensional. p(s(t)) indicates the coordinates of the laser on the object. For example, p(s(t)) can be set in a Cartesian coordinate system, which describes the coordinates of the laser on the object. p(s(t)) can also be represented as a vector based on the time-dependent path progression s(t).

[0015] In order to compensate for the position-dependent ideal velocity curve v max (s) The position-dependent achievable speed curve v(s) is calculated, and the time-dependent second power curve L2(s(t)) is calculated, wherein the position-dependent first power curve L1(s) is reduced so that the achievable speed curve v(s) is less than the position-dependent ideal speed curve v maxAt the position of the processing path (s), the position-dependent energy input curve E(s) is obtained.

[0016] The position-dependent energy input curve E(s) or simply the target energy input or energy input describes the effect that the user wishes to achieve on the workpiece during the process. The distance energy required for this can be constant (energy per unit distance), so that the average power of the laser can be selected proportionally to the feed rate (scanning speed). However, more complex situations may also be involved, in which, for example, the color change can be adjusted by the interaction of certain process parameter combinations of the laser (such as feed, focus and / or laser power), between which there are nonlinear relationships that must be taken into account in order to ensure the required energy input. The laser power can be adjusted, in particular by means of the pulse energy, the pulse duration and / or the pulse frequency.

[0017] It is advantageous if the time-dependent path progression s(t) is calculated in such a way that the processing time of the processing path is minimized. This allows for a time optimization of the laser processing.

[0018] It is advantageous if the time-dependent path progression s(t) is calculated in such a way that the achievable velocity profile v(s) of the laser is at least partially maximized. The maximum achievable velocity profile v(s) reduces the machining time of the machining path.

[0019] It is advantageous if an optimization method and / or an optimization algorithm is used to calculate the time-dependent path progression s(t). Various optimization methods are known for this purpose. For example, in an optimization method, several different time-dependent path progressions s(t) can be determined and the time-dependent path progression s(t) which provides the shortest machining time for the machining path can be selected. Additionally or alternatively, the time-dependent path progression s(t) can be calculated using a system of differential equations, in which the dynamic limit, the position-dependent energy input curve E(s), the position-dependent first power curve L1(s) and / or the position-dependent ideal speed curve v max (s) are incorporated as boundary conditions.

[0020] It is advantageous if the time-dependent path progression s(t) is calculated in such a way that the dynamic limits of the laser, in particular the dynamic limits of the laser power, are respected. For example, one dynamic limitation of a laser is that the laser power cannot be changed infinitely fast. For example, the time rate of change dL(t) / dt or the position rate of change dL(s) / ds is finite.

[0021] It is advantageous if the time-dependent path course s(t) is calculated in such a way that acceleration limits and / or jerk limits are observed as dynamic limits of the galvanometer scanning system. Acceleration limits and / or jerk limits are based in particular on the inertia of the deflection mirrors of the galvanometer scanning system, on the physical limits of the galvanometer drive and its current supply, and on the characteristics of the position control method for adjusting the angular position of the deflection mirror. Velocity, acceleration and / or jerk are generally vector variables. In this case, the above-mentioned dynamic limits for acceleration and / or jerk can be observed for at least one vector component. In addition or alternatively, when observing the dynamic limits, the value of the acceleration vector and / or the jerk vector can also be taken into account. This means that at least one, in particular all, of the vector components and / or the values ​​of the corresponding vectors observe the dynamic limits.

[0022] The dynamic limits are in particular limitations on the value ranges of the first, second and third time derivatives of p(s(t)). The dynamic limits can in particular be limitations on the value ranges of the second and third derivatives of the p(s(t)) vector components. The dynamic limits correspond to upper limits on the jerk and / or acceleration of the galvanometer drive in the galvanometer scanner.

[0023] This procedure is particularly advantageous when using a scanning system with a regulator without drag distortion, in which the dynamic limits including the acceleration and / or jerk limits of the galvanometer scanner can be described with a good approximation and the temporal behavior of the system can be predicted with high accuracy when executing acceleration and / or jerk limited trajectories.

[0024] It is advantageous if the time-dependent path progression s(t) is calculated so that the achievable speed profile and / or a profile derivable therefrom, in particular via a time derivative, is continuous and / or continuously differentiable, in particular smooth. A profile derivable from the time-dependent path progression s(t) is, for example, the time derivative ds(t) / dt as the achievable speed profile, d 2 s(t) / dt 2 As the acceleration curve, d 3 s(t) / dt 3 As jerk curves, even higher-order derivatives. In addition or alternatively, other curves can also be derived from the path course s(t) if the path course s(t) is referenced to calculate further curves. These curves have advantageous motion characteristics if they are continuous and / or continuously differentiable, in particular smooth.

[0025] If the position-dependent first power curve L1(s), the achievable speed curve v(s) calculated as a function of the time-dependent path progression, the curve derivable from the time-dependent path progression s(t) (in particular by means of a time derivative) and / or the time-dependent second power curve L2(s(t)) are calculated and / or specified such that they are step-shaped and / or constant in at least some path sections along the processing path, the calculation and / or control of the laser and / or galvanometer scanning system can be simplified.

[0026] For example, the position-dependent energy input curve E(s), the position-dependent first power curve L1(s) and / or the position-dependent ideal speed curve v can be calculated or specified in a stepped and / or constant manner. max (s), in order to respond to a specific processing task or if the restriction on the processing task is sufficient. This makes it easier to optimize. In addition, due to the step-shaped and / or partially constant performance of the curve or function, the associated values ​​can be stored and / or in a value table. The optimization performed by discrete values ​​allows simpler calculations and adaptive calculations. If the smaller the step-shaped and / or partially constant part is selected, the more accurately the corresponding curve and function can be determined. If the larger the step-shaped and / or partially constant part is selected, the less time is required for calculation. In addition, the size of the partially constant part of the curve or function can be selected so that their processing or processing is within the clock range of the controller. In particular, it is advantageous if the position-dependent energy input curve E (s) is step-shaped and / or constant in at least some path sections along the processing path. This position-dependent energy input curve E (s) can exist as input data or preset values, such as input data or preset values ​​in a table. This can be provided for special customer requirements, wherein the use of discrete position-dependent energy input curves E (s) can also be used to optimize time in a simpler way.

[0027] If the time-dependent second power curve L2(s) and / or the position-dependent ideal speed curve v is calculated max (s), it is advantageous to also calculate the orientation and / or positioning of the galvanometer scanning system relative to the processing path. For example, if the object is tilted relative to the galvanometer scanning system and thus relative to the laser, the laser will fall obliquely on the object and the processing path. As a result, the full power is no longer applied to the processing path. In addition, the processing path can no longer be traversed at the highest possible speed, so the position-dependent ideal speed curve v can also be taken into account in the calculation. max (s).

[0028] It is advantageous if a time-dependent coordinate p(s(t)) is determined relative to the machining path of the galvanometer scanning system based on the time-dependent path progression s(t) so that the laser moves along the machining path (4). For example, the coordinate p(s(t)) is a two-dimensional vector whose components are p x (s(t)) and p y (s(t)). Alternatively, p(s(t)) can also be a three-dimensional vector, which can describe the position of the laser in space. These indicate the coordinates of the processing path on the object. These can then be converted by the galvanometer scanning system into the time-dependent angular position of the deflection mirrors of the galvanometer scanning system, thereby moving the laser along the processing path.

[0029] It is advantageous if the method for time-optimizing laser processing of an object is executed by a computing unit.

[0030] It is advantageous if the coordinates p(s(t)) are transmitted to a current control unit of a galvanometer scanning system and the time-dependent second power curve L(s(t)) is transmitted to a laser controller. The laser controller controls the power of the laser so that a time-dependent second power curve L(s(t)) is formed. It is advantageous if the object is processed after the transmission. Preferably, the object is processed after the time optimization of the processing path. In this case, the time optimization of further processing paths of the entire processing task or path segments can optionally also be carried out simultaneously with the already calculated paths or path segments. For this purpose, the entire processing task can be decomposed into path segments, which represent the processing path in the sense of the method according to the invention. In this process, higher requirements are placed on the time efficiency of the method for time optimization.

[0031] It is advantageous if, in preparation for the actual laser processing of the object, a computer-aided simulation of the laser processing is used. Here, for example, the operating behavior of the laser processing can be simulated. This can be used to simulate and determine, for example, whether the laser processing can be performed with an achievable speed curve v(s) and / or a time-dependent second power curve L2(s(t)). In addition or as an alternative, it can be simulated and / or determined whether the processing path can be processed correctly or within tolerance limits using the determined curves. In addition or as an alternative, the time required for processing the processing path can be simulated and / or determined. This simulation has the advantage that if it is discovered during the simulation that the laser processing is not progressing correctly or as expected, this can be corrected. An adaptation and adjustment of the laser processing or the time optimization method can then be carried out.

[0032] It is advantageous if the actual laser processing of the object takes place after the time optimization of the laser processing. Additionally or alternatively, the actual laser processing of the object can be carried out after the simulated laser processing. This has the advantage that the laser processing process is known, thereby avoiding errors in the processing process.

[0033] Furthermore, a method, in particular a computer-implemented method, for optimizing the laser processing time of an object of a laser processing device having a galvanometer scanning system and a laser is proposed, in which the energy input of the laser is specified or predetermined by the laser for several path sections of a predetermined processing path, wherein the corresponding energy output depends on the laser power and the laser speed of the laser in the corresponding path section. In the method, a first, in particular constant and / or maximum laser power of the laser is specified on the path section. Subsequently, taking into account the first laser power, a corresponding ideal laser speed is determined for the path section, with the aid of which the corresponding energy input in the path section is achieved. Thereafter, taking into account at least one dynamic limit of the galvanometer scanning system and / or the laser, the corresponding ideal laser speed is adjusted to the corresponding achievable laser speed, in particular reduced. Subsequently, in those path sections that have been adjusted to the corresponding achievable laser speed, the first laser power is adjusted to, in particular reduced to, the second laser power in order to achieve the corresponding energy input. It is advantageous if the energy input, the laser power, in particular the first and / or second laser power, the laser speed, in particular the ideal and / or achievable laser speed, are preferably present as time- and / or position-dependent mathematical functions. Additionally or alternatively, it is advantageous if the energy input, the laser power, in particular the first and / or second laser power, the laser speed, in particular the ideal and / or achievable laser speed, are constant in at least one path section or vary in a time- and / or position-dependent manner.

[0034] Preferably, the invention is constructed additionally or alternatively in accordance with the preceding and / or following description, wherein said features may exist alone or in any combination.

[0035] To process a workpiece, a laser can be guided along a path on the workpiece using a galvanometer scanner. Optionally, the position of the focus along the laser beam can be adjusted synchronously using a dynamically variably adjustable focusing device to guide the laser focus along the path in three spatial dimensions. In this case, the path p(s(t)) is described by the three vector components of p, for example p=(p x , p y , p z ). The processing result can be, for example, laser marking, laser ablation, laser welding or laser cutting. In order to achieve the desired processing result with high precision, the path of the laser focus relative to the workpiece and the characteristics of the laser beam must be controlled synchronously to adjust the local energy input. Depending on the design of the laser, for example, the change in energy input can be set by varying the average power or energy, duration or frequency of the pulses or the duty cycle during modulation. The laser can be switched off in areas that do not need to be processed and switched on along the processing path.

[0036] Due to the high demands on synchronicity, the laser, the galvanometer scanner, the focusing device and / or other movement axes, if necessary, are controlled by specially designed real-time control systems (eg SCANLAB RTC6).

[0037] Depending on the application, the processing goal can be, for example, a uniform energy input along a path or a targeted local variation, such as laser marking on the image lines of a rasterized image with grayscale information (bitmap) that is available for each image point, i.e. spatial resolution.

[0038] Therefore, a method is proposed having one or more of the following features:

[0039] - the starting point is a predetermined geometry of the machining path p(s) and a target energy input E(s) value for at least one support point at a point along the path. (s: path parameter, i.e. distance along the path); and / or

[0040] - use of a galvanometer scanning system with a "trajectory regulator" (i.e. a position regulation method with no drag distortion); and / or

[0041] - using a laser that allows variable adjustment of at least one parameter within an adjustment range and within the dynamic limits of the laser to influence the average laser power. The laser power parameter L required to achieve the energy input E is described by the functional relationship L(E); and / or

[0042] - using a computing unit to optimize the process plan; and / or

[0043] - Optimize the execution time of a path (trajectory) p(s(t)) by maximizing the path progress speed curve ds / dt using an optimization algorithm,

[0044] - wherein the position and dynamic limits (axis position, velocity, acceleration and possibly jerk) of the galvanometer scanning system are preferably not exceeded (necessary for the algorithm: conversion into ds / dt and d 2 s / dt 2 ,d 3 s / dt 3 limits); and / or

[0045] - not exceed the minimum required processing time determined by the target energy input E(s), the maximum available laser power L(E) and / or the dynamic limits of the laser power variation (dL / dt, ...) and therefore not exceed the maximum possible path speed depending on s (i.e. an additional upper limit of ds / dt); and / or

[0046] - using the function s(t) determined in this way to calculate the time curve of the scanner controller p(s(t)) and / or the time curve of the laser power preset value L(E(s(t))) based on the predetermined path energy input E(s); and / or

[0047] - transmitting the pre-planned trajectory p(s(t)) and the time profile of the laser power L(E(s(t))) to a real-time control system; and / or

[0048] - Synchronous output of the control data of the galvanometer scanning system and the laser based on the precalculated trajectory and the time curve of the laser power (wherein the signal preparation / signal conversion / temporal rasterization of the corresponding signals of the laser and the galvanometer scanner can be carried out independently of each other).

[0049] Depending on the process and the laser used in conjunction with the deflection system (scanner), different relationships arise between the target energy input E and the laser parameter L to be adjusted, which must be described by the function L(E). This often nonlinear relationship can be approximated by means of a characteristic curve.

[0050] The function L(E) depends in particular on the path speed of the laser and can therefore be a function of these two parameters L(E, v). Typically, at higher path speeds, a higher laser power L is set to achieve the same energy input. This relationship depends on the physics of the interaction of the laser with the object material and can be approximately linear or nonlinear with respect to speed.

[0051] A correction that depends on the positioning of the laser focus generated by the scanning system may be included in the function L(E) (optical image compensation).

[0052] The function E(s) can be defined as being partially constant or linear and / or having equidistant support points. For example, marking lines on a bitmap (grayscale image) or at the start and end of a welding process.

[0053] The predetermined geometric path may be a row of a grayscale rasterized image, and / or the predetermined target energy input may be partially constant and equidistant to reflect the grayscale values ​​of the grayscale image pixels.

[0054] For galvanometer scanning systems with trajectory regulators, the pre-calculated trajectory is preferably at least velocity and acceleration limited. Path accuracy can be improved by additional jerk limiting.

[0055] It is advantageous if the maximum possible velocity profile is determined in advance based on the geometrical profile of the path, in which all dynamic limits of the galvanometer scanning system can be observed. In the case of curved paths, a local maximum velocity can be determined at which the acceleration and jerk limits of the galvanometer scanning system are not exceeded in any axis direction.

[0056] The calculation complexity of the time-optimized jerk-limited trajectory is high. In order to save calculation time, smoothing methods can be used when calculating the acceleration- and / or jerk-limited trajectory, in particular when calculating the jerk-limited trajectory from the acceleration-limited trajectory.

[0057] The galvanometer scanning system may include two galvanometer axes, and the path may be a 2D path in a processing plane located in the focal plane.

[0058] The galvanometer scanning system may include a dynamic focusing system, and the path may be a 3D path, particularly a path arranged on a non-planar surface of the workpiece.

[0059] The galvanometer scanning system can be combined with other slower redundant axes, where the plan can include the division of motion between the galvanometer scanner's axes and the redundant axes. The other axes can move the workpiece relative to the galvanometer scanning system, or the galvanometer scanning system relative to the workpiece.

[0060] The galvanometer scanning system can include four galvanometer axes to dynamically adjust the beam position and beam direction in two axes. The process parameters then also specifically include the angle of attack of the laser beam. This 4-axis system can be expanded around the focusing unit so that the processing distance is also variable.

[0061] In the pre-planning, beam defocusing that may occur due to the limited dynamic function of the focusing device can be compensated by adjusting (increasing) the laser power. Alternatively, a buffer setting range of the laser power can be provided, which is prepared in the pre-planning so that real-time compensation can be performed during the execution of the processing.

[0062] Deterministic pre-planning and control is more advantageous than real-time compensation or iterative optimization manipulation by trial and error. Advantageously, processing is performed at a lower target energy input to a location along a path at a higher scanning speed. Furthermore, it is advantageous if the limits of the available laser power are taken into account (in an intermediate step) by calculating a curve of the local maximum speed from the local target energy input. Furthermore, it is advantageous if the execution time is optimized taking into account the dynamic limits of the scanning system and / or the laser.

[0063] It is advantageous if a drag-free regulator of a galvanometer scanning system is used, when this preferably requires a precalculation of a target value curve (trajectory planning) in order to limit the acceleration and / or jerk along the path. Due to the drag distortion, it is practically impossible to correctly precalculate the time curve of the laser power of the scanning system when changing speed, because the time behavior of the regulator is difficult to model and thus the exact path-dependent introduction of the target energy input cannot be reasonably precalculated. In such a case, such a system can choose an iterative process for further optimization, in particular by at least one of the following steps: speed change and recording curve, calculation and verification of parameter L, checking the optimality and / or checking the speed change again or accepting the solution. In addition, for fast changes in the scanning speed, the dynamics of the scanning system with a classical regulator can only be used within a limited range.

[0064] One application of the above invention is, for example, the use of a laser to create a bitmap (a rasterized image with grayscale information), for example, in particular for creating patterns on textiles or generating photos on smart card IDs, as described below. Preferably, the invention is additionally or alternatively constituted according to the following description, wherein the features described can exist alone or in any combination. The bitmap consists of a large number of pixels, which are marked unidirectionally or bidirectionally line by line using a laser scanner, in particular a galvanometer scanning system. Typically, the line speed is kept constant and the laser energy per pixel varies depending on the required grayscale value. The dynamics of the scanning system are independent of the image quality because the acceleration phase is outside the range of the bitmap. Drag distortion can be compensated by line offset. In this way, high-resolution bitmaps can be accurately laser printed. In order to save process time and increase productivity at a given laser power, the scanning speed within the line can be adapted to the grayscale value curve. The required laser energy is determined by the grayscale value of the pixel, from which the maximum possible pixel speed of the pixel position can be calculated together with the maximum available laser power. For each row of pixel lines, a time-optimized trajectory, i.e. a time-dependent position curve, can be determined. The acceleration limit and jerk limit of the system must be considered. The velocity of the trajectory at each pixel position must be less than the maximum possible pixel velocity calculated from the grayscale value, otherwise the grayscale value cannot be achieved.

[0065] The laser power can be used optimally as long as the maximum laser power is always used and the grayscale values ​​are only achieved by the scanning speed. The grayscale values ​​can change arbitrarily between pixels, which would require very high accelerations of the scanning system. The task now is to determine a time-optimized trajectory under given boundary conditions. The possible time savings depend largely on the grayscale value distribution of the bitmap, and in the case of fabric structuring the conditions are generally favorable.

[0066] In order to be able to follow the calculated trajectory exactly, the scanner controller needs to be pre-controlled. This can be done with a scanning system (such as SCANLAB exceliSCAN) or offline. The laser control of the pixels must be done exactly at each pixel position. Due to the power-on time of the laser and the pixel speed, the pixels extend over a certain length that is smaller than the pixel distance. The laser control should be done in such a way that the pixel is located in the center of the desired pixel grid.

[0067] In the following, the gray value of a pixel i within a row is referred to as value. The value range of G is between 0 and 1. Bitmap files usually use 8-bit gray values, i.e. the gray values ​​must first be normalized to 1. Gray value 0 corresponds to "white" and therefore to maximum laser energy. However, depending on the method, the opposite relationship may also exist.

[0068] When using a device with maximum power P max The maximum pixel energy E is generated when using a laser, especially a CW laser diode. i =P max ·D / v i , where D is the pixel distance and v i The velocity at pixel i. This is only an approximation, as the velocity is not constant. However, since the pixel spacing and the velocity variation between pixels is small, this approximation can be used.

[0069] Assuming that there is a linear relationship between grayscale value and pixel energy, the maximum possible velocity v at pixel i can be determined from this: i . E max represents the pixel energy required to achieve a grayscale value of 0. Actual speeds are usually lower and grayscale values ​​must be achieved by reducing the laser power or briefly switching it off. A grayscale value of 1 would allow infinitely high speeds, but this is limited by the dynamic limits, i.e., only a pixel frequency of 100kHz can be produced, for example, preferably by an RTC control card in microvector mode: v limit =D·100kHz

[0070] E i =(1-Gi)·E max

[0071]

[0072] CO2 lasers can be used. These are operated, for example, by PWM signals in the frequency range of 0-100 kHz, typically the duty cycle must not exceed a value of, for example, 55%. Due to the high time constant, in particular about 60 μs, a non-linear relationship between the laser on time and the pixel energy can be assumed and therefore corrected using, for example, a gamma correction γ.

[0073] Calculate the laser on time LaserOn on pixel i based on the gray value of the white pixel and the maximum laser on time i .

[0074] LaserOn i =(1-G i ) γ LaserOnMax

[0075] The maximum pulse frequency is calculated using the maximum allowed duty cycle, and the maximum possible pixel speed is calculated using the pixel distance.

[0076]

[0077] The trajectory can be calculated with acceleration and / or jerk limitations, whereby it can also be performed by a scanning system. Since the third-order derivatives for trajectory planning are relatively complex and time-consuming, a simpler formula can be chosen for some applications. First, the second-order derivatives of the trajectory are calculated. Then, jerk limitation is achieved by forming a smoothed average. This method is much simpler and faster, but does not provide an optimal solution.

[0078] Speed ​​planning can be carried out within a location area, ie depending on the location.

[0079]

[0080] The maximum possible velocity change between two adjacent pixels at a distance D can be determined by integration.

[0081]

[0082] By planning forward and backward along a single pass of a pixel row, velocity planning limited by acceleration can be done very quickly. i+1 , and compares it to the maximum possible speed for the pixel. If the speed is allowed, it is used for that pixel, otherwise the maximum possible speed is used.

[0083] From the optimized position-dependent velocity profile over the pixel rows, the trajectory x(t) can now be easily calculated.

[0084] The maximum possible acceleration of the scanning system for in-line trajectory planning does not necessarily have to be used. A reduced value may be sufficient to achieve the desired performance improvement.

[0085] If jerk limitation is required, a smoothed average can be applied to the trajectory. The average is performed over 2·M+1 time steps, where the symbol j indicates the position in a 10 microsecond time step. The acceleration jumps are thus distributed over 2·M+1 time steps and the jerk is limited.

[0086]

[0087] The mean value formation acts like a low-pass filter on the pixel speed, so that the speed can rise at certain locations and slightly exceed the maximum possible speed. This can also be avoided by suitable improvement of the planning method.

[0088] Furthermore, the overtravel must be added at the start and end of the line. The length is given by the pixel velocity at the first and last pixel multiplied by the onward travel time of M·10μs. The laser control signal for the pixel can now be determined from the trajectory, where the signal travel time must be taken into account.

[0089] The jump from the end of one row to the start of the next row should also be time-optimized. The calculation can be done with a 3rd or 2nd order and averaging filter. The start and end speeds of the pixel rows must be considered separately. In addition, the start and end positions of the rows may change. For time reasons, the rows are bidirectionally labeled. Fast algorithms are available for the jump planning in the form of commercial or free libraries. Depending on the dynamic limits of the scanning system and the available adjustment range of the laser power, the speed may also have to be limited explicitly and not just implicitly via the grayscale values ​​of the pixel rows.

[0090] Before the scanner and laser control are controlled according to the plan set forth here along the lines of the bitmap, an optimization plan of the path can optionally also be performed according to the image content. This may mean that the lines of the image do not necessarily have the same length and do not necessarily always have the same direction / orientation. The image can also consist of several areas with different processing strategies. These areas can also overlap or interlock. However, the method set forth here for planning scanner movements and laser control can be applied here in the same or similar manner.

[0091] In order to avoid missing the first pixel in the first row due to laser dynamics, the laser steering should be planned appropriately to move the laser safely to the beginning of the row in a state that enables the start of the row to be achieved on time.

[0092] Furthermore, a computer program is proposed which, when executed by a computing unit, causes it to perform at least some of the method steps of the method for time optimization according to one or more method steps described in the preceding and / or subsequent description. For example, the computer program can be stored in a memory in the computing unit and / or can be executed by the computing unit. The computing unit can also be formed by a computer.

[0093] A computer-readable storage medium comprising a computer program is also proposed, wherein, when executed by a computer, the computer program causes it to perform a method for time optimization according to one or more method steps described in the preceding and / or subsequent description. The computer-readable storage medium may be the above-mentioned memory. The computer program may be designed according to the features described in the preceding and / or subsequent description.

[0094] Furthermore, a laser processing device is proposed, which has a laser for laser processing and is used for processing an object with the laser. The laser light can be generated by a laser unit.

[0095] Additionally, the laser processing apparatus includes a galvanometer scanning system for directing the laser light over the object.

[0096] Furthermore, the laser processing device comprises a computing unit, wherein the computing unit is designed to execute the method according to at least one method step described in the preceding and / or subsequent description.

[0097] In the following, a possible calculation of a processing path for time-optimized passage or processing of an object is presented.

[0098] The geometry route is set as:

[0099]

[0100] Where x(s), y(s), z(s) are the path coordinates with respect to the previous arc length s∈[0,L], and the vector velocity, acceleration and jerk can be expressed as follows.

[0101]

[0102]

[0103] in

[0104]

[0105] The optimization problem is formulated by maximizing speed.

[0106]

[0107] The solution is limited by the maximum achievable velocity, acceleration, and jerk.

[0108]

[0109] Since t is unknown, it cannot be calculated explicitly and So a replacement was made. get,

[0110]

[0111] By definition, is restricted to be positive, so q is also restricted to be positive. If q is now restricted to be the second-order derivative of the B-spline, then at the support points of the spline [a1, a2…a K ]The derivatives q',q" can be considered to be linear.

[0112]

[0113] Among them, N i,2 (s) are the basis functions of the spline. These are determined with the aid of the De Boor algorithm.

[0114] Support point [a1,a2…a K ] is now the optimization parameter. In order to make the optimization task completely linear, it is necessary to An approximate value is found. To this end, the optimization problem without jerk constraints is solved. The q thus determined * represents the upper limit of all achievable jerk-constrained trajectories. The solution space of the problem may be smaller or equal to the already determined q due to the additional condition (jerk) * With the aid of this approximation, the condition can be introduced as a pseudo-jerk in the second optimization step.

[0115]

[0116] Because q * ≥q and This leads to two necessary steps for determining a jerk-limited trajectory,

[0117] Step 1:

[0118]

[0119] and

[0120]

[0121] Step 2;

[0122]

[0123] and

[0124]

[0125] In the case of using the matrix representation of B-splines,

[0126]

[0127] Efficient LP solvers can be used. With this representation, it is now also possible to predetermine an explicit v for each route parameter s max,i .

[0128] BRIEF DESCRIPTION OF THE DRAWINGS

[0129] Other advantages of the present invention are described in the following embodiments. The accompanying drawings show:

[0130] Figure 1 : Processing of the object 1 by means of a laser 2 along a processing path 4 by means of a galvanometer scanning system 3,

[0131] Figure 2 : Example of position-dependent energy input curve E(s) and position-dependent ideal and achievable speed curve v(s)

[0132] Figure 3 : Another example of a position-dependent energy input curve E(s) and a position-dependent ideal and achievable speed curve v(s). DETAILED DESCRIPTION

[0133] Figure 1 The laser processing of an object 1 by means of a laser processing device 19 with a laser 2 along a processing path 4 is shown. A laser source 20 generates a laser beam (hereinafter referred to as "laser"). The laser 2 is guided over the object 1 and along the processing path 4 by means of a galvanometer scanning system 3. In the following, it is explained that the laser 2 is moved over the object 1. For this purpose, the beam of the laser 2 or the laser beam or the laser focus of the laser 2 is moved over the object 1 and along the processing path 4.

[0134] According to the present embodiment, the processing path 4 has a path start point 5 and a path end point 6. Therefore, the laser 2 passes through the processing path 4, or simply referred to as the path 4, from the path start point 5 to the path end point 6. In particular, the laser 2 starts processing the object 1 at the path start point 5 and ends processing the object 1 at the path end point 6.

[0135] Galvanometer scanning system 3 means here that the galvanometer scanning device 3 has at least one, in particular at least two galvanometer drivers 9, 10, which will be explained in more detail below. Galvo is here and hereinafter the abbreviation for galvanometer driver.

[0136] In order to deflect the laser 2 onto the object 1 and to be able to move the laser 2, the galvanometer scanning system 3 comprises two deflection mirrors 7, 8 in this embodiment. With the aid of the two deflection mirrors 7, 8, the laser 2 can be moved on the surface of the object 1. For example, the laser 2 can be moved in one direction by means of the first deflection mirror 7, while the laser 2 can be moved in a transverse direction relative thereto by the second deflection mirror 8. Overall, the laser 2 can be moved on the object 1 thereby.

[0137] As shown in this example, the laser beam can be focused onto the plane of the workpiece, optionally using scanning optics 11 located in the beam path after the deflection mirrors 7, 8. For this purpose, in particular, an f-theta lens can be used, in which the position of the focus in the plane depends essentially linearly on the deflection angle of the two deflection mirrors.

[0138] To be able to move the two deflection mirrors 7, 8, a galvanometer drive 9, 10, or galvanometer 9, 10 for short, is assigned to each deflection mirror 7, 8. The first galvanometer drive 9 can be used to move the first deflection mirror 7, and the second galvanometer drive 10 can be used to move the second deflection mirror 8.

[0139] The galvanometer scanning system 3 receives position data p(t) from the controller 22, which represent the coordinates of a processing path 4 on the object 1. The processing path 4 is determined by the position data p(t). These position data p(t) are used to control the galvanometer drives 9, 10 in order to predetermine the target position of the deflection mirrors 7, 8 at the time t. In this case, the respective orientation of the deflection mirrors 7, 8 also depends on the orientation and / or position of the galvanometer scanning system 3 relative to the object 1. In addition, in the formula p(t)=p(s(t)), the position data p(t) are determined with the aid of the time-dependent path course s(t). The calculation of the time-dependent path course s(t) is described in the following description and / or explained in the previous description.

[0140] Furthermore, the laser processing system (NR) comprises a focusing unit 21, by means of which the divergence of the laser beam 2 can be changed. As a result, the object 1 can be processed not only in the width and length direction, but also in the depth direction. Furthermore, the laser 2 can be widened by means of the focusing unit 21, so that the laser 2 processes the object 1 not only at one point or in a relatively small area, but also in a relatively large area due to the widening. Thus, the laser 2 can be defocused. As shown here, the focusing unit 21 is located behind the laser source 20 and / or is spaced apart from the galvanometer scanning system 3. As can be seen here, the focusing unit 21 is located in front of the laser coupling device 12 explained below. Optionally, the focusing unit 21 can also be a component path segment of the galvanometer scanning system 3.

[0141] Furthermore, the galvanometer scanning system 3 has a laser coupling device 12. The laser light 2 can be generated remotely and coupled into the galvanometer scanning system 3 for guidance on the object 1. Alternatively, the laser light 2 can also be generated directly in or at the galvanometer scanning system 3. The laser source 20 shown here can thus also be a component path section of the galvanometer scanning system 3.

[0142] The laser processing device 19 comprises a controller 22, which in the present example is located outside the galvanometer scanning system 3, but can optionally also be integrated into the galvanometer scanning system 3, in particular into the galvanometer control unit 13. The controller 22 can also be integrated into the computing unit 14. As controller 22, a slot card can be inserted into the computer unit 14, for example, or a SCANLAB RTC6 Ethernet box connected to the computer unit 14 via Ethernet.

[0143] The controller 22 can advantageously control the laser source 22 so that the laser power can be adjusted. The controller 22 can be connected to means for modulating the laser power, which in the present example can be arranged in the laser source 22. Here, this can be, for example, an acousto-optic modulator or a modulation of the laser diode current. The controller 22 is also connected to the galvo scanning system 3 to send the position data p(t) to the galvo control unit 13. Optionally, the controller 22 can also be connected to the focusing unit 21 to change the focus of the laser 2.

[0144] Furthermore, the galvo scanning system 3 according to the present embodiment comprises a galvo control unit 13, which controls the galvo scanning system 3 and thus controls the movement of the laser 2. In particular, the galvo control unit 13 can adjust the angular position of the two galvos 9, 10 or the two galvo drives 9, 10 so that the two deflection mirrors 7, 8 are moved according to the target position curve and the laser 2 is guided along the processing path 4. In order to follow the processing path 4 as accurately as possible, a control method without drag distortion is preferably implemented in the galvo control unit 13, in particular a so-called state controller or a so-called trajectory controller. The control method may require that the target position curve adheres to dynamic limits, in particular dynamic limits for speed and / or acceleration and / or jerk, which correspond to the physical limits of the galvo drives 7, 8 and / or the limits of their current supply. For example, in order to increase the stability of the control algorithm, these dynamic limits can also be narrower than the corresponding physical limits.

[0145] As shown here, the processing path 4 has a continuously changing thickness. The thickness here shall denote the energy input E into the object 1 along the processing path 4. Thicker locations on the processing path 4 shall denote a high energy input E, while thinner locations denote a correspondingly low energy input E. The energy input E is also the energy introduced into the object 1 by the laser 2 along the processing path 4. The energy input E depends on the laser power L of the laser 2. A high laser power L leads to a high energy input E, a low laser power L to a low energy input E. Furthermore, the energy input E also depends on time. The longer the laser 2 is located at a location on the processing path 4 or at a region along the processing path 4, the more energy is input.

[0146] By means of the energy input E or by means of the laser 2, for example, the object 1 can be engraved, ablated, marked, cut and / or welded. Depending on the energy input E generated by the laser 2 at a certain location or a certain area, the object 1 is also processed accordingly. For example, with a small energy input E, for example in the area of ​​the path starting point 5, the object can be processed so that the properties of the object 1 change only slightly at this location. For example, if material is to be ablated, only a small amount of material is ablated with a small energy input E, as shown here at the path starting point 5. In the middle area of ​​the processing path 4, where the processing path is shown thicker here, more energy is input or the energy input E is higher compared to the path starting point 5.

[0147] The object 1 is thus processed more intensively here, provided, of course, that the object 1 has corresponding constant properties or consists of the same material, for example, along the processing path 4 shown here. If material is also ablated, more material is ablated due to the higher energy input E, or the properties of the object 1 change more intensively at locations where the energy input E is higher. As already mentioned, the assumption here is that the entire object 1 has constant properties for easier explanation.

[0148] The machining of the object 1 along the machining path 4 thus depends on the energy input E along the machining path 3. Furthermore, the machining of the object 1 also depends on other factors, such as the material etc., which have not been considered here. However, these factors may be taken into account in the method described in the following description and / or in the method described in the preceding description.

[0149] As another example, the object 1 can be a textile material to which a pattern is to be applied. With the aid of the laser 2, a piece of fabric, in particular a dyed piece of fabric, can be faded along a processing path 4. The energy input E of the laser 2 can destroy the dye, so that the fabric becomes lighter at these locations. Thus, a pattern can be applied by the energy input E along the processing path 4. By changing the energy input E along the processing path 4, different degrees of fading can therefore also be achieved, so that the piece of fabric becomes lighter or darker (with respect to the color of the fabric) accordingly along the processing path 4. If the processing path 4 is planned accordingly, various two-dimensional patterns or, for example, letters can be formed.

[0150] Since industrial processing of an object 1 may be involved, it is advantageous if the processing of the object 1 by a laser 2, i.e. the laser processing of the object 1, is carried out as quickly as possible. To this end, a method for time optimization of laser processing of an object 1, in particular a computer-implemented method, is proposed, wherein the laser 2 is guided along a predetermined processing path 4 on the object 1 by a galvanometer scanning system 3.

[0151] It has been described that the object 1 can be a piece of fabric or generally a workpiece. Additionally or alternatively, the object 1 can also be a body part, whereby laser processing is used for medical purposes.

[0152] Furthermore, a calculation unit 14 is shown here, with which the method explained in the previous description and / or in the following description can be carried out. The control data determined by the method can be sent from the calculation unit 14 to the controller 22. The controller 22 then controls at least the modulation of the laser power and the position of the laser 2 synchronously.

[0153] Furthermore, a laser processing device 19 is shown at least in sections. The laser processing device 19 comprises a laser 2 or a laser source 20 generating the laser 2 , a focusing unit 21 , a galvo scanning system 3 , a computing unit 14 and preferably a controller 22 .

[0154] Figure 2 Examples of a position-dependent energy input curve E(s) 15 , a position-dependent ideal speed curve v(s) 16 and power curves 23 , 24 are shown.

[0155] Figure 2 Three graphs or coordinate systems are shown. In two of the coordinate systems, the abscissa axis represents the position s or the location along the machining path 4. In the first, upper graph or coordinate system, the ordinate axis represents the energy or energy input E. In the second, lower graph or coordinate system, the ordinate axis represents the speed v.

[0156] The position-dependent energy input curve E(s) 15 here shows how high the energy input must be at position s of the machining path 4. It should be noted here that Figure 2 The position-dependent energy input curve E(s)15 in Figure 1 The machining path 4 in FIG. 1 is modeled. Therefore, the position-dependent energy input curve E(s) 15 starts from the path start point 5 and ends at the path end point 6, and the same is true for the position-dependent ideal velocity curve v(s) 16. Therefore, if the position s in the curve diagram or coordinate system passes from the path start point 5 to the path end point 6, the machining path 4 also passes from the path start point 5 to the path end point 6.

[0157] because Figure 2 The position-dependent energy input curve E(s)15 is based on Figure 1 The machining path 4 is modeled so that the position-dependent energy input curve E(s) 15 increases from the path starting point 5 to the middle area. Figure 1 This is indicated by the increasing thickness of the machining path 4. Afterwards, the position-dependent energy input curve E(s) 15 decreases again, which is Figure 1 It is represented by the continuously tapering processing path 4.

[0158] Advantageously, the position-dependent energy input curve E(s)15 is predetermined. This means that the position-dependent energy input curve E(s)15 serves as input data. The position-dependent energy input curve E(s)15 can also advantageously be determined from preset values ​​or from a position-dependent preset value curve. For example, a customer predetermines an ablation curve along the processing path 4 as a preset value curve, wherein the ablation curve predetermines how much material is to be ablated along the processing path 4. A fading curve can also be predetermined as a preset value curve, which indicates how intensely the fading of the fabric piece should be along the processing path 4. Based on this preset value curve, the position-dependent energy input curve E(s)15 can be determined, wherein for this purpose preferably at least the material properties are simultaneously taken into account. Obviously, for plastics and metals such as iron or steel, different energy inputs are required to achieve similar material ablation.

[0159] Preferably, the energy input E into the object 1 in the region along the processing path 4 is determined from the position-dependent energy input curve E(s) 15 by means of the integration of the position-dependent energy input curve E(s) 15 in the corresponding region of the curve diagram.

[0160] For time-optimized processing of the processing path 4, it is advantageous if the laser 2 is moved as quickly as possible through the processing path 4. The maximum speed of the laser 2 along the processing path 4 is limited on the one hand by the position-dependent energy input curve E(s) 15 that must be achieved. On the other hand, the laser 2 has only a limited laser power.

[0161] In this method, a position-dependent first power curve L1(s) 23 of the laser 2 along the machining path 4 is determined. The position-dependent first power curve L1(s) 23 is always lower than or equal to the maximum possible power of the laser 2. Advantageously, the position-dependent first power curve L1(s) 23 is set as high as possible, so that the machining time of the machining path 4 is shortened, in particular to be set constant and equal to the available maximum power that the laser 20 can generate. As shown here, it is advantageous if the position-dependent first power curve L1(s) 23 of at least a path section is set constant. Based on the position-dependent first power curve L1(s) 23 and the position-dependent energy input curve E(s) 15, the position-dependent ideal velocity curve v is calculated. max (s) 16. Position-dependent ideal velocity curve v max A possible relationship between the position-dependent first power curve L1(s) 23 and the position-dependent energy input curve E(s) 15 can be, for example, as follows:

[0162]

[0163] The ideal speed at position s can be proportional in particular to the ratio of the position-dependent power curve L1 and the desired energy input E at position s. However, other, in particular nonlinear, relationships between power L, energy E and speed v are also applicable to the laser process. For the method of the invention, it is only important that there is a value range within which the speed v can be assigned to the energy input E and the power L by a known relationship. In this relationship, other parameters, such as the focal size or the angle of incidence of the laser 2, can also be included. This relationship can be determined, for example, by experimental parameter studies, in which laser processing is carried out with at least several different parameter (power and speed) values ​​and the energy input (i.e., ablation or color change, for example) is evaluated. The values ​​thus determined experimentally can be recorded in a value table and interpolation can be performed between the values ​​of the table. The value table or the appropriate expression of the interpolation is stored in the calculation unit 14 and allows the calculation of one of the three variables (speed v, power L and / or energy input E) if the other two variables are given. Since the dependence of the energy input E on the parameters speed v and power L usually has a strictly monotonic trend, the calculation of the above-mentioned quantities can usually be realized in a single-valued manner. If the relationships of various other parameters of the laser process are determined experimentally in a similar manner, the influence of these parameters can also be taken into account in the method. For example, the influence of the angle of incidence on the object surface or the influence of the defocusing of the laser beam can also be stored in the value table.

[0164] exist Figure 2 The ideal velocity curve v related to position max (s) 16 is shown in the middle diagram or coordinate system. As can be seen, the position-dependent ideal velocity curve v max (s) 16 is relatively low in the middle region. Here, the position-dependent ideal speed curve v max (s) 16 is lower in the middle region for the following reasons. The position-dependent energy input curve E(s) 15 is high in the middle region, which is related to Figure 1 The thick middle region of the processing path 4 corresponds to this. This is because a relatively high energy input E to the processing path 4 should be achieved here. However, since the laser power L or the position-dependent first power curve L1(s) 23 cannot be arbitrarily high or limited, the position-dependent ideal speed curve v max (s) 16 must be low in this region in order to be able to achieve a high energy input E. The laser 2 must be moved accordingly slowly along the processing path 4. At the path start 5 and the path end 6, the position-dependent ideal speed profile v max (s) 16 is higher, because here the position-dependent energy input curve E(s) 15 is lower. Here again, the processing path 4 of the laser 2 can be passed faster, so that the predetermined energy input E can still be achieved. Position-dependent ideal speed curve v maxThe "ideal" in (s) 16 here means that this is the maximum speed v that the laser 2 can have along the processing path 4 according to the position s. max , so that in the case of the position-dependent first power curve L1(s) 23 , the position-dependent energy input curve E(s) 15 is exactly reached.

[0165] In this method, the time-dependent path progression s(t) is also calculated, so that the achievable velocity profile v(s) 17 of the laser 2 along the processing path 4 is subject to the position-dependent ideal velocity profile v max (s) 16 and observe at least one dynamic limit of the galvanometer scanning system 3 along the processing path 4. The at least one dynamic limit arises from the fact that the deflection mirrors 7, 8 cannot be moved arbitrarily fast and / or their orientation cannot be changed arbitrarily fast. In particular, the galvanometer drives 9, 10 also have inertia, so that the laser 2 cannot, for example, be accelerated arbitrarily fast to achieve the desired velocity profile v max (s) 16. Thus, the at least one dynamic limit may be an acceleration limit and / or a jerk limit of the galvanometer scanning system 3, in particular of the deflection mirrors 7, 8 and / or of the galvanometer drives 9, 10.

[0166] Therefore, the achievable speed curve v(s)17 can at most reach the ideal speed curve v max (s) 16. Therefore, the achievable speed curve v(s) 17 is subject to the ideal speed curve v max (s)16 restrictions.

[0167] In order to achieve an achievable speed profile v(s) 17, the method calculates the time-dependent path course s(t). The time-dependent path course s(t) indicates how far the laser 2 has advanced along the processing path 4 at a given time. Therefore, the time-dependent path course s(t) indicates the position s of the laser 2 at time t.

[0168] Advantageously, the time-dependent path course s(t) is calculated by an optimization method so that the passage time of the laser 2 through the processing path 4 or the processing time of the processing path 4 is minimized. In addition or alternatively, the time-dependent path course s(t) can be calculated by an optimization method so that at least in certain sections, in particular everywhere, the achievable speed curve v(s) 17 reaches a maximum value.

[0169] exist Figure 2 In the middle figure of FIG. 1 , an exemplary achievable speed curve v(s) 17 is also shown as a dashed line. It can be seen that the achievable speed curve v(s) 17 is affected by the position-dependent ideal speed curve v max(s) 16. At least in the region of the path starting point 5 shown here, the two curves 16, 17 can be at least approximately identical, for example because the laser 2 has already been accelerated in advance. However, it should be noted that if, for example, a linear processing path 4 has a constant energy input E, so that the laser power L and / or the speed v do not need to be changed, the achievable speed curve v(s) 17 can be very close to the position-dependent ideal speed curve v max (s) 16. In this exemplary case, the achievable speed profile v(s) 17 can approximate the position-dependent ideal speed profile v max (s)16.

[0170] In the method, a second time-dependent power curve L2(s(t)) 24 is also calculated, wherein the position-dependent first power curve L1(s) 23 is reduced so that the achievable speed curve v(s) 17 in the machining path is less than the ideal position-dependent speed curve v max At the location or position of (s) 16, the position-dependent energy input curve E(s) 15 is realized. Since the achievable speed curve v(s) 17 is lower than the position-dependent ideal speed curve v max (s) 16, the laser 2 also stays longer in the area of ​​the processing path 4, so that while maintaining the position-dependent first power curve L1 (s) 23, the energy input E becomes too large, in particular exceeds the preset value. Therefore, in order to balance the speed curve v (s) 17 and the position-dependent ideal speed curve v max The position-dependent first power curve L1(s) 23 is reduced to a time-dependent second power curve L2(s(t)) 24 by a decrease compared to the position-dependent energy input curve E(s) 16. The correlation between the achievable speed curve v(s) 17, the position-dependent energy input curve E(s) 15 to be achieved and the time-dependent second power curve L2(s(t)) 24 can be, for example, as follows:

[0171] L2(s(t))~E(s)*v(s)

[0172] The time-dependent path course s(t) can be defined as a normalized path course, i.e. s(t) represents the distance traveled on the path until time t. The time-dependent path course s(t) can therefore be converted unambiguously to t(s). This applies:

[0173]

[0174] Using the inverse function t(s), v(s) can be determined as v(s)=v(t(s)).

[0175] The two power curves 23, 24 are shown here in the lower figure. As can be seen, the position-dependent first power curve L1(s) 23 is higher than or equal to the time-dependent second power curve L2(s(t)) 24. It can also be seen that the second power curve L2(s(t)) 24 decreases more strongly relative to the power curve L1(s) 23 at locations where the achievable speed curve 17 deviates further downward from the ideal speed curve 16. It can also be seen that along the entire processing path 4, the second power curve L2(s(t)) 24 is only slightly lower than the first power curve L1(s) 23. This shows that, by optimization, the available power of the laser can be utilized to a large extent in the process.

[0176] Figure 3 An example of a machining path 4 is shown, which has a discrete position-dependent energy input curve E(s) 15. Here, Figure 3 In the upper area of ​​the figure, six boxes are shown, which are exemplary path segments 18 along the processing path 4. The case where the processing path 4 has only six path segments 18 is only for the purpose of simplifying the description. The processing path 4 can include more path segments 18, for example up to 10,000, because the size of the path segments can be several micrometers. Due to this number of path segments 18, the discrete position-dependent energy input curve E(s) 15 shown here becomes an approximately continuous position-dependent energy input curve E(s) 15, which can therefore also be calculated using a continuous function.

[0177] For example, the position-dependent energy input curve E(s) 15 can be changed from a preset value curve to a derivation from grayscale values. Each path segment 18 can be assigned a grayscale value, which reflects, for example, the color tone on the textile piece. With the help of the preset value curve or from the grayscale value of each path segment, the position-dependent energy input curve E(s) 15 can be determined so that the corresponding color tone is obtained on the textile piece. For example, if more color needs to be eliminated, the energy input E must be higher, so that the fabric piece becomes lighter at this location. The boxes or path segments 18 shown here can be compared with pixels of a bitmap or derived from pixels of a bitmap.

[0178] Each path segment 18 is assigned an energy input E, which can be determined based on the position-dependent energy input curve E(s) 15. Thus, the first path segment 18a has a first energy input E1=2. The second path segment 18b has a second energy input E2=2. The third path segment 18c has a third energy input E3=7. The fourth path segment 18d has a fourth energy input E4=10. The sixth path segment 18e has a fifth energy input E5=6. The sixth path segment 18f has a sixth energy input E6=3. Thus, the position-dependent energy input curve E(s) 15 is similar to Figure 1 and Figure 2 The position-dependent energy input curve E(s) 15 shown here is predetermined or determined according to a preset value curve.

[0179] Based on the position-dependent energy input curve E(s) 15 and the position-dependent first power curve L1(s) 23, the position-dependent ideal speed curve v is determined. max (s) 16. Since the position-dependent energy input curve E(s) 15 is discrete, i.e. constant along the corresponding path segment 18, the position-dependent ideal speed curve v max (s) 16 is also constant in sections, ie in the corresponding path section 18. Position-dependent ideal speed profile v max (s) 16 and the position-dependent energy input curve E(s) 15 are here step-shaped or step functions.

[0180] The achievable velocity profile v(s) 17 is indicated by a dashed line. It is calculated here so that it is linear in the corresponding path section 18, wherein the temporal variation of the velocity v is limited here by the dynamic limits of the galvanometer scanning system 3δ, in this case in particular the acceleration limits. Furthermore, it is calculated so that it is subject to the position-dependent ideal velocity profile v max (s) 16. It can also be seen that in the fourth path section 18d, the achievable speed profile v(s) 17 and the position-dependent ideal speed profile v max (s) 16 are equal, wherein the achievable speed profile v(s) 17 is still subject to the position-dependent ideal speed profile v max (s)16 restrictions.

[0181] exist Figure 3 In the example shown, the limits of acceleration can be seen in the first two path sections 18a, 18b. In the first path section 18a, the achievable speed profile v(s) 17 does not follow the ideal speed profile v max (s) 16, otherwise the required acceleration value in the region of the second path section 18b would exceed the acceleration limit of the galvanometer scanning system 3 or even if the maximum available acceleration is used in the second path section 18b, the ideal velocity curve v in the third path section 18c would be exceeded. max (s) 16. In order to avoid this, in the method, a speed change v is already selected in the first path segment 18a so that the required total speed change Δv is assigned to the first two path segments 18a, 18b until the beginning of the third path segment 18c. If this is not sufficient to comply with the acceleration limit (as in the example), then a speed lower than the ideal speed v must be selected in the method already at the beginning of the first path segment 18a. max(s) 16 in order to be able to determine the achievable speed curve v(s) 17.

[0182] Since the achievable speed profile v(s) 17 is at least partially lower than the position-dependent ideal speed profile v max (s) 16, so that the position-dependent first power curve L1(s) 23 is reduced, wherein the time-dependent second power curve L2(s(t)) 24 is calculated, so that the energy input curve E(s) is achieved or the energy input E into the corresponding path segment 18 is achieved and in particular not exceeded. In this case, in particular in the case of very short path segments 18, a constant energy input E for each path segment 18 can be calculated from the intermediate speed v of the path segment 18, which energy input E corresponds to the time-averaged energy input E of the path segment 18.

[0183] In the third figure, the achievable velocity curve v(s) 17 is shown again, but this time it is smoothed. This has the advantage that the achievable velocity curve v(s) 17 is free of inflection points, so that the velocity v does not change abruptly. Due to this smoothing, the jerk corresponding to the time derivative of the acceleration is limited to a finite value. By appropriately selecting the smoothing parameters, it is also possible to achieve compliance with another dynamic limit of the Galvo scanning system 3, namely the jerk limit.

[0184] The path process s(t) determined using this method can be used to determine the control data p(t)=p(s(t)) for the Galvo scanning system 3 and the control data L2(t)=L2(s(t)) for modulating the laser power L, which can be transmitted from the computing unit 14 to the controller 22.

[0185] For the calculation, the same relationships between the process variables (energy input E, speed v and power L) can be used, which already calculates the ideal speed v max (s)16 is described above.

[0186] For execution by the controller 22, these data can optionally be interpolated or scaled by the controller 22 or the computing unit 14 to correspond to the clock of the controller 22 and / or the internal coordinate system of the galvo scanning system 3. For example, the controller 22 can transmit two digital values ​​with a resolution of 20 bits to the galvo scanning system 3 every 10 μs, which digital values ​​describe the target angular position of the galvo drivers 7, 8 and synchronously modulate the power of the laser 2 at a higher clock rate (e.g. 8 MHz or 64 MHz). The control of the laser modulation can also be performed by the controller 22, for example using an analog signal obtained from the digital signal, which is generated by the controller 22 based on the time-dependent second power curve L2 (t).

[0187] The controller 22 can also digitally modulate the laser source 20 between two powers, in particular can also switch it on and off. The controller 22 can control the laser source 20, for example, using a pulse width modulated signal, using which a time-dependent second power curve L2(t) 24 is formed. In this case, via the duty cycle, an average laser power can be set, which corresponds to the time-dependent second power curve L2(t).

[0188] The lower figure again shows two power curves L1(s) 23, L2(s(t)) 24. Here again it can be seen that the achievable speed curve v(s) 17 deviates from the ideal speed curve v max The more the area with higher speed (s) 16, the more the second power curve L2 (s (t)) 24 decreases compared to the position-dependent first power curve L1 (s) 23. For example, in the fourth path segment 18d, the time-dependent second power curve L2 (s (t)) 24 and the position-dependent first power curve L1 (s) 23 are identical, because the achievable speed curve v (s) 17 and the ideal speed curve v max In the first and second path sections 18a, 18b, the time-dependent second power curve L2(s(t)) 24 deviates relatively strongly from the position-dependent first power curve L1(s) 23, because the achievable speed curve v(s) 17 deviates relatively strongly from the ideal speed curve v max (s) 16. In addition, in the comparison of the fifth and sixth path sections 18e, 18f, the deviation is smaller. Here, the reduction of the second power curve L2(s(t)) 24 relative to the position-dependent first power curve L1(s) 23 can depend on the achievable speed curve v(s) 17 and the ideal speed curve v max (s)16.

[0189] The invention is not limited to the embodiments shown and described. Modifications and combinations of features are possible within the scope of the claims, even if these are presented and described in different embodiments.

[0190] Reference numerals list

[0191] 1 Object

[0192] 2 Laser

[0193] 3Galvanometer scanning system

[0194] 4 Processing paths

[0195] 5. Trail start

[0196] 6 Path End

[0197] 7First deflecting mirror

[0198] 8 Second deflection mirror

[0199] 9First galvanometer driver

[0200] 10 Second galvanometer driver

[0201] 11 Scanning optics

[0202] 12 Laser coupling device

[0203] 13 Ammeter control unit

[0204] 14 computing units

[0205] 15 Position-dependent energy input curve E(s)

[0206] 16 Position-dependent ideal velocity curve v max (s)

[0207] 17Achievable speed curve v(s)

[0208] 18 Path segment s(t)

[0209] 19Laser processing equipment

[0210] 20 Laser source

[0211] 21 Focusing Unit

[0212] 22 Controller

[0213] 23 Position-dependent first power curve L1(s)

[0214] 24 Time-dependent second power curve L2 (s (t))

[0215] s position

[0216] Energy Input

[0217] Speed

[0218] t time

Claims

1. A method for time-optimization of laser processing of an object (1), in particular a computer-implemented method, wherein: A laser (2) can be guided along a predetermined processing path (4) on the object by means of a current scanning system (3); wherein an energy input curve E(s) (15) associated with a position of entering the object along the predetermined processing path (4) is determined by the laser; wherein a first position-dependent power curve L1(s)(23) of the laser (2) along the machining path (4) is determined, and wherein a position-dependent ideal velocity curve v is calculated based on the first position-dependent power curve L1(s)(23) and the position-dependent energy input curve E(s)(15) max (s); wherein the time-dependent path process s(t) is calculated so that the achievable velocity curve v(s)(17) of the laser (2) along the processing path (4) is subject to the position-dependent idealized velocity curve v max (s) and comply with at least one dynamic limit of the current scanning system (3); wherein a time-dependent second power curve L2(s(t))(24) is calculated, wherein the position-dependent first power curve L1(s)(23) is reduced so that the achievable speed curve v(s)(17) on the machining path (4) is less than the idealized speed curve v max The position (s) of (s)(16) is obtained to obtain the energy input curve E(s)(15) related to the position.

2. The method according to the preceding claim, characterized in that The time-dependent path progress s(t) is calculated so that the machining time of the machining path (4) is minimized.

3. The method according to any one or more of the preceding claims, characterized in that The time-dependent path progression s(t) is calculated such that the achievable speed profile v(s) (17) reaches a maximum at least in sections, in particular everywhere.

4. The method according to any one or more of the preceding claims, characterized in that The time-dependent path progression s(t) is calculated by an optimization method, wherein preferably a plurality of different time-dependent path progressions s(t) are calculated in the optimization method and the time-dependent path progression s(t) which provides the shortest processing time for the processing path (4) is selected.

5. The method according to any one or more of the preceding claims, characterized in that The time-dependent path progression s(t) is calculated such that the dynamic limits of the laser (2), in particular the dynamic limits of the laser power, are observed.

6. Method according to one or more of the preceding claims, characterized in that The time-dependent path progression s(t) is calculated such that acceleration limits and / or jerk limits are observed as dynamic limits of the current scanning system (3).

7. Method according to one or more of the preceding claims, characterized in that The time-dependent path progression s(t) is calculated such that the achievable speed profile (17) and / or a profile derivable therefrom, in particular via a time derivative, is continuous and / or continuously differentiable, in particular smooth.

8. Method according to one or more of the preceding claims, characterized in that Calculate and / or specify the position-dependent energy input curve E(s) (15) and / or the position-dependent first power curve L1(s) (23) and / or the position-dependent idealized speed curve v max (s)(16) such that these are stepped and / or constant along at least some path portions of said machining path (4).

9. Method according to one or more of the preceding claims, characterized in that In calculating the time-dependent second power curve L2(s(t))(24) and / or the position-dependent idealized speed curve v max (s)(16), the orientation and / or positioning of the current scanning system (3) relative to the processing path (4) is calculated together.

10. Method according to one or more of the preceding claims, characterized in that The time-dependent coordinate p(s(t)) of the current scanning system (3) relative to the processing path (4) is calculated from the time-dependent path course s(t) in order to move the laser (2) along the processing path (4).

11. Method according to one or more of the preceding claims, characterized in that The method for time-optimized laser processing of an object (1) is executed by a computing unit (14).

12. Method according to one or more of the preceding claims, characterized in that The coordinates p(s(t)) are transmitted to a control unit of the current scanning system (3), and the time-dependent second power curve L2(s(t)) (24) is transmitted to a laser controller.

13. Method according to one or more of the preceding claims, characterized in that In preparation for the actual laser processing of the object (1), the laser processing, in particular with regard to the operating behavior, is simulated computer-aided and / or after time optimization of the laser processing and / or simulation of the laser processing, the object (1) is actually laser processed.

14. A computer program which, when executed by a computing unit (14), causes it to perform the method according to one or more of the preceding claims.

15. A computer-readable storage medium comprising a computer program, wherein: The computer program, when executed by a computing unit (14), causes it to perform a method according to one or more of the preceding claims.

16. A laser processing device for laser processing an object (1), comprising a laser (2) for laser processing, a current scanning system (3) for guiding the laser (2) on the object (1), and a computing unit (14), characterized in that: The computing unit (14) is designed to carry out the method according to at least one of the preceding claims.

Images