A five-axis linkage-based adaptive laser focusing method for complex curved surfaces

Abstract

This invention provides an adaptive laser focusing method for complex curved surfaces based on five-axis linkage, relating to the field of laser precision machining technology. The method first acquires the surface model and process parameters, and plans the five-axis scanning trajectory. During machining, the surface normal vector and curvature are acquired in real time, and the focus following distance for focus position compensation and the beam incident optimization angle for energy distribution optimization are dynamically calculated accordingly. Subsequently, the position of the focusing lens group is synchronously adjusted to precisely control the focus, and the beam is pointed in real time by the five-axis system. Finally, the calculated target space vector and position are converted into control commands to drive adaptive machining. This invention effectively overcomes the problems of focus drift and uneven energy distribution in the machining of complex curved surfaces through online sensing and dynamic adjustment, significantly improving machining accuracy, uniformity, and process robustness.

Description

Technical Field

[0001] This invention relates to the field of laser precision machining technology, and in particular to an adaptive laser focusing method for complex curved surfaces based on five-axis linkage. Background Technology

[0002] Laser processing technology, with its advantages of high precision, non-contact operation, and high flexibility, plays an increasingly important role in the cutting, welding, cladding, and surface treatment of complex curved surfaces in fields such as aerospace, mold manufacturing, and medical devices. Five-axis CNC technology makes it possible to achieve arbitrary pointing of the laser beam in three-dimensional space, and is a key equipment foundation for achieving high-quality laser processing of complex curved surfaces. However, existing five-axis CNC-based laser processing methods still face a series of severe challenges in terms of processing quality and process stability when dealing with complex curved surfaces.

[0003] Existing technologies typically rely on fixed toolpaths generated through offline programming for machining. During the programming phase, the system plans the spatial trajectory and beam direction of the laser head based on the theoretical CAD model of the workpiece. However, in actual machining, there are unavoidable manufacturing errors, clamping errors, and thermal deformation between the actual shape of the workpiece and the theoretical model. These factors cause deviations between the theoretically programmed position and the actual surface position, making it impossible for the laser focus to accurately land on the target surface. When the laser focus deviates from the optimal focusing position, the size and power density of the laser spot acting on the material surface change significantly, severely affecting machining accuracy and consistency, such as causing uneven kerf width, fluctuating weld penetration, or unstable cladding thickness.

[0004] A more prominent problem is that the geometric characteristics of complex curved surfaces themselves constitute dynamic interference to the laser processing. First, the continuous change in surface curvature alters the incident angle and reflection conditions of the laser beam. Even if the focal point is mechanically maintained at a theoretical point on the surface, the shape and energy distribution of the projected beam spot on the surface will be distorted with the curvature. On convex surfaces, the beam spot is compressed, leading to excessively high local energy density, which may cause overheating or even vaporization and perforation of the material; on concave surfaces, the beam spot is elongated, resulting in insufficient energy density, causing incomplete cutting or incomplete welding. Second, existing methods often employ a fixed beam incident angle (usually perpendicular to the local tangent plane or maintaining a fixed angle), which ignores the significant differences in the absorption rate of laser energy by the material at different incident angles (i.e., the Fresnel effect). In regions where the surface curvature changes drastically, the fixed incident angle strategy causes drastic fluctuations in energy absorption efficiency, making heat input difficult to control, and consequently affecting the microstructure and mechanical properties.

[0005] Furthermore, the entire processing is a dynamic and variable system. Fluctuations in laser power, acceleration and deceleration of feed speed, and local differences in material surface conditions (such as oxide layers) all have a real-time impact on the final processing result. Current mainstream technologies lack the ability to sense and compensate for these dynamic factors online in real time. Processing parameters are often set based on experience or limited process experiments, and once set, they remain unchanged throughout the entire processing. This open-loop control method cannot adapt to random disturbances during processing and makes it difficult to guarantee uniform, stable, and high-quality processing results across the entire processing area of ​​complex curved surfaces.

[0006] In summary, existing technologies have significant shortcomings in terms of geometric adaptability, energy control precision, and process dynamic stability when dealing with laser processing of complex curved surfaces with geometric complexity and dynamic physical processes. Therefore, seeking an adaptive focusing method capable of real-time sensing of processing status, intelligent decision-making, and dynamic adjustment of beam parameters has become an urgent technical need to improve the quality and reliability of laser processing of complex curved surfaces. Summary of the Invention

[0007] To address the technical problems in existing technologies, such as laser focus misalignment, uneven beam energy distribution, and unstable processing quality caused by geometric errors, dynamic changes in surface curvature, and disturbances during processing, this invention provides an adaptive laser focusing method for complex curved surfaces based on five-axis linkage.

[0008] The technical solution provided by this invention is as follows: This invention provides an adaptive laser focusing method for complex curved surfaces based on five-axis linkage, comprising: S1. Obtain the three-dimensional geometric model data of the complex curved surface to be processed and the target laser processing parameters; S2. Based on the three-dimensional geometric model data, plan the spatial scanning trajectory of the laser processing head under the five-axis linkage CNC system. The spatial scanning trajectory includes the sequence of spatial position points of the processing head and the theoretical initial vector of the laser beam at each position point. S3. During the movement along the spatial scanning trajectory, the actual surface information at the current processing point is collected in real time by a non-contact ranging sensor installed on the laser processing head. The actual surface information includes at least the surface normal vector and the local curvature. S4. Based on the actual surface information and the target laser processing parameters, dynamically calculate and generate two key control quantities: the focus following distance, used to compensate for the deviation between the actual position of the laser focus and the theoretical surface position, and the beam incident optimization angle, used to optimize the energy distribution of laser energy on the surface; wherein, the calculation of the focus following distance takes into account at least the local curvature, and the calculation of the beam incident optimization angle takes into account at least the surface normal vector. S5. Based on the calculated focus following distance, adjust the axial position of the laser focusing lens group in real time so that the laser focus falls precisely on the target point determined by the focus following distance offset from the theoretical curved surface position along the normal vector; S6. Simultaneously, based on the calculated beam incident optimization angle and combined with the surface normal vector at the current processing point, the target space vector of the laser beam is calculated. S7. The target space vector and the spatial position of the target action point are converted into real-time control commands for each motion axis in the five-axis linkage CNC system to drive the laser processing head and the workpiece to move, thereby realizing adaptive focusing processing of the laser beam and complex curved surfaces.

[0009] The beneficial effects of the technical solution provided by this invention include at least the following: (1) In this invention, the actual surface normal vector and local curvature of the processing point are collected online in real time, and the focus following distance is dynamically calculated based on this, thereby adjusting the position of the focusing lens group in real time. This technical means directly solves the problem of laser focus drift caused by the actual pose deviation of the workpiece and the change of surface curvature. It can ensure that the laser focus always accurately follows the real contour of the surface, and compensates for the error between the theoretical model and the actual object, as well as the focus offset caused by curvature. The technical effect achieved is that the actual size of the laser spot on the material surface remains highly consistent throughout the entire complex surface processing area, which significantly improves the geometric accuracy of the processing contour, such as achieving uniform kerf width or cladding layer thickness, and fundamentally overcomes the processing quality fluctuation caused by focus misalignment in traditional static focusing methods.

[0010] (2) In this invention, the beam incident angle is dynamically optimized by analyzing real-time surface information and material properties, and the spatial direction of the beam is adjusted in real time by combining a five-axis linkage system. This technique directly addresses the problem of uneven laser energy absorption efficiency caused by surface curvature changes and the Fresnel effect. The system adjusts the incident angle according to the local curvature and optimal energy coupling model, optimizes the interaction conditions between the beam and the surface, and makes the energy absorption rate tend to be stable. The resulting technical effect is that the distribution of laser energy on complex surfaces is actively controlled and optimized, avoiding defects such as overheating of convex surfaces and lack of fusion on concave surfaces, achieving precise control of heat input, thereby obtaining a high-quality processing area with uniform structure and consistent performance, and improving the overall mechanical properties of the weld or cladding layer.

[0011] (3) In this invention, a closed-loop adaptive control loop including real-time process monitoring and online parameter self-tuning is established to provide real-time feedback and correction for power, speed, and surface morphology deviations during the processing. This technical approach solves the problem of process instability caused by internal and external disturbances (such as power fluctuations, speed changes, and material inhomogeneity). The system can sense minute deviations in the processing effect and automatically backtrack to adjust the calculation logic of the core control parameters, enabling the entire processing system to have self-learning and self-adaptive capabilities. The resulting technical effect is that the robustness and stability of the laser processing of complex curved surfaces are greatly enhanced, enabling the continuous output of high-quality and highly consistent processing results under various uncertainties, reducing the dependence on operator experience, and improving the reliability and repeatability of the process. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating a complex surface adaptive laser focusing method based on five-axis linkage, provided in an embodiment of the present invention; Figure 2 A flowchart illustrating the dynamic calculation of the focus following distance in a complex curved surface adaptive laser focusing method based on five-axis linkage, provided for an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the process of dynamically calculating the optimal incident angle of a complex curved surface in an adaptive laser focusing method based on five-axis linkage, as provided in an embodiment of the present invention. Detailed Implementation

[0014] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0015] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0016] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0017] In embodiments of the present invention, sometimes the subscript is as follows: It may be mistakenly written as a non-subscript form such as W1. When the distinction is not emphasized, the meaning they express is the same.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] Reference manual attached Figure 1 The diagram shows a flowchart of a complex surface adaptive laser focusing method based on five-axis linkage provided by an embodiment of the present invention.

[0020] This invention provides an adaptive laser focusing method for complex curved surfaces based on five-axis linkage. The processing flow may include the following steps: S1. Obtain the three-dimensional geometric model data of the complex curved surface to be processed and the target laser processing parameters.

[0021] Step S1 involves data acquisition and input via a software module integrated within the CNC system. The 3D geometric model data of the complex surface to be processed originates from the workpiece's 3D CAD design file or 3D scanned point cloud data. The system directly reads the mathematical expression of the surface or performs point cloud triangulation to reconstruct the surface. The target laser processing parameters are set by the operator through the human-machine interface, including at least the laser power, nominal spot diameter, initial beam incident angle, scanning feed speed, and target values ​​for the depth or width of material removal. These data are uniformly loaded into the system's core processing unit, serving as the benchmark for all subsequent adaptive control processes.

[0022] S2. Based on the three-dimensional geometric model data, plan the spatial scanning trajectory of the laser processing head under the five-axis linkage CNC system. The spatial scanning trajectory includes the sequence of spatial position points of the processing head and the theoretical initial vector of the laser beam at each position point.

[0023] Step S2 is executed by the trajectory planning module of the CNC system. Based on the input 3D geometric model, this module first generates initial path points covering the entire surface to be processed using the spatial isoparametric line method or the section method. For each path point, the module calculates the spatial coordinates of the laser processing head end and the theoretical initial vector of the laser beam axis based on the theoretical surface normal vector at that point and the preset initial beam incident angle. The theoretical initial vector is described by a homogeneous transformation matrix from the tool coordinate system to the machine coordinate system. Finally, the module outputs an ordered sequence of spatial position points and the corresponding theoretical initial vector sequence, forming the basic data for the five-axis linkage spatial scanning trajectory.

[0024] S3. During the movement along the spatial scanning trajectory, the actual surface information at the current processing point is collected in real time by a non-contact ranging sensor installed on the laser processing head. The actual surface information includes at least the surface normal vector and local curvature.

[0025] Step S3 is achieved through a coaxial or paraaxial non-contact laser rangefinder integrated into the laser processing head. As the processing head moves along the planned trajectory, the CNC system synchronously triggers the sensor in each control cycle. The sensor emits a measurement beam towards the actual curved surface region in front of the current theoretical processing point and receives its reflected signal. By measuring the time of flight or phase change of the measurement beam, the sensor calculates the precise distance from the measurement point to the sensor probe in real time. Based on the measured distance value and the sensor's own pose parameters, combined with the real-time pose of the processing head, the system obtains the actual surface normal vector at the current processing point in real time through spatial geometric calculation, and further estimates the local curvature value of that point through the difference in normal vectors between adjacent measurement points. The actual surface normal vector and local curvature are immediately updated to the data buffer.

[0026] S4. Based on the actual surface information and the target laser processing parameters, two key control quantities are dynamically calculated and generated: the focus following distance, used to compensate for the deviation between the actual position of the laser focus and the theoretical surface position, and the beam incident optimization angle, used to optimize the energy distribution of laser energy on the surface. The calculation of the focus following distance considers at least the local curvature, and the calculation of the beam incident optimization angle considers at least the surface normal vector.

[0027] Step S4 is executed by the adaptive focusing control module. This module reads the real-time actual surface information provided in step S3 within each control cycle, including the surface normal vector n and local curvature κ, and simultaneously calls the target laser processing parameters set in step S1. The module's built-in core algorithm calculates two control variables in parallel. The first control variable is the focus following distance. The purpose of this calculation is to overcome focus drift caused by the curvature of the surface, ensuring that the actual projected size of the beam spot on the curved surface meets the process requirements. This calculation uses the local curvature κ as the main input, following the fundamental physical principle that the focus compensation amount is inversely proportional to the absolute value of the curvature. The second control quantity is the optimized beam incident angle. The purpose of this calculation is to optimize laser energy coupling efficiency, reduce reflection loss, and control heat input distribution. The calculation uses the real-time surface normal vector *n* as the main input, and is achieved by dynamically adjusting the angle relative to the normal. The dynamic calculation of the two control variables is a continuous process based on real-time sensor data, and its output values ​​are continuously updated with the instantaneous changes in the surface geometry.

[0028] S5. Based on the calculated focus following distance, adjust the axial position of the laser focusing lens group in real time so that the laser focus falls precisely on the target point determined by the focus following distance offset from the theoretical curved surface position along the normal vector.

[0029] Step S5 is achieved by controlling the linear motion mechanism of the laser focusing lens assembly along the optical axis. The adaptive focusing control module uses the calculated focus following distance. This position command is sent as a positional command to the servo motor or piezoelectric ceramic driver that drives the focusing lens assembly. The target position of this command is to ensure that the laser focus falls precisely on the theoretical curved surface position point along the direction of the surface normal vector measured in real time at that point, and offset from it. At the designated spatial point, the motion mechanism, upon receiving the command, moves the focusing lens group to the specified axial position within a very short control cycle, thereby changing the focal length of the laser beam in real time, completing the closed-loop adjustment of the focal point's spatial position, and ensuring that the laser energy is precisely applied to the target surface.

[0030] S6. Simultaneously, based on the calculated beam incident optimization angle and combined with the surface normal vector at the current processing point, the target space vector of the laser beam is calculated.

[0031] Step S6 is completed in the kinematics conversion module of the CNC system. This module receives the beam incident optimization angle from the adaptive focusing control module. And the actual surface normal vector n from the current point in step S3. The module first constructs a local coordinate system with n as the reference axis. Then, according to A desired beam axis direction is determined within this local coordinate system. Finally, the direction vector in this local coordinate system is transformed to the machine tool's global coordinate system through coordinate transformation, thereby calculating the target space vector of the laser beam in the machine tool coordinate system. This vector directly defines the direction in space that the laser beam should point to achieve optimal energy coupling.

[0032] S7. The target space vector and the spatial position of the target action point are converted into real-time control commands for each motion axis in the five-axis linkage CNC system, driving the laser processing head and workpiece to move, and realizing adaptive focusing processing of the laser beam and complex curved surfaces.

[0033] Step S7 is completed collaboratively by the interpolator and axis controller of the CNC system. The system then uses the target space vector obtained in step S6. and the spatial coordinates of the target point of action determined in step S5 As input, based on the kinematic model of the five-axis machine tool, the inverse kinematics module calculates in real time the conditions for simultaneously satisfying the arrival of the laser head tip. Point and beam axis aligned The interpolator determines the precise positions that the machine tool's three linear axes and two rotary axes need to reach. Based on these axis position commands, the interpolator generates smooth, continuous motion trajectories for each axis. Finally, the axis controller converts these trajectory commands into current or voltage signals to drive the motors, propelling the laser processing head and table to perform complex five-axis linkage motions, thereby achieving dynamic, adaptive focusing of the laser beam relative to complex curved surfaces. In one possible implementation, such as Figure 2 As shown, the dynamic calculation of the focus following distance includes: S401. Based on the local curvature at the current processing point and the spot diameter requirement in the target laser processing parameters, calculate the basic offset required to ensure effective spot coverage. S402. Based on the material removal mode and laser power in the target laser processing parameters, determine a process adjustment factor; S403. The basic offset and the process adjustment factor are coupled to obtain the final focus following distance. The coupling operation is configured to make the focus following distance negatively correlated with the absolute value of the local curvature and positively correlated with the relative magnitude of the instantaneous feed rate relative to a characteristic speed threshold.

[0034] The calculation of the basic offset in step S401 is based on the principles of geometric optics and surface geometry. The system considers the reciprocal of the local curvature κ as the local radius of curvature R. To ensure effective coverage of the spot diameter d on the surface, it is necessary to compensate for the focal position deviation caused by curvature. This basic offset Δbase is calculated using the following formula: Calculated, where It is a laser beam quality factor The tiny correction constant determined by the beam divergence angle.

[0035] Repair normal quantity Through formula To make an estimate, among which The wavelength of the laser. Let be the Rayleigh length of the laser beam. Let be the diameter of the laser beam waist. This formula quantifies the focal expansion effect caused by the combined effect of beam divergence and surface curvature.

[0036] In step S402, the process adjustment factor η is determined based on a pre-established process database. This database categorizes material removal modes into fine cutting, deep penetration welding, or surface cladding, and associates them with different laser power ranges P, mapping them to specific... Value. Process adjustment factor. The mapping rules follow the following physical principles: for fine cutting modes that are primarily based on vaporization cutting, The value of approaches the lower limit of the range of 0.8 to 1.2 to maintain a high peak power density; for deep penetration welding modes that are mainly based on melting, The value of is in the middle range of 1.2 to 1.5 to achieve a molten pool with a suitable aspect ratio; for surface cladding or heat treatment modes that rely primarily on heat conduction, The value of is close to the upper limit of the range of 1.5 to 2.0 to obtain a wide and shallow energy distribution. Specific values ​​in the mapping rules are determined through process calibration experiments for specific material and laser power combinations and stored in a database.

[0037] The core of the coupling operation in step S403 lies in constructing a real-time scaling relationship. The focus following distance Dff is given by the formula... Given, among which This refers to the dynamic scaling factor. Scaling factor Constructed to be inversely proportional to |κ| and to the instantaneous feed rate and characteristic velocity threshold The ratio is related, specifically when When it increases, It grows in a non-linear manner to achieve correlation.

[0038] In one possible implementation, such as Figure 3 As shown, the dynamic calculation of the beam incident optimization angle includes: S411. Using the surface normal vector at the current processing point as a reference, determine an initial beam vector according to the initial incident angle specified in the processing parameters. S412. Based on local curvature and material properties, analyze the reflection and absorption characteristics of laser energy in the local region of the curved surface under the initial beam vector. S413. Based on the analysis results, the initial incident angle is dynamically optimized and adjusted to obtain the beam incident optimized angle; wherein, the dynamic optimization adjustment is based at least on the deviation between the local radius of curvature and a preset optimal radius of curvature, and the spatial rate of change of the surface normal vector.

[0039] Determining the initial beam vector in step S411 is a coordinate transformation process. The system establishes a local reference coordinate system with the real-time acquired surface normal vector n as the Z-axis. The initial incident angle is set according to the process parameters. In the plane formed by the normal vector n and the tangent of the feed direction, the direction vector of the beam axis in the local coordinate system is calculated, and then transformed back to the machine coordinate system by a rotation matrix to obtain the initial beam vector. .

[0040] The analysis of reflection and absorption characteristics in step S412 is based on the electromagnetic wave-matter interaction model. The system calls the material property database to obtain the complex refractive index of the currently processed material. This complex refractive index is expressed as... Where n is the real part of the refractive index, k is the extinction coefficient, and i is the imaginary unit. Combining this with the local radius of curvature R, the Fresnel formula is used to calculate the value at the initial incident angle. and polarization-dependent reflectivity under current local geometric conditions And based on this, the energy absorption rate is estimated. The absorption rate A serves as the direct basis for subsequent optimization and adjustments.

[0041] The dynamic optimization adjustment in step S413 aims to bring the absorption rate A closer to its maximum or the target value required by the process. The optimization process uses the local radius of curvature R and the preset optimal radius of curvature as a reference. The difference ΔR and the change Δn / Δs of the surface normal vector n on the unit feed path are used as key inputs. The incident angle adjustment that optimizes the absorption characteristics is solved by a multidimensional optimization function.

[0042] In one possible implementation, the coupling operation in step S403 specifically includes: Calculate a baseline value consisting of the base offset; A modulation factor is introduced, which is a nonlinear function of the difference between the instantaneous feed rate and the characteristic velocity threshold; The baseline value is dynamically scaled using a modulation factor to obtain the focus following distance; when the instantaneous feed rate is greater than the characteristic velocity threshold, the growth rate of the focus following distance driven by the modulation factor increases accordingly.

[0043] The coupling operation is implemented through an explicit mathematical process. The baseline value is the base offset Δbase calculated in step S401. Modulation factor Defined as instantaneous feed rate With characteristic velocity threshold The difference is a nonlinear function. A specific functional form is: ,in It is a positive constant related to the material's sensitivity to heat input, and tanh is the hyperbolic tangent function. This function ensures that when... hour, Value greater than 1 and follows It increases monotonically, with a relatively high growth rate near the origin, then gradually saturates. Ultimately, the focus-following distance is determined by the formula... The calculation yielded the result. This formula directly reflects the dynamic scaling of the baseline value, and the scaling factor... growth rate at Exceed It did improve and met the technical requirements.

[0044] In one possible implementation, the dynamic optimization adjustment in step S413 specifically involves: Calculate the first correction amount required for the initial incident angle based on the deviation between the local radius of curvature and the preset optimal radius of curvature; The second correction required to calculate the spatial rate of change based on the surface normal vector; The first correction amount and the second correction amount are combined to obtain the final adjustment amount of the initial incident angle, thereby determining the optimal incident angle of the beam.

[0045] The specific implementation of state optimization adjustment is completed in two steps. The first correction amount... The local radius of curvature R and the preset optimal radius of curvature The deviation is driven by [the following]. Its calculation formula is: , where β is a proportionality coefficient, the sign and magnitude of which are determined through process experiments, so that the incident angle decreases at convex surfaces and increases at concave surfaces, in order to balance the energy distribution.

[0046] The sign of the scaling factor β is determined as follows: when machining convex surfaces (i.e., when the local radius of curvature R is positive), β takes a negative value; when machining concave surfaces (i.e., when the local radius of curvature R is negative), β takes a positive value. The absolute value of the scaling factor β is inversely proportional to the thermal diffusivity of the material and the laser power density. For heat-sensitive materials or high power density conditions, a smaller absolute value within the range of 0.1 to 0.3 should be used; for heat-insensitive materials or low power density conditions, a larger absolute value within the range of 0.4 to 0.8 can be used. The setting of the adjustment gain λ is related to the expected geometric complexity of the surface. For workpieces with drastic changes in normal vectors, such as free-form surfaces, λ takes a larger value within the range of 0.5 to 1.0 degrees per millimeter; for gently sloping surfaces, λ takes a smaller value within the range of 0.1 to 0.3 degrees per millimeter. The specific values ​​of the scaling factor β and the adjustment gain λ are calibrated after machining the test piece and measuring the uniformity of the heat-affected zone.

[0047] Second correction amount It is determined by the spatial rate of change of the surface normal vector. This rate of change is approximated by calculating the magnitude of the difference quotient between the normal vectors of the current point and the previous point, i.e. . The calculation formula is , where λ is the adjustment gain, and the arctan function is used to limit over-adjustment when the rate of change is too large. The final optimized beam incident angle is obtained through synthesis: This additive synthesis method clearly implements the technical logic of first correcting for curvature deviation and then making secondary adjustments based on the rate of geometric change.

[0048] In one possible implementation, the characteristic velocity threshold is dynamically set based on the laser power in the target laser processing parameters and the material thermophysical properties of the current processing point.

[0049] Feature velocity threshold It is not a fixed constant, but a variable that is dynamically set according to the processing conditions. Its setting is based on the laser power P and the specific thermophysical properties of the material. These properties include at least the material's thermal diffusivity α and evaporation temperature. The system uses relational expressions. An estimation is performed. Among them, ρ is the material's absorptivity to the laser wavelength, c is the material density, ΔT is the effective temperature rise, and w is the spot radius. The physical meaning of this formula is the maximum scanning speed per unit time that allows for effective processing given the laser power and material properties. Therefore, Directly related to the balance between heat input and heat dissipation, when power P increases or the material's thermal diffusivity α decreases, The corresponding increase.

[0050] In one possible implementation, the preset optimal radius of curvature is set based on the laser wavelength in the target laser processing parameters and the absorption rate of the material to be processed at that wavelength.

[0051] Preset optimal radius of curvature The setting is closely related to the interaction mechanism between lasers and materials. Its core basis lies in the laser wavelength λ and the bulk absorption coefficient of the material at that wavelength. The absorption coefficient is determined by the extinction coefficient k of the material, and the relationship is as follows: . This is determined using an optical model that considers the incident, scattering, and energy penetration depth of the beam on the curved surface. A specific functional relationship is as follows: ,in It is a dimensionless morphology coupling factor, obtained through calibration experiments. Its physical significance lies in the fact that for high-absorption materials, energy deposition is shallow, requiring a smaller [factor / component]. To focus energy; for low-absorption materials, energy penetrates deeper, allowing for greater energy penetration. To achieve a more uniform lateral energy distribution, the system adjusts the input wavelength accordingly. and absorption coefficient retrieved from materials database Dynamically calculate and set .

[0052] In one possible implementation, step S2, which involves planning the spatial scanning trajectory, further includes: S201. Slice and layer the three-dimensional geometric model data to generate multiple parallel cross-sectional contours; S202. For each cross-sectional profile, the path points of the laser processing head are calculated using the equal residual height algorithm; S203. Connect the path points of each layer according to the processing sequence, and smooth the motion commands of each axis to form a continuous five-axis linkage spatial scanning trajectory.

[0053] The trajectory planning refinement steps are implemented through the CAM module of the CNC system. Slicing and layering are performed along one principal axis of the workpiece coordinate system, intersecting the 3D model with a series of parallel planes to generate 2D cross-sectional profile loops representing different heights. For each cross-sectional profile, the implementation of the equal residual height algorithm first defines the maximum allowable residual height. The method iteratively calculates the spacing between adjacent scanning paths based on the geometry of the current contour, the effective diameter of the laser spot, and the overlap rate between rows, ensuring that the height of residual material between any two points in the scanning direction does not exceed [a certain value]. This generates a series of path points. Finally, the module connects these layered and segmented path points according to the machining sequence from bottom to top and from inside to outside, and uses a B-spline curve interpolation algorithm to smoothly fit the connected tool points. At the same time, it filters and smooths the motion angles of each rotation axis to generate a continuous, smooth spatial scan trajectory file that meets the requirements of five-axis linkage interpolation.

[0054] In one possible implementation, after step S7, a closed-loop adaptive control step is also included: S8. During the processing, monitor the actual laser power, actual feed rate, and surface morphology features fed back by auxiliary sensors in real time. S9. Compare the monitored actual parameters with the target laser processing parameters. If the deviation exceeds the preset tolerance, trigger the online self-tuning of the control parameters. S10. Online self-tuning includes: calculation logic for adjusting the focus following distance based on surface roughness deviation, or calculation logic for adjusting the beam incident optimization angle based on heat-affected zone morphology deviation.

[0055] Closed-loop adaptive control is achieved through an independent monitoring and feedback loop. In real-time monitoring, the actual laser power is sampled by a power meter along the beam path, the actual feed rate is calculated based on feedback from the encoders of each axis servo motor, and the surface morphology is measured online using a confocal microscope or white light interferometer separate from the main machining head to obtain the micro-roughness. And characteristic values ​​such as weld width W. In the online comparison step, the system will monitor... W and the target value of the process parameters Compare and calculate the absolute deviation When any deviation exceeds its preset tolerance, online self-tuning is triggered. The self-tuning action does not directly modify the final... Instead of outputting, it retrospectively adjusts the process adjustment factor. The mapping rules, or adjusting the proportional coefficient β and the value of the gain λ. Online self-tuning adopts a rule-based proportional-integral adjustment strategy. When the surface roughness consistently above the target value At that time, the system uses the formula Update process adjustment factor ,in These are the preset proportional and integral adjustment coefficients. Similarly, when the width W of the heat-affected zone deviates from the target value... At that time, the system uses the formula The coefficients β and λ are adjusted in the opposite direction to reduce the deviation. Proportional adjustment coefficient. The value is retrieved from the preset process knowledge base based on the material type during system initialization.

[0056] In one possible implementation, after step S1 and before step S2, the method further includes: S11. Based on the motion envelope of the laser processing head and focusing lens group, as well as the geometric model of the workpiece and fixture, the planned spatial scanning trajectory is simulated and tested for motion and interference in a virtual environment. S12. If potential interference is detected, the spatial scanning trajectory is automatically optimized until an interference-free safe processing trajectory is generated.

[0057] Trajectory safety prediction and optimization are completed during the offline programming stage or the preprocessing stage before machining. Motion simulation is based on a virtual environment containing a precise 3D model of the laser machining head, movable focusing lens, workpiece, and fixture. The system drives the virtual model to move cyclically according to the planned spatial scanning trajectory, strictly following the kinematic model of the five-axis machine tool. Interference verification is achieved by calculating the minimum spatial distance between all moving and stationary parts in the virtual environment, using a collision detection algorithm based on bounding box trees or triangular facet intersection detection. If the distance between any two points is less than the safety margin, it is considered a potential interference. Optimization strategies are automatically selected based on the type of interference: for minor regional interference, a trajectory translation strategy is used, i.e., fine-tuning the tool path in the normal plane; for interference caused by beam angle, a beam vector local reconstruction strategy is used, making minor adjustments to the incident angle within the allowable process window; for dynamic interference that may be caused by high-speed motion, a feed rate attenuation strategy is used, automatically reducing the feed rate in dangerous path segments. The system iteratively executes a simulation-detection-optimization loop until a safe machining trajectory verified as interference-free is generated. The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: (1) In this invention, the actual surface normal vector and local curvature of the processing point are collected online in real time, and the focus following distance is dynamically calculated based on this, thereby adjusting the position of the focusing lens group in real time. This technical means directly solves the problem of laser focus drift caused by the actual pose deviation of the workpiece and the change of surface curvature. It can ensure that the laser focus always accurately follows the real contour of the surface, and compensates for the error between the theoretical model and the actual object, as well as the focus offset caused by curvature. The technical effect achieved is that the actual size of the laser spot on the material surface remains highly consistent throughout the entire complex surface processing area, which significantly improves the geometric accuracy of the processing contour, such as achieving uniform kerf width or cladding layer thickness, and fundamentally overcomes the processing quality fluctuation caused by focus misalignment in traditional static focusing methods.

[0058] (2) In this invention, the beam incident angle is dynamically optimized by analyzing real-time surface information and material properties, and the spatial direction of the beam is adjusted in real time by combining a five-axis linkage system. This technique directly addresses the problem of uneven laser energy absorption efficiency caused by surface curvature changes and the Fresnel effect. The system adjusts the incident angle according to the local curvature and optimal energy coupling model, optimizes the interaction conditions between the beam and the surface, and makes the energy absorption rate tend to be stable. The resulting technical effect is that the distribution of laser energy on complex surfaces is actively controlled and optimized, avoiding defects such as overheating of convex surfaces and lack of fusion on concave surfaces, achieving precise control of heat input, thereby obtaining a high-quality processing area with uniform structure and consistent performance, and improving the overall mechanical properties of the weld or cladding layer.

[0059] (3) In this invention, a closed-loop adaptive control loop including real-time process monitoring and online parameter self-tuning is established to provide real-time feedback and correction for power, speed, and surface morphology deviations during the processing. This technical approach solves the problem of process instability caused by internal and external disturbances (such as power fluctuations, speed changes, and material inhomogeneity). The system can sense minute deviations in the processing effect and automatically backtrack to adjust the calculation logic of the core control parameters, enabling the entire processing system to have self-learning and self-adaptive capabilities. The resulting technical effect is that the robustness and stability of the laser processing of complex curved surfaces are greatly enhanced, enabling the continuous output of high-quality and highly consistent processing results under various uncertainties, reducing the dependence on operator experience, and improving the reliability and repeatability of the process.

[0060] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0061] The following points need to be explained: (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0062] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0063] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0064] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for adaptive laser focusing on complex curved surfaces based on five-axis linkage, characterized in that, include: S1. Obtain the three-dimensional geometric model data of the complex curved surface to be processed and the target laser processing parameters; S2. Based on the three-dimensional geometric model data, plan the spatial scanning trajectory of the laser processing head under the five-axis linkage CNC system. The spatial scanning trajectory includes the sequence of spatial position points of the processing head and the theoretical initial vector of the laser beam at each position point. S3. During the movement along the spatial scanning trajectory, the actual surface information at the current processing point is collected in real time by a non-contact ranging sensor installed on the laser processing head. The actual surface information includes at least the surface normal vector and the local curvature. S4. Based on the actual surface information and the target laser processing parameters, dynamically calculate and generate two key control quantities: the focus following distance, used to compensate for the deviation between the actual position of the laser focus and the theoretical surface position, and the beam incident optimization angle, used to optimize the energy distribution of laser energy on the surface; wherein, the calculation of the focus following distance takes into account at least the local curvature, and the calculation of the beam incident optimization angle takes into account at least the surface normal vector. S5. Based on the calculated focus following distance, adjust the axial position of the laser focusing lens group in real time so that the laser focus falls precisely on the target point determined by the focus following distance offset from the theoretical curved surface position along the normal vector; S6. Simultaneously, based on the calculated beam incident optimization angle and combined with the surface normal vector at the current processing point, the target space vector of the laser beam is calculated. S7. The target space vector and the spatial position of the target action point are converted into real-time control commands for each motion axis in the five-axis linkage CNC system to drive the laser processing head and the workpiece to move, thereby realizing adaptive focusing processing of the laser beam and complex curved surfaces.

2. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 1, characterized in that, The dynamic calculation of the focus following distance includes: S401. Based on the local curvature at the current processing point and the spot diameter requirement in the target laser processing parameters, calculate the basic offset required to ensure effective spot coverage. S402. Based on the material removal mode and laser power in the target laser processing process parameters, determine a process adjustment factor; S403. The basic offset is coupled with the process adjustment factor to obtain the final focus following distance; wherein the coupling operation is configured to make the focus following distance negatively correlated with the absolute value of the local curvature and positively correlated with the relative magnitude of the instantaneous feed rate relative to a characteristic speed threshold.

3. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 1, characterized in that, The dynamic calculation of the beam incidence optimization angle includes: S411. Using the surface normal vector at the current processing point as a reference, determine an initial beam vector according to the initial incident angle specified in the processing parameters. S412. Based on the local curvature and material properties, analyze the reflection and absorption characteristics of laser energy in the local region of the curved surface under the initial beam vector; S413. Based on the analysis results, the initial incident angle is dynamically optimized and adjusted to obtain the beam incident optimized angle; wherein, the dynamic optimization adjustment is based at least on the deviation between the local radius of curvature and a preset optimal radius of curvature, and the spatial rate of change of the surface normal vector.

4. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 2, characterized in that, The coupling operation in step S403 is specifically as follows: Calculate a baseline value consisting of the aforementioned base offset; A modulation factor is introduced, which is a nonlinear function of the difference between the instantaneous feed rate and the characteristic rate threshold; The baseline value is dynamically scaled using the modulation factor to obtain the focus following distance; wherein, when the instantaneous feed rate is greater than the characteristic speed threshold, the modulation factor drives the growth rate of the focus following distance to increase accordingly.

5. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 3, characterized in that, The dynamic optimization adjustment in step S413 specifically includes: Calculate the first correction amount required to determine the initial incident angle based on the deviation between the local radius of curvature and the preset optimal radius of curvature; Calculate the second correction amount required for the spatial rate of change based on the surface normal vector; The first correction amount and the second correction amount are combined to obtain the final adjustment amount of the initial incident angle, thereby determining the optimized incident angle of the beam.

6. A method for adaptive laser focusing on complex curved surfaces based on five-axis linkage according to claim 2 or 4, characterized in that, The characteristic velocity threshold is dynamically set based on the laser power in the target laser processing parameters and the material thermophysical properties at the current processing point.

7. A method for adaptive laser focusing on complex curved surfaces based on five-axis linkage according to claim 3 or 5, characterized in that, The preset optimal radius of curvature is set based on the laser wavelength in the target laser processing parameters and the absorption rate of the material to be processed at that wavelength.

8. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 1, characterized in that, The step S2 of planning the spatial scanning trajectory also includes: S201. The three-dimensional geometric model data is sliced ​​and layered to generate multiple parallel cross-sectional contours; S202. For each cross-sectional profile, the path points of the laser processing head are calculated using the equal residual height algorithm; S203. Connect the path points of each layer according to the processing sequence, and smooth the motion commands of each axis to form a continuous five-axis linkage spatial scanning trajectory.

9. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 8, characterized in that, Following step S7, a closed-loop adaptive control step is also included: S8. During the processing, monitor the actual laser power, actual feed rate, and surface morphology features fed back by auxiliary sensors in real time. S9. Compare the monitored actual parameters with the target laser processing parameters. If the deviation exceeds the preset tolerance, trigger the online self-tuning of the control parameters. S10. The online self-tuning includes: calculation logic for adjusting the focus following distance based on surface roughness deviation, or calculation logic for adjusting the beam incident optimization angle based on heat-affected zone morphology deviation.

10. The adaptive laser focusing method for complex curved surfaces based on five-axis linkage according to claim 1, characterized in that, After step S1 and before step S2, it also includes: S11. Based on the motion envelope of the laser processing head and focusing lens group, as well as the geometric model of the workpiece and fixture, the planned spatial scanning trajectory is simulated and tested for motion and interference in a virtual environment. S12. If potential interference is detected, the spatial scanning trajectory is automatically optimized until an interference-free safe processing trajectory is generated.

Images