A method and system for processing a titanium alloy material curved surface

Abstract

This invention discloses a method and system for machining curved surfaces of titanium alloy materials, comprising the following steps: acquiring initial blank parameters, surface design model data, and preset machining trajectory planning data corresponding to the titanium alloy material through a five-axis linkage machining unit; collecting initial contact state data using an ultrasonic vibration auxiliary unit and combining it with reference contour scanning data generated by a real-time contour monitoring unit to construct an initial parameter set for titanium alloy surface machining; generating a sequence of collaborative control parameters for the machining process based on the initial parameter set; performing dynamic compensation calculations on the sequence of collaborative control parameters to generate adaptive machining parameters; driving the five-axis linkage machining unit, ultrasonic vibration auxiliary unit, and cooling and lubrication unit to operate collaboratively according to the adaptive machining parameters, generating a precision-machined surface of the titanium alloy material and performing accuracy testing, and generating a surface machining quality assessment report. This invention enables efficient and precise machining of complex curved surfaces of titanium alloys.

Description

Technical Field

[0001] This invention relates to the field of material surface processing technology, specifically to a method and system for processing curved surfaces of titanium alloy materials. Background Technology

[0002] Titanium alloys, with their high strength, low density, corrosion resistance and excellent high-temperature mechanical properties, have become key materials for manufacturing complex curved surface core components (such as industrial robot parts and mechanical equipment parts). However, the requirements for the surface contour accuracy, surface roughness and mechanical properties after processing of these components are becoming increasingly stringent, and traditional processing technologies can hardly meet the processing needs of high precision and high efficiency.

[0003] Existing methods form curved surfaces by combining preset tool paths and machining parameters, and then verify the machining accuracy according to fixed inspection standards. However, due to factors such as the low thermal conductivity of titanium alloys, significant work hardening effect, and easy tool wear, thermal deformation, surface micro-cracks, or dimensional deviations may occur during the machining process, resulting in a decline in component machining quality. At the same time, traditional machining processes can only adjust parameters for a single machining stage, which is difficult to cover the accuracy control requirements of the entire process from roughing to finishing of complex curved surfaces, and the machining efficiency is low. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for machining curved surfaces of titanium alloy materials, comprising the following steps: Step S1: Obtain the initial blank parameters, surface design model data and preset machining trajectory planning data corresponding to the titanium alloy material through the five-axis linkage machining unit, collect the initial contact state data between the tool and the titanium alloy material using the ultrasonic vibration auxiliary unit, and combine the reference contour scanning data generated by the real-time contour monitoring unit to construct the initial parameter set for titanium alloy surface machining. Step S2: Based on the initial parameter set for machining titanium alloy curved surfaces, real-time dynamic adjustment parameters for spindle speed and feed rate are generated through a five-axis linkage machining unit. At the same time, high-frequency vibration parameters are output using an ultrasonic vibration auxiliary unit. The real-time dynamic adjustment parameters for spindle speed and feed rate and the high-frequency vibration parameters are then fused to generate a sequence of collaborative control parameters for the machining process. Step S3: Continuously collect real-time contour data of the surface machining area through the real-time contour monitoring unit, and combine it with the temperature field distribution data of the machining area fed back by the cooling and lubrication unit to calculate the surface machining contour deviation value and thermal deformation influence coefficient; Based on the surface machining contour deviation value and thermal deformation influence coefficient, perform dynamic compensation calculation on the machining process collaborative control parameter sequence to generate adaptive machining parameters including spindle speed correction, feed rate compensation value and ultrasonic vibration frequency adjustment. Step S4: Drive the five-axis linkage machining unit, ultrasonic vibration auxiliary unit and cooling and lubrication unit to operate in coordination according to the adaptive machining parameters to generate a precision machined surface of titanium alloy material. Then, use the real-time contour monitoring unit to detect the accuracy of the precision machined surface of titanium alloy material and generate a surface machining quality assessment report.

[0005] Furthermore, the present invention also provides a titanium alloy material curved surface machining system, including a computer-readable storage medium, a processor, a communication interface, and a computer program stored on the computer-readable storage medium and executable on the processor. When the processor executes the computer program, it implements the titanium alloy material curved surface machining method as described above.

[0006] The beneficial effects of this application are as follows: by accurately acquiring core information such as initial blank parameters and surface design model data through a five-axis linkage machining unit, and by capturing the initial contact state between the tool and the titanium alloy with the help of an ultrasonic vibration auxiliary unit, and then combining the reference scanning data of the real-time contour monitoring unit to construct an initial parameter set, the initial machining conditions can be fully grasped. This multi-dimensional data fusion parameter construction method can effectively avoid machining deviations caused by missing or inaccurate initial parameters, laying the foundation for precise control of the subsequent machining process. Especially for the characteristics of titanium alloys, such as low thermal conductivity and significant work hardening effect, mastering the initial contact state in advance can avoid excessively rapid initial tool wear and reduce the probability of subsequent thermal deformation and surface microcracks. Secondly, based on the initial parameter set, dynamic adjustment parameters for spindle speed and feed rate are generated through a five-axis linkage machining unit. At the same time, high-frequency vibration parameters are output using an ultrasonic vibration auxiliary unit. These three parameters are integrated to generate a coordinated control parameter sequence, realizing the linkage control of the core machining unit. The dynamically adjusted spindle speed and feed rate can be adjusted in real time according to the actual situation of the machining area, avoiding local machining overload caused by fixed parameters and reducing thermal deformation. The high-frequency vibration parameters can reduce the friction coefficient between the tool and the titanium alloy material by utilizing the characteristics of ultrasonic vibration, alleviating the work hardening effect, reducing tool wear, and extending tool life. This multi-parameter coordinated control mode breaks the limitations of traditional single-parameter adjustment, improves the stability and efficiency of the machining process, and can better adapt to the entire process requirements of complex titanium alloy surfaces from roughing to finishing. Then, the real-time contour data of the machining area is continuously collected by the real-time contour monitoring unit. Combined with the temperature field distribution data fed back by the cooling and lubrication unit, the contour deviation value and thermal deformation influence coefficient are accurately calculated. This allows for dynamic compensation of the collaborative control parameter sequence, generating adaptive machining parameters. This real-time monitoring and dynamic compensation mechanism can promptly identify the root cause of deviations during machining, adjust the cooling and lubrication strategy based on the temperature field distribution to reduce the impact of thermal deformation, and correct the spindle speed, feed rate, and ultrasonic vibration frequency in real time according to the contour deviation, effectively curbing further expansion of the deviation. For example, when excessively high local temperatures are detected, heat accumulation can be reduced by adjusting the feed rate or enhancing cooling and lubrication. When contour deviations exceed limits, the spindle speed and ultrasonic vibration frequency are corrected promptly to ensure contour accuracy, avoiding the lag problem of "post-processing verification" in traditional machining, and significantly reducing the risks of dimensional deviations, thermal deformation, and surface microcracks in titanium alloy curved surface machining.Finally, the five-axis linkage, ultrasonic vibration assistance, and cooling and lubrication units are driven by adaptive machining parameters to ensure that each unit maintains precise coordination during the machining process. This avoids machining errors caused by asynchronous unit operation. After generating a precision-machined surface, the accuracy is detected by a real-time contour monitoring unit, and a quality assessment report is generated. The collaborative operation mechanism ensures the continuity and stability of the machining process, further improving the surface machining accuracy. The comprehensive quality assessment report not only intuitively presents whether the machining accuracy meets the standards, but also provides feedback on the adaptability of parameters in each machining stage, providing a basis for optimizing subsequent machining parameters. At the same time, continuous optimization of machining parameters improves the overall machining quality and efficiency, meeting the high-quality requirements of titanium alloy precision components. Attached Figure Description

[0007] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of the titanium alloy material curved surface processing method in this embodiment; Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1. Figure 3 for Figure 1 A schematic diagram illustrating the overall technical process of step S2; Figure 4 This is a schematic diagram of the titanium alloy material curved surface processing system in this embodiment. Detailed Implementation

[0008] The following drawings disclose several embodiments of the present invention. For clarity, many practical details will be described in the following description. However, it should be understood that these practical details are not intended to limit the invention. That is, in some embodiments of the invention, these practical details are not essential. Furthermore, for the sake of simplicity, some conventional structures and components will be shown in the drawings in a simple schematic manner.

[0009] To further understand the invention's content, features, and effects, the following embodiments are provided, and detailed descriptions are given below in conjunction with the accompanying drawings: Reference Figure 1 , Figure 1 This is a flowchart of the titanium alloy material curved surface processing method in this embodiment. The titanium alloy material curved surface processing method in this embodiment includes the following steps: Step S1: Obtain the initial blank parameters, surface design model data and preset machining trajectory planning data corresponding to the titanium alloy material through the five-axis linkage machining unit, collect the initial contact state data between the tool and the titanium alloy material using the ultrasonic vibration auxiliary unit, and combine the reference contour scanning data generated by the real-time contour monitoring unit to construct the initial parameter set for titanium alloy surface machining. In this embodiment of the invention, the initial parameters of the titanium alloy billet (length 200mm × width 150mm × height 80mm, density 4.51g / cm³) are obtained by the laser measurement component (laser wavelength 532nm, scanning interval 0.05mm) of the five-axis linkage machining unit. 3 Ra 1.6μm; read the surface design model data (hemispherical radius 50mm + cylinder diameter 80mm, shape tolerance 0.03mm), and plan the preset machining trajectory (residual height 0.01mm, node coordinate accuracy 0.001mm). The force sensor of the ultrasonic vibration auxiliary unit (frequency 20kHz, amplitude 15μm) collects the initial contact state data (radial force 250N, axial force 180N, contact coordinates X100mm Y75mm Z80mm). The laser scanning probe of the real-time contour monitoring unit (scanning speed 50mm / s, point cloud density 100 points / mm) 2 Generate baseline contour scanning data (1.5 million points, coordinate accuracy 0.005mm). Integrate five types of data, unify the format (coordinates mm, force N, angle °), and construct an initial parameter set for titanium alloy surface machining, containing 128 specific parameters.

[0010] Step S2: Based on the initial parameter set for machining titanium alloy curved surfaces, real-time dynamic adjustment parameters for spindle speed and feed rate are generated through a five-axis linkage machining unit. At the same time, high-frequency vibration parameters are output using an ultrasonic vibration auxiliary unit. The real-time dynamic adjustment parameters for spindle speed and feed rate and the high-frequency vibration parameters are then fused to generate a sequence of collaborative control parameters for the machining process. In this embodiment of the invention, the curvature of the surface (0.02mm for a hemispherical surface) is analyzed by a five-axis linkage machining unit based on an initial parameter set. -1The spindle speed (cylindrical surface 0), contour complexity (transition region 0.3358), and feed rates (0.034 mm / r, 0.025 mm / r, 0.061 mm / r) were calculated in real time, with the goal of minimizing tool wear in low-complexity regions and ensuring stable machining in high-complexity regions. The calculations were based on these parameters. An ultrasonic vibration-assisted unit, combined with the peak cutting force (280 N) and material hardness (HRC35), calculated high-frequency vibration parameters (cylindrical surface amplitude 0.025 mm / frequency 135 Hz, transition region 0.0188 mm / frequency 136.69 Hz). Time-series synchronous analysis (sampling interval 0.1 s, 300 data sets) was performed on the spindle speed, feed rate, and vibration parameters. These parameters were then aligned and fused according to timestamps to generate a sequence of collaborative control parameters for the machining process, ensuring real-time coordination among the three.

[0011] Step S3: Continuously collect real-time contour data of the surface machining area through the real-time contour monitoring unit, and combine it with the temperature field distribution data of the machining area fed back by the cooling and lubrication unit to calculate the surface machining contour deviation value and thermal deformation influence coefficient; Based on the surface machining contour deviation value and thermal deformation influence coefficient, perform dynamic compensation calculation on the machining process collaborative control parameter sequence to generate adaptive machining parameters including spindle speed correction, feed rate compensation value and ultrasonic vibration frequency adjustment. In this embodiment of the invention, real-time contour data is collected every 0.5 seconds by a real-time contour monitoring unit. The deviation is calculated by registering the data with the CAD model (maximum 0.018mm, average 0.011mm in the transition area). The surface machining contour deviation value of 0.0555 is obtained using the formula (maximum deviation × 0.4 + average deviation × 0.3 + deviation area percentage × 0.3). The temperature sensor array (20 sensors, accuracy ±0.5℃) of the cooling and lubrication unit generates a temperature field distribution (maximum 78℃ in the transition area), combined with the titanium alloy's thermal expansion coefficient of 10.8 × 10⁻⁶. -6 / ℃, calculate the thermal deformation influence coefficient (0.00907 in the high-temperature zone). Both indicators exceed the threshold (deviation 0.05, coefficient 0.008), triggering the correction mechanism: decompose the deviation into normal 0.048mm and tangential 0.0278mm, calculate the contour deviation compensation; adjust the cooling flow rate to 12L / min, and convert the temperature compensation coefficient to 0.00181. Dynamically compensate and coordinate control parameters to generate adaptive machining parameters (spindle speed 509.95r / min, feed 0.06095mm / r, vibration frequency 136.75Hz).

[0012] Step S4: Drive the five-axis linkage machining unit, ultrasonic vibration auxiliary unit and cooling and lubrication unit to operate in coordination according to the adaptive machining parameters to generate a precision machined surface of titanium alloy material. Then, use the real-time contour monitoring unit to detect the accuracy of the precision machined surface of titanium alloy material and generate a surface machining quality assessment report.

[0013] In this embodiment of the invention, the five-axis linkage machining unit is driven to run along a compensated trajectory (feed depth -0.048mm, trajectory offset +0.0278mm) according to adaptive machining parameters. The ultrasonic vibration auxiliary unit vibrates at 136.75Hz / 0.0188mm, and the cooling and lubrication unit sprays at a flow rate of 12L / min. After machining, the real-time contour monitoring unit scans again to detect the accuracy (shape tolerance 0.025mm, surface roughness Ra 0.7μm), and the cooling unit records the final temperature (70℃ in the transition area). Integrating the detection data, a surface machining quality assessment report is generated: it is clear that the machined surface meets the design requirements (accuracy compliance rate 98%), the machining accuracy in the transition area is marked as optimal (deviation 0.009mm), and it is suggested that the vibration frequency can be finely adjusted to 137Hz for the next machining to further reduce surface roughness, providing a reference for subsequent machining.

[0014] Furthermore, refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1 in this embodiment. Step S1 in this embodiment includes the following steps: Step S11: Obtain the three-dimensional dimensional parameters, material uniformity distribution data, and surface roughness parameters of the initial titanium alloy blank through the laser measurement component corresponding to the five-axis linkage machining unit, so as to generate the initial blank parameters; In this embodiment of the invention, the initial titanium alloy billet is three-dimensionally scanned using the laser measurement component of a five-axis linkage machining unit. The laser wavelength is set to 532nm, the scanning interval is 0.05mm, and the three-dimensional dimensional parameters (length 200mm, width 150mm, height 80mm, dimensional error ±0.02mm) are obtained. The material density is then measured using a material density detection module (detection accuracy 0.01g / cm³). 3 Obtain material uniformity distribution data (density 4.51 g / cm³). 3 Distribution deviation ≤ 0.03 g / cm 3 The surface roughness was detected using a stylus-type measuring instrument (measuring force 0.75mN), with a sampling length of 0.8mm and an evaluation length of 4mm, to obtain roughness parameters (Ra 1.6μm, Rz 6.3μm). These parameters were then integrated to generate initial blank parameters, including size, material distribution, and surface quality data.

[0015] Step S12: Obtain the CAD model of the surface design corresponding to the initial billet of titanium alloy material based on the initial billet parameters, and extract the corresponding surface curvature change parameters, feature contour line data and accuracy tolerance requirements parameters according to the CAD model of the surface design to generate surface design model data; In this embodiment of the invention, a surface design CAD model (a combination of a hemispherical and cylindrical surface, with a hemispherical radius of 50mm and a cylindrical diameter of 80mm) is called based on the initial blank parameters. The surface curvature variation parameters (hemispherical curvature 0.02mm) are extracted using a model analysis tool. -1 The cylindrical surface has a curvature of 0, and the curvature gradually changes from 0.005 to 0.02 mm in the transition region. -1 Extract feature contour data (hemispherical equatorial circle diameter 100mm, cylindrical axis offset from hemispherical center 10mm); read the model annotation accuracy tolerance requirements parameters (contour accuracy 0.03mm, position accuracy 0.05mm, surface roughness Ra 0.8μm), generate surface design model data, and clarify the processing target shape and accuracy standards.

[0016] Step S13: Plan the initial machining trajectory based on the surface design model data, and obtain the trajectory node coordinate data, tool attitude angle parameters and path interval distribution data based on the initial machining trajectory to generate preset machining trajectory planning data; In this embodiment of the invention, the initial machining trajectory is planned using an equal residual height algorithm based on the surface design model data. The residual height is set to 0.01mm. The coordinate data of the trajectory nodes are calculated (the hemispherical surface is divided into polar coordinates, with one node taken every 5°, for a total of 72 nodes, with a coordinate accuracy of 0.001mm). The tool attitude angle parameters are determined according to the surface normal direction (the angle between the tool axis and the normal is 5° when machining the hemispherical surface, and 0° when machining the cylindrical surface, with an angle accuracy of 0.01°). The path interval distribution data is calculated (the path interval is 1mm in the hemispherical area, 2mm in the cylindrical area, and 1-2mm in the transition area), generating preset machining trajectory planning data, which includes complete tool path parameters.

[0017] Step S14: Use the force sensor corresponding to the ultrasonic vibration auxiliary unit to collect the initial cutting force data and contact position coordinate parameters at the moment the tool comes into contact with the titanium alloy material, and generate initial contact state data; In this embodiment of the invention, the ultrasonic vibration auxiliary unit is activated with a vibration frequency of 20kHz and an amplitude of 15μm. The force sensor sampling frequency is 10kHz. At the instant the cutting tool (10mm diameter carbide ball end mill) contacts the surface of the titanium alloy billet, the initial cutting force data (radial force 250N, axial force 180N, tangential force 120N, force error ±5N) is collected. Simultaneously, the contact position coordinate parameters (X100mm Y75mm Z80mm, coordinate error ±0.01mm) are recorded to generate initial contact state data, reflecting the mechanical and positional characteristics of the tool and material at the instant of contact.

[0018] Step S15: The initial titanium alloy blank is scanned across the entire surface area by the laser scanning probe corresponding to the real-time contour monitoring unit to obtain the reference contour point cloud data. Based on the reference contour point cloud data and the preset processing trajectory planning data, coordinate matching is performed to generate the reference contour scanning data. In this embodiment of the invention, the laser scanning probe of the real-time contour monitoring unit (scanning line speed 50 mm / s, point cloud density 100 points / mm) is used. 2 A full-area surface scan of the initial titanium alloy billet is performed to obtain reference contour point cloud data (containing 1.5 million points, each with a coordinate accuracy of 0.005 mm). The reference contour point cloud data is matched with the preset machining trajectory planning data using an iterative nearest point algorithm. The matching error is controlled within 0.01 mm to generate reference contour scan data and establish the positional relationship between the initial machining state and the theoretical trajectory.

[0019] Step S16: The initial blank parameters, surface design model data, preset machining trajectory planning data, initial contact state data and reference contour scanning data are fused to construct the initial parameter set for titanium alloy surface machining.

[0020] In this embodiment of the invention, the initial blank parameters (dimensions, material roughness), surface design model data (curvature profile tolerance), preset machining trajectory planning data (node ​​coordinates, tool posture path interval), initial contact state data (cutting force contact coordinates), and reference profile scanning data (point cloud matching results) are fused together. A weighted average method is used to process duplicate parameters (such as taking the average value of laser measurement and scanning probe for coordinate data), and a unified data format (coordinate unit mm, force unit N, angle unit °) is established to construct an initial parameter set for titanium alloy surface machining containing 128 parameters, providing a complete initial data foundation for subsequent machining processes.

[0021] Furthermore, refer to Figure 3 , Figure 3 for Figure 1 A schematic diagram illustrating the overall technical process of step S2 in this embodiment shows that step S2 includes the following steps: Step S21: Extract the corresponding surface curvature change parameters and preset machining trajectory planning data from the initial parameter set of titanium alloy surface machining, and calculate the curvature radius change rate and contour complexity coefficient corresponding to different machining areas based on the surface curvature change parameters and preset machining trajectory planning data combined with the reference contour scanning data. In this embodiment of the invention, the curvature variation parameter of the surface (hemispherical curvature 0.02mm) is extracted from the initial parameter set of the titanium alloy curved surface machining. -1 Cylindrical surface with 0% curvature; transition area 0.005-0.02mm. -1 Combined with preset machining trajectory planning data (node ​​coordinates, path interval 1-2mm), and reference contour scanning data (point cloud coordinate accuracy 0.005mm), the rate of change of curvature radius is calculated using the formula: Curvature radius change rate = (curvature difference between adjacent nodes) / path interval. The rate of change for the cylindrical surface is 0mm / mm. 2 , hemispherical surface 0.03mm / mm 2 Transition area 0.3mm / mm 2 Based on the rate of curvature change, trajectory density (10 tracks / 10mm for hemispherical surfaces, 5 tracks / 10mm for cylindrical surfaces), and contour deviation dispersion (0.286 for the transition region), the contour complexity coefficients are calculated by weighting: 0.0384 for cylindrical surfaces, 0.0798 for hemispherical surfaces, and 0.3358 for the transition region, thus clarifying the curvature and complexity characteristics of different processing areas.

[0022] Step S22: Based on the rate of change of curvature radius and contour complexity coefficient corresponding to different machining areas, and combined with the surface design model data, the five-axis linkage machining unit is used to generate the corresponding real-time spindle speed and feed rate dynamic adjustment parameters. In this embodiment of the invention, the curvature radius change rate of each region (0.3 mm / mm in the transition region) is used as the basis for the method. 2 The model is calculated using the highest complexity coefficient (0.3358 for the transition region, representing medium complexity) and the surface design model's accuracy tolerance (0.008mm for the transition region's shape tolerance, the strictest). The five-axis machining unit then calls the parameter calculation model. With the goal of minimizing tool wear in the low-complexity region (cylindrical surface), the spindle speed is calculated to be 334 r / min and the feed rate to be 0.034 mm / r. In the medium-complexity region (transition region), with the goal of machining stability, the spindle speed is calculated to be 510 r / min and the feed rate to be 0.061 mm / r, combined with curvature correction. For the hemispherical surface (low complexity), the spindle speed is calculated to be 371 r / min and the feed rate to be 0.025 mm / r. Real-time dynamic adjustment parameters for spindle speed and feed rate are generated to ensure that the parameters match the machining requirements of the region.

[0023] Step S23: Using the ultrasonic vibration auxiliary unit, the initial amplitude and frequency parameters of the ultrasonic vibration are calculated based on the initial contact state data and the hardness parameters of the titanium alloy material and the characteristic parameters of the tool material, so as to generate high-frequency vibration parameters. In this embodiment of the invention, the initial contact state data (peak cutting force 280N / 200N / 140N, contact coordinates) is read by an ultrasonic vibration auxiliary unit, combined with the titanium alloy hardness HRC35 and tool characteristics (carbide HRC65, radius 5mm). The initial amplitude of ultrasonic vibration is calculated according to the formula: amplitude = (peak cutting force × tool hardness) / (material hardness × 1000), with 0.025mm for the cylindrical surface, 0.0125mm for the hemispherical surface, and 0.0188mm for the transition region. Combining the cutting force fluctuation frequency (100-150Hz) with the spindle speed matching, the initial frequencies are determined as follows: 135Hz for the cylindrical surface, 137.7Hz for the hemispherical surface, and 136.69Hz for the transition region. High-frequency vibration parameters are integrated to ensure that the vibration is compatible with the cutting state.

[0024] Step S24: Perform time series synchronous analysis on the real-time spindle speed, feed rate dynamic adjustment parameters and high-frequency vibration parameters, output the real-time spindle speed change curve, feed rate dynamic change sequence and high-frequency vibration parameter optimization sequence, and fuse them to generate a machining process collaborative control parameter sequence.

[0025] In this embodiment of the invention, time-series synchronous analysis is performed on real-time spindle speed (334-510 r / min), feed rate (0.025-0.061 mm / r), and high-frequency vibration parameters (frequency 135-137.7 Hz, amplitude 0.0125-0.025 mm), with a sampling interval of 0.1 s, generating 300 sets of synchronous data. The output includes a real-time spindle speed variation curve (smoothly transitioning from 334 r / min on the cylindrical surface to a transition region of 510 r / min), a dynamic feed rate variation sequence (0.034→0.025→0.061 mm / r), and an optimized high-frequency vibration parameter sequence (amplitude 0.025→0.0125→0.0188 mm, frequency 135→137.7→136.69 Hz). These three sequences are aligned and merged according to timestamps to generate a sequence of collaborative control parameters for the machining process, ensuring real-time coordination among the spindle, feed, and vibration to meet the machining requirements of each region.

[0026] Furthermore, the step S21, which calculates the rate of change of curvature radius and contour complexity coefficient corresponding to different processing areas based on surface curvature change parameters, preset processing trajectory planning data, and reference contour scanning data, includes the following steps: By extracting principal curvature values, curvature direction angles, and Gaussian curvature values ​​corresponding to each sampling point from the surface curvature variation parameters, and constructing a surface curvature feature matrix; and reading the corresponding trajectory node coordinate data and path interval distribution data from the preset processing trajectory planning data, a three-dimensional coordinate sequence of the processing trajectory and a trajectory density distribution map are generated. In this embodiment of the invention, sampling points (300 in total) are extracted from the surface curvature variation parameters of the surface design model data at 1mm intervals. The principal curvature value is calculated for each sampling point (the first principal curvature of the hemispherical region is 0.02mm). -1 Second principal curvature 0.02mm -1 The first principal curvature of the cylindrical surface region is 0 mm. -1 The second principal curvature is 0.0125mm. -1 The first principal curvature of the transition region is 0.005-0.02mm. -1 The second principal curvature is 0.006-0.02mm. -1 ), curvature direction angle (uniformly distributed from 0-360° in the hemispherical region, and fixed at 90° in the cylindrical region), Gaussian curvature value (0.0004mm in the hemispherical region). -2 Cylindrical area 0mm -2 The transition zone is 0.00003-0.0004mm. -2 The three types of parameters from 300 sampling points are arranged in the format of "sampling point number-principal curvature 1-principal curvature 2-direction angle-Gaussian curvature" to construct a 300×5 dimension surface curvature feature matrix, ensuring that the curvature features of each sampling point can be accurately located. Next, the trajectory node coordinate data (72 nodes for hemispherical surfaces, coordinates such as (100,75,130), (104.33,75,127.5)...; 48 nodes for cylindrical surfaces, coordinates such as (100,75,80), (100,75,82)...) are read from the preset machining trajectory planning data and arranged according to the machining order to generate a three-dimensional coordinate sequence of the machining trajectory (a total of 120 coordinate points, coordinate accuracy 0.001mm). Read the path interval distribution data (1mm for hemispherical region, 2mm for cylindrical region), count the number of trajectory nodes within every 10mm length (10 nodes / 10mm for hemispherical region, 5 nodes / 10mm for cylindrical region), and present the data in the form of a heat map. Dense areas of trajectory nodes are marked in red (hemispherical region), and sparse areas are marked in blue (cylindrical region), generating a trajectory density distribution map to intuitively reflect the trajectory distribution density of different curved surface regions.

[0027] Furthermore, the reference contour point cloud data in the reference contour scanning data is spatially registered with the three-dimensional coordinate sequence of the processing trajectory to calculate the spatial distance deviation between the trajectory nodes and the reference contour points, and generate trajectory-contour matching deviation parameters; based on the principal curvature values ​​in the surface curvature feature matrix and the trajectory-contour matching deviation parameters, a differential geometry algorithm is used to calculate the initial value of the curvature radius corresponding to each trajectory node. In this embodiment of the invention, spatial registration is performed between the reference contour point cloud data (1.5 million points, coordinate accuracy 0.005 mm) in the reference contour scanning data and the three-dimensional coordinate sequence of the machining trajectory (120 points). An iterative nearest-point algorithm is used: for each trajectory node, the three nearest reference contour points are searched in the point cloud data, and the average distance is calculated as the spatial distance deviation. For example, the hemispherical node (100, 75, 130) corresponds to an average distance of 0.008 mm from the reference point cloud, the cylindrical node (100, 75, 80) corresponds to an average distance of 0.005 mm, and the transition region node corresponds to an average distance of 0.006-0.008 mm. The deviation values ​​of the 120 trajectory nodes are recorded as "node number - spatial distance deviation" to generate a trajectory-contour matching deviation parameter. The smaller the deviation value, the higher the fit between the trajectory and the actual blank contour. Secondly, based on the principal curvature values ​​(taking the first principal curvature) in the surface curvature feature matrix and the trajectory-contour matching deviation parameter, a differential geometry algorithm is used to calculate the initial value of the curvature radius of each trajectory node. The algorithm formula is: Curvature radius = 1 / Principal curvature value × (1 - Spatial distance deviation / Preset accuracy tolerance), with a preset accuracy tolerance of 0.03mm. For example, the principal curvature of a hemispherical node is 0.02mm. -1 The deviation is 0.008mm, and the radius of curvature is 1 / 0.02×(1-0.008 / 0.03)=50×0.733≈36.65mm; the principal curvature of the cylindrical surface node is 0.0125mm. -1 With a deviation of 0.005 mm, the radius of curvature is calculated as 1 / 0.0125 × (1 - 0.005 / 0.03) = 80 × 0.833 ≈ 66.64 mm. The initial values ​​of the radius of curvature for 120 trajectory nodes are calculated, and a sequence of radius of curvature is generated to ensure that the values ​​match the actual curvature and contour deviation of the surface.

[0028] Furthermore, the initial values ​​of the curvature radius of adjacent trajectory nodes are differentially calculated, and the trajectory step size parameter in the path interval distribution data is combined to generate the curvature radius change gradient parameter; the curvature change direction vector field is constructed based on the curvature radius change gradient parameter and the curvature direction angle, and the curvature radius change rate corresponding to different processing areas is analyzed based on the curvature change direction vector field; In this embodiment of the invention, the initial values ​​of the curvature radii of adjacent trajectory nodes are differentially calculated using the formula: Difference = Curvature radius of the next node - Curvature radius of the previous node. For example, the radii of adjacent nodes in a hemispherical region are 36.65 mm and 36.70 mm, with a difference of 0.05 mm; the radii of adjacent nodes in a cylindrical region are 66.64 mm and 66.64 mm, with a difference of 0 mm; and the radii of adjacent nodes in the transition region change from 45 mm to 50 mm, with a difference of 0.5-1 mm. The trajectory step length parameters (1 mm for the hemispherical region and 2 mm for the cylindrical region) are read from the path interval distribution data, and the curvature radius change gradient is calculated using the formula: Curvature radius change gradient = Difference / Trajectory step length. The gradient for the hemispherical region is 0.05 mm / mm, the gradient for the cylindrical region is 0 mm / mm, and the gradient for the transition region is 0.25-0.5 mm / mm. The gradient values ​​of 119 adjacent nodes are arranged in the processing order to generate the curvature radius change gradient parameters, reflecting the rate of change of the curvature radius as the trajectory progresses. Secondly, a curvature change direction vector field is constructed based on the curvature radius change gradient parameters (0.05 mm / mm for hemispherical surfaces, 0 mm / mm for cylindrical surfaces, and 0.25-0.5 mm / mm for the transition region) and curvature direction angles (0-360° for hemispherical surfaces, 90° for cylindrical surfaces). The vector of each trajectory node is determined by the gradient value and direction angle: the vector magnitude of hemispherical nodes is 0.05, and the direction changes with the node position from 0 to 360°, forming a ring vector field; the vector magnitude of cylindrical nodes is 0, and the direction is fixed at 90°, forming a zero vector field; the vector magnitude of transition region nodes is 0.25-0.5, and the direction gradually changes from 90° to 360°, forming a gradual vector field. The rate of change of curvature radius is calculated based on the vector field, using the formula: rate of change = vector magnitude × cosine of direction angle. The rate of change for hemispherical surfaces is 0.05 × cosθ (θ is the node direction angle), ranging from -0.05 to 0.05 mm / mm. 2 The rate of change for the cylindrical surface is 0; the rate of change for the transition region is 0.25-0.5×cosθ, ranging from -0.5 to 0.5 mm / mm. 2 This clarifies the rate of curvature change in different regions.

[0029] Furthermore, Gaussian curvature values ​​are extracted from the surface curvature feature matrix, and the convex-concave transition region of the surface is identified by analyzing the sign change corresponding to the Gaussian curvature values, generating region boundary feature parameters; the processing region is segmented based on the region boundary feature parameters, and the standard deviation of the trajectory-contour matching deviation parameter in each sub-region is calculated to generate the contour deviation dispersion coefficient. In this embodiment of the invention, the Gaussian curvature values ​​(0.0004 mm for a hemispherical surface) of 300 sampling points are extracted from the surface curvature feature matrix. -2 Cylindrical surface 0mm -2 Transition zone 0.00003-0.0004mm -2Analysis of sign changes: The Gaussian curvature values ​​in all regions are non-negative, with no abrupt change in sign, but the Gaussian curvature in the transition region changes from 0.00003 mm. -2 Gradient to 0.0004mm -2 The abrupt change point (Gaussian curvature 0.0002mm) -2 The boundary of the convex-concave transition region is defined at the point (e.g., (100, 75, 105), (102, 75, 104)...). The coordinates of 10 sampling points at the boundary are extracted, and the length of the boundary line segment (5mm) and the angle between the boundary and the axis (30°) are calculated to generate the region boundary feature parameters, clarifying the spatial location and geometric shape of the transition region. Next, based on the region boundary feature parameters, the processing area is divided into three sub-regions: a hemispherical sub-region (containing 72 trajectory nodes), a transition sub-region (containing 20 trajectory nodes), and a cylindrical sub-region (containing 48 trajectory nodes). The standard deviation of the trajectory-contour matching deviation parameter within each sub-region is calculated using the formula: Standard Deviation = √[Σ(deviation value - mean deviation)]. 2 / (number of nodes - 1)]. The mean deviation of the hemispherical sub-region is 0.008 mm, and the standard deviation is √[Σ(deviation - 0.008)]. 2 / 71]≈0.001mm; the mean deviation of the transition sub-region is 0.007mm, and the standard deviation is ≈0.002mm; the mean deviation of the cylindrical surface sub-region is 0.005mm, and the standard deviation is ≈0.0008mm. Dividing the standard deviation by the mean deviation of each region generates the profile deviation dispersion coefficient: hemispherical surface 0.125, transition region 0.286, cylindrical surface 0.16. The larger the coefficient, the more dispersed the deviation distribution within the region.

[0030] Furthermore, the initial contour complexity coefficients corresponding to different processing areas are obtained by weighting the gradient parameter of the radius of curvature change, the rate of change of the radius of curvature, and the dispersion coefficient of the contour deviation; the initial contour complexity coefficients are spatially weighted and corrected by the trajectory density distribution map to obtain the contour complexity coefficients corresponding to different processing areas.

[0031] In this embodiment of the invention, the following weights are set: 0.4 for the curvature radius change gradient parameter, 0.3 for the curvature radius change rate, and 0.3 for the contour deviation dispersion coefficient. The parameter values ​​for each sub-region are: hemispherical gradient 0.05, change rate 0.03 mm / mm. 2 The dispersion is 0.125, and the weighted average is 0.4 × 0.05 + 0.3 × 0.03 + 0.3 × 0.125 = 0.0665; the gradient of the transition region is 0.4, and the rate of change is 0.3 mm / mm. 2The initial contour complexity coefficients are: hemispherical surface 0.0665, transition region 0.3358, and cylindrical surface 0.048. The gradient, rate of change, and dispersion of the cylindrical surface are 0.16, and the weighted calculation is 0 + 0 + 0.3 × 0.16 = 0.048. Next, the trajectory density weights for each sub-region are obtained from the trajectory density distribution map: hemispherical region: 10 trajectories / 10mm, weight 1.2; transition region: 8 trajectories / 10mm, weight 1.0; cylindrical surface: 5 trajectories / 10mm, weight 0.8. The initial contour complexity coefficients are multiplied by the corresponding density weights for spatial weighting correction: hemispherical surface = 0.0665 × 1.2 ≈ 0.0798; transition region = 0.3358 × 1.0 = 0.3358; cylindrical surface = 0.048 × 0.8 ≈ 0.0384, yielding the final contour complexity coefficients. The transition region has the highest coefficient, indicating that this region is the most difficult to process, providing a basis for subsequent processing parameter adjustments.

[0032] Furthermore, the step of identifying the convex-concave transition region of the surface through the sign change analysis corresponding to the Gaussian curvature value includes the following steps: Based on the Gaussian curvature values ​​and the corresponding three-dimensional coordinate parameters of each sampling point in the surface design model, a coordinate-Gaussian curvature association dataset is constructed. The three-dimensional coordinate parameters include X-axis coordinate values, Y-axis coordinate values, and Z-axis coordinate values, and the Gaussian curvature values ​​include positive curvature values, negative curvature values, and zero curvature values. Based on the coordinate-Gaussian curvature association dataset, the sampling points are divided into grids along the Z-axis coordinate direction in ascending order of X-axis coordinate values ​​and ascending order of Y-axis coordinate values, generating sampling grid matrices for each surface. Each grid cell contains at least 3 adjacent sampling points and their corresponding Gaussian curvature values. In this embodiment of the invention, the Gaussian curvature values ​​(0.0004 mm for a hemispherical surface) of 300 sampling points are extracted from the surface curvature feature matrix. -2 (Positive), Cylindrical surface 0mm -2 (Zero), Transition zone 0.00003-0.0004mm -2 (Positive) Ten new sampling points were added in the local depression area, with a Gaussian curvature of -0.0002mm. -2(Negative)), combined with the three-dimensional coordinate parameters of each sampling point (such as hemispherical surface (100,75,130), cylindrical surface (100,75,80), concave region (110,85,110)), arranged in the format of "X coordinate-Y coordinate-Z coordinate-Gaussian curvature value-curvature sign", to construct a coordinate-Gaussian curvature association dataset containing 310 data points, each data point precisely corresponding to the spatial location and curvature attribute of the sampling point. Secondly, based on the coordinate-Gaussian curvature association dataset, along the Z-axis coordinate value direction (80-130mm), the sampling points are divided into a grid at 5mm intervals for X-axis coordinate values ​​(90-120mm) and 5mm intervals for Y-axis coordinate values ​​(70-90mm), resulting in a total of 10×4×10=400 grid units (one layer every 5mm along the Z-axis). Each grid cell measures 5mm × 5mm × 5mm, ensuring that each cell contains at least 3 adjacent sampling points. For example, a grid cell with Z = 105-110mm, X = 100-105mm, and Y = 75-80mm contains 3 sampling points (100, 75, 108), (102, 77, 106), and (104, 79, 109) and their corresponding Gaussian curvature values, generating a matrix of 400 surface sampling grids, clearly defining the spatial range and sampling point information of each grid.

[0033] Furthermore, the Gaussian curvature sign of adjacent sampling points in each grid cell of each surface sampling grid matrix is ​​determined, and the number of positive curvature sampling points, negative curvature sampling points, and zero curvature sampling points in each grid cell are counted to generate grid cell curvature sign distribution parameters. Based on the grid cell curvature sign distribution parameters, the sign change coefficient of each grid cell is calculated, which is the ratio of the number of pairs of adjacent sampling points with different curvature signs to the total number of pairs of adjacent sampling points. If the sign change coefficient is greater than a preset threshold, the grid cell is marked as a potential convex-concave transition grid, and a potential transition grid label dataset is generated. In this embodiment of the invention, the Gaussian curvature sign of adjacent sampling points within each grid cell of each surface sampling grid matrix is ​​determined. For example, in the target grid cell of the Z=105-110mm layer, the curvature signs of the three sampling points are positive, positive, and negative, respectively. The number of positive curvature sampling points in the cell is counted as 2, negative curvature as 1, and zero curvature as 0, recorded as (2,1,0). In another cylindrical surface grid cell (Z=80-85mm), the curvature of all three sampling points is zero, recorded as (0,0,3). All 400 grid cells are counted in this format to generate grid cell curvature sign distribution parameters, which intuitively reflect the distribution of curvature signs within the cell. Next, the sign change coefficient is calculated based on the grid cell curvature sign distribution parameters. The formula is: sign change coefficient = number of pairs of adjacent sampling points with different curvature signs / total number of pairs of adjacent sampling points. Within the target grid cell, three sampling points form three adjacent pairs (point 1-point 2: positive-positive, point 1-point 3: positive-negative, point 2-point 3: positive-negative). The number of pairs with different signs is 2, the total number of pairs is 3, and the coefficient = 2 / 3 ≈ 0.67. A preset threshold of 0.5 is set; 0.67 > 0.5, and the grid cell is marked as a potential convex-concave transition grid. Grid cells in the cylindrical region have zero-zero adjacent pairs, a coefficient of 0, and are not marked. A total of 68 potential convex-concave transition grids are marked, generating a potential transition grid labeled dataset, including grid coordinate ranges and labeling results.

[0034] Furthermore, the three-dimensional coordinate range of potential convex-concave transition meshes is extracted from the potential transition mesh marker dataset. Combined with the feature contour data of the surface design model, it is determined whether the potential convex-concave transition mesh coincides with the surface feature boundary in the design. If they coincide, the mesh is removed, and a filtered transition mesh dataset is generated. For each transition mesh in the filtered transition mesh dataset, the Gaussian curvature values ​​of all sampling points inside it are extracted. The Gaussian curvature difference between adjacent sampling points is calculated to generate a curvature difference sequence. At the same time, the sampling point coordinate difference corresponding to each difference is recorded. In this embodiment of the invention, the three-dimensional coordinate range of potential convex-concave transition meshes (e.g., target mesh Z=105-110mm, X=100-105mm, Y=75-80mm) is extracted from the potential transition mesh marker dataset. Combined with the feature contour data of the surface design model (hemispherical equatorial circle Z=105mm, X95-105mm, Y70-80mm), the target mesh coordinate range is determined to have an 80% overlap with the equatorial circle contour, thus being identified as a design feature boundary mesh and discarded. Another potential mesh (Z=110-115mm, X=110-115mm, Y=85-90mm) has a 10% overlap with the design feature contour and is retained. Finally, 42 non-feature boundary transition meshes are selected, generating a filtered transition mesh dataset for accurate localization of the actual convex-concave transition region. Secondly, from the 42 transition grids in the filtered transition grid dataset, one typical transition grid (Z=110-115mm, X=110-115mm, Y=85-90mm) was selected, and the Gaussian curvature values ​​of four sampling points inside were extracted: point A(110,85,112) 0.0001mm -2 Point B (112, 87, 113) - 0.0001mm -2 Point C(114,89,114)-0.0002mm -2 Point D(111,88,111)0mm -2 Calculate the Gaussian curvature difference between adjacent sampling points: AB = 0.0002 mm -2 BC = -0.0001mm -2 CD=0.0002mm -2 DA=0.0001mm -2 A curvature difference sequence [0.0002, -0.0001, 0.0002, 0.0001] is generated. The coordinate differences of the sampling points corresponding to each difference are recorded synchronously: AB coordinate difference (2,2,1)mm, BC (2,2,1)mm, CD (-3,-1,-3)mm, DA (-1,-3,-1)mm, to ensure that the difference is associated with the spatial location.

[0035] Furthermore, based on the curvature difference sequence and the corresponding sampling point coordinate difference, the curvature change gradient is calculated to generate the curvature change gradient matrix for each transition grid. According to the curvature change gradient matrix, the transition sampling point with the largest curvature change gradient in each transition grid is identified. At the same time, the number and distribution density of transition sampling points in each transition grid are counted to generate the distribution parameters of transition feature points. In this embodiment of the invention, the curvature gradient is calculated based on the curvature difference sequence and the corresponding coordinate difference, using the formula: curvature change gradient = curvature difference / coordinate difference magnitude. The coordinate difference magnitude is calculated as √(ΔX). 2+ΔY 2 +ΔZ 2 For example, the gradient of AB = 0.0002 / √(2). 2 +2 2 +1 2 = 0.0002 / 3 ≈ 6.67 × 10 -5 mm -3 BC gradient = -0.0001 / 3 ≈ -3.33 × 10⁻⁶ -5 mm -3 CD gradient = 0.0002 / √((-3)) 2 +(-1) 2 +(-3) 2 )≈4.56×10 -5 mm -3 DA gradient = 0.0001 / √11 ≈ 3.02 × 10 -5 mm -3 Arrange the four gradient values ​​in adjacent pairs to construct a 4×1 dimension curvature change gradient matrix [6.67×10]. -5 -3.33×10 -5 4.56×10 -5 3.02×10 -5 Each matrix element corresponds to the rate of curvature change of a segment of adjacent points within the grid. Next, by comparing the absolute values ​​of the four gradients based on the curvature change gradient matrix, the absolute value of the AB gradient is 6.67 × 10⁻⁶. -5 Maximum, corresponding to transition sampling points A and B; absolute value of gradient CD 4.56 × 10⁻⁶ -5 Next, corresponding to transition sampling points C and D, four transition sampling points (A, B, C, and D) were identified within the transition grid. The total number of transition sampling points was 4, and the grid cell volume was 125 mm². 3 (5mm×5mm×5mm), distribution density = 4 / 125 = 0.032 particles / mm 3 This operation is performed on all 42 transition grids. For example, if another transition grid contains 3 transition sampling points, the distribution density is 0.024 points / mm. 3 Generate transition feature point distribution parameters, including the number of transition points per grid (3-5) and the distribution density (0.024-0.04 points / mm). 3 ).

[0036] Furthermore, spatial connectivity analysis is performed on the transition feature points within adjacent transition grids. The least squares method is used to fit the spatial curve of the transition feature points. If the radius of curvature of the fitted curve is within a preset range and is continuous without discontinuities, the curve is marked as the preliminary convex-concave transition boundary line, generating a preliminary boundary line dataset. The preliminary boundary line dataset is compared with the CAD data of the surface design model, and the spatial deviation between the preliminary boundary line and the design theoretical boundary line is calculated. If the deviation is less than a preset accuracy threshold, the preliminary boundary line is retained; otherwise, the fitting parameters of the transition feature points are corrected based on the deviation value, and the corrected transition boundary line is regenerated. In this embodiment of the invention, spatial connection analysis is performed on the transition feature points (A, B, C, D in grid 1, E, F, G, H in grid 2) within two adjacent transition grids (grid 1: Z=110-115mm, grid 2: Z=115-120mm) to obtain the three-dimensional coordinates of eight transition feature points. The least squares method is used to fit a spatial curve. Let the equations of the fitted curve be parametric equations X=a1t+b1, Y=a2t+b2, Z=a3t+b3 (t is a parameter). Substituting the coordinates of the eight points, we obtain a1=2, b1=110, a2=2, b2=85, a3=1, b3=112. The corresponding radius of curvature is calculated as √((a1...)... 2 +a2 2 +a3 2 ) 3 ) / (√((a2a3'-a3a2') 2 +(a3a1'-a1a3') 2 +(a1a2'-a2a1') 2 Since the parametric equation is a straight line (derivative is 0), the radius of curvature approaches infinity, is within a preset range (>100mm), and is continuous without discontinuities, this curve is marked as the preliminary convex-concave transition boundary line. Fitting all adjacent transition mesh combinations generates 15 preliminary boundary lines, creating a preliminary boundary line dataset containing the parametric equation and coordinate range of each boundary line. Next, by comparing the preliminary boundary line dataset (15 boundary lines) with the CAD data of the surface design model, one preliminary boundary line is selected (parametric equation X=2t+110, Y=2t+85, Z=t+112, t∈[0,5]), whose corresponding theoretical boundary line CAD data is X=2t+110.1, Y=2t+85.1, Z=t+112.05, t∈[0,5]. The spatial deviation between the two boundary lines is calculated using 10 uniformly distributed sampling points (t=0,0.5,…,5), calculated as follows: Spatial deviation = √[(ΔX)] 2 +(ΔY) 2 +(ΔZ) 2 Calculate, for example, at t=0, the deviation = √[(0.1)]. 2 +(0.1)2 +(0.05) 2 The deviation is approximately 0.14 mm at t=5, with an average deviation of 0.14 mm. A preset accuracy threshold of 0.15 mm is set; 0.14 mm < 0.15 mm, so this initial boundary line is retained. Another initial boundary line has an average deviation of 0.16 mm. Based on the deviation value, the fitting parameters are adjusted (e.g., the X coefficient is adjusted from 2 to 2.01), and the corrected transition boundary line is refitted to ensure that the deviation of all boundary lines is less than the threshold.

[0037] Furthermore, the starting coordinates, ending coordinates, length, curvature variation range, and angle with adjacent surfaces of the corrected transition boundary lines are extracted. Simultaneously, the area of ​​the convex-concave transition region corresponding to each boundary line is calculated to generate a basic parameter set for the transition boundary. Combining this basic parameter set with the previously generated curvature variation gradient matrix, the boundary region stability coefficient is calculated. Specifically, this coefficient is the ratio of the length to the area of ​​the corresponding convex-concave transition region, and the product of the curvature variation range and the average curvature variation gradient within the region. The boundary region stability coefficient is then fused with the basic parameter set to generate regional boundary feature parameters.

[0038] In this embodiment of the invention, basic parameters are extracted from the 12 retained preliminary boundary lines and 3 corrected transition boundary lines. Taking the target boundary line as an example: the starting point coordinates (t=0) are (110, 85, 112), and the ending point coordinates (t=5) are (120, 95, 117); the length is calculated using the formula: length = √[(ΔX)]. 2 +(ΔY) 2 +(ΔZ) 2 Calculate ΔX=10mm, ΔY=10mm, ΔZ=5mm, length=√(100+100+25)=15mm; the curvature variation range is [0,0] because the boundary line is a straight line and the curvature is always 0; the angle with the adjacent hemisphere is calculated by the vector dot product, the boundary line vector is (10,10,5), the hemisphere normal vector is (0,0,1), and the angle is arccos(5 / (15×1))≈70.5°; the area of ​​the corresponding convex-concave transition region is calculated by the product of the boundary line length and the region width (5mm), and the area is 15×5=75mm. 2 This process is repeated for all 15 boundary lines to generate a basic parameter set for the transition boundary, containing 6 basic parameters for each boundary line. Next, the basic parameter set for the transition boundary is combined with the curvature change gradient matrix (the gradient matrix of the target boundary line corresponding to the mesh [6.67×10⁻⁶]). -5 -3.33×10 -5 4.56×10 -5 3.02×10 -5 ]), calculate the stability coefficient of the boundary region. First, calculate the ratio of the length to the area of ​​the transition region = 15 / 75 = 0.2 mm.-1 The average curvature gradient within the calculated region is calculated as (6.67 + 3.33 + 4.56 + 3.02) × 10⁻⁶. -5 / 4≈4.395×10 -5 mm -3 The product of the curvature variation range and the average gradient is 0 × 4.395 × 10⁻⁶. -5 =0; Calculated using the formula Stability Coefficient = (Length-to-Area Ratio × 0.6) + (Product × 0.4), the Stability Coefficient is 0.2 × 0.6 + 0 = 0.12. The Stability Coefficient is then integrated with the basic parameters of the transition boundary (start point, end point, length, etc.) and added to the regional boundary feature parameters. For example, a new "Stability Coefficient 0.12" is added to the regional boundary feature parameters of the target boundary line, providing a basis for the stability of the boundary region in subsequent processing parameter adjustments.

[0039] Furthermore, step S22 includes the following steps: Based on the rate of change of curvature radius corresponding to different processing areas and combined with the accuracy tolerance requirements parameters in the surface design model data, including shape tolerance, position tolerance and surface roughness tolerance, a curvature-tolerance correlation matrix is ​​constructed, in which each element corresponds to the matching relationship between the rate of change of curvature radius of a single processing area and the accuracy tolerance of that area, generating the accuracy requirement mapping parameters for the processing area. In this embodiment of the invention, the curvature radius change rate (cylindrical surface 0 mm / mm) of different processing areas is used as the basis. 2 , hemispherical surface 0.03mm / mm 2 Transition area 0.3mm / mm 2 Combining the accuracy tolerance requirements of the surface design model (cylindrical surface: shape tolerance 0.02mm, position tolerance 0.04mm, surface roughness Ra 0.8μm; hemispherical surface: shape tolerance 0.01mm, position tolerance 0.03mm, surface roughness Ra 0.6μm; transition region: shape tolerance 0.008mm, position tolerance 0.02mm, surface roughness Ra 0.4μm), a 3×3 curvature-tolerance correlation matrix is ​​constructed. The matrix elements are "curvature radius change rate - accuracy tolerance matching degree", such as cylindrical surface elements (0, 0.02, 0.04, 0.8), hemispherical surface elements (0.03, 0.01, 0.03, 0.6), and transition region elements (0.3, 0.008, 0.02, 0.4), generating accuracy requirement mapping parameters for the machining area and clarifying the correspondence between curvature and accuracy in different regions.

[0040] Furthermore, the complexity level of different processing areas is determined based on the contour complexity coefficient, where a contour complexity coefficient ≤ 0.3 is low complexity, 0.3 < contour complexity coefficient ≤ 0.7 is medium complexity, and contour complexity coefficient > 0.7 is high complexity. At the same time, the feature contour line data in the surface design model, including spheres, parabolic surfaces and irregular free surfaces, are combined to generate processing area complexity-feature correlation parameters. In this embodiment of the invention, the complexity level is determined based on the contour complexity coefficients (cylindrical surface 0.0384, hemispherical surface 0.0798, transition region 0.3358): cylindrical surface ≤ 0.3 is low complexity, hemispherical surface ≤ 0.3 is low complexity, and transition region 0.3 < 0.3358 ≤ 0.7 is medium complexity. Combining the feature contour data of the surface design model (cylindrical surface belongs to cylindrical feature, hemispherical surface belongs to spherical feature, transition region belongs to irregular freeform surface feature), a processing area complexity-feature correlation parameter is generated: cylindrical surface (low complexity - cylindrical surface), hemispherical surface (low complexity - spherical surface), transition region (medium complexity - irregular freeform surface), establishing the correlation between complexity and surface features.

[0041] Furthermore, the tool database of the five-axis linkage machining unit is called to obtain the material property parameters of the tool currently being machined, including hardness, wear resistance coefficient and thermal expansion coefficient, and structural parameters, including tool radius, cutting edge angle and number of cutting edges. Combined with the accuracy requirement mapping parameters of the machining area, the tool-accuracy adaptation coefficient is calculated. In this embodiment of the invention, by calling the tool database of the five-axis linkage machining unit, the tool currently being machined is a carbide ball end mill with the following material properties: hardness HRC65, wear resistance coefficient 0.8, and thermal expansion coefficient 12×10⁻⁶. -6 / ℃; Structural parameters: tool radius 5mm, cutting edge angle 30°, number of cutting edges 2. Based on the precision requirements of the machining area and mapping parameters, the tool-precision adaptation coefficient is calculated using the formula: tool-precision adaptation coefficient = (tool hardness / material hardness) × (tool radius / shape tolerance). For titanium alloy with a hardness of HRC35, the adaptation coefficient for cylindrical surfaces is (65 / 35) × (5 / 0.02) ≈ 46.43, for hemispherical surfaces it is (65 / 35) × (5 / 0.01) ≈ 92.86, and for transition areas it is (65 / 35) × (5 / 0.008) ≈ 116.07. A higher coefficient indicates higher tool adaptation precision.

[0042] Furthermore, the complexity-feature correlation parameters of the machining area and the tool-precision adaptation coefficient are input into the initial calculation model of the spindle speed. With the goal of minimizing tool wear rate in low complexity area and optimizing machining stability in high complexity area, the initial spindle speed values ​​of different machining areas are calculated. At the same time, the initial speed values ​​are corrected based on the rate of change of curvature radius, and real-time dynamic adjustment parameters of spindle speed are generated. In this embodiment of the invention, by inputting the machining area complexity-feature correlation parameters and the tool-precision adaptation coefficient into the initial calculation model of the spindle speed, the low complexity area (cylindrical surface, hemispherical surface) aims to minimize the tool wear rate, and the initial speed is calculated as (1000 × tool hardness) / (material hardness × tool radius). The initial speed of the cylindrical surface is approximately 371 r / min (1000 × 65) / (35 × 5) and the hemispherical surface is approximately 371 r / min. The medium complexity area (transition area) aims to optimize machining stability, and the initial speed is approximately 464 r / min (1200 × adaptation coefficient) / (curvature radius change rate × 1000) ≈ (1200 × 116.07) / (0.3 × 1000). Based on the rate of change of curvature radius, the rotational speed is corrected as follows: cylindrical surface correction coefficient 0.9 (low rate of change), final rotational speed 334 r / min; hemispherical surface correction coefficient 1.0, final rotational speed 371 r / min; transition region correction coefficient 1.1 (high rate of change), final rotational speed 510 r / min, generating real-time dynamic adjustment parameters for spindle speed.

[0043] Furthermore, based on the real-time spindle speed dynamic adjustment parameters and the mapping parameters of the machining area accuracy requirements, the feed rate constraint threshold is calculated, specifically the maximum feed rate upper limit that meets the accuracy tolerance requirements. The feed rate adjustment weight is determined by combining the contour complexity coefficient, and feed rate constraint parameters are generated. The servo system performance parameters of the five-axis linkage machining unit are called, and combined with the feed rate constraint parameters, the machining trajectory deviation under different feed rates is simulated by the kinematic simulation algorithm. The feed rate range with trajectory deviation less than the accuracy tolerance requirements is selected, and feed rate dynamic adjustment parameters are generated.

[0044] In this embodiment of the invention, based on the real-time spindle speed dynamic adjustment parameters and the accuracy requirement mapping parameters, the feed rate constraint threshold is calculated according to the formula: (spindle speed × tool radius × shape tolerance) / (1000 × surface roughness). The threshold values ​​are: cylindrical surface threshold = (334 × 5 × 0.02) / (1000 × 0.8) ≈ 0.042 mm / r, hemispherical surface threshold = (371 × 5 × 0.01) / (1000 × 0.6) ≈ 0.031 mm / r, and transition area threshold = (510 × 5 × 0.008) / (1000 × 0.4) ≈ 0.051 mm / r. The adjustment weights are determined by combining the contour complexity coefficient (0.8 for low complexity, 1.2 for medium complexity). The feed rates are: cylindrical surface = 0.042 × 0.8 ≈ 0.034 mm / r, hemispherical surface = 0.031 × 0.8 ≈ 0.025 mm / r, and transition region = 0.051 × 1.2 ≈ 0.061 mm / r. The servo system performance parameters (maximum feed rate 0.1 mm / r) are called, and kinematic simulation algorithms are used to select the feed rate ranges where the trajectory deviation is less than the shape tolerance (cylindrical surface 0.02-0.034 mm / r, transition region 0.04-0.061 mm / r), generating dynamic adjustment parameters for the feed rate.

[0045] Furthermore, step S23 includes the following steps: The ultrasonic vibration-assisted unit is used to perform time-domain and frequency-domain analysis on the initial cutting force data in the initial contact state data to extract the peak cutting force parameters, fluctuation frequency characteristics and duration distribution; In this embodiment of the invention, the initial cutting force data (radial force 250N, axial force 180N, tangential force 120N, sampling frequency 10kHz) in the initial contact state data is read by the ultrasonic vibration auxiliary unit, and time-domain analysis is performed: the peak cutting force parameters (peak radial force 280N, peak axial force 200N, peak tangential force 140N) are extracted, and the duration distribution is statistically analyzed (peak radial force lasts 0.02s, axial force 0.015s, tangential force 0.01s); frequency-domain analysis is performed: the time-domain signal is converted into a frequency-domain signal through Fourier transform, the fluctuation frequency characteristics are identified (radial force main frequency 150Hz, axial force 120Hz, tangential force 100Hz, and harmonic frequencies are all integer multiples of 50Hz), and a cutting force feature dataset containing peak value, frequency, and duration is generated, providing a mechanical basis for vibration parameter calculation.

[0046] Furthermore, the hardness parameters of the titanium alloy material and the characteristic parameters of the tool material are obtained and a material-tool characteristic matrix is ​​constructed. At the same time, based on the peak cutting force parameters, fluctuation frequency characteristics and duration distribution and the material-tool characteristic matrix, the energy input parameters and amplitude threshold range required for ultrasonic vibration are calculated. In this embodiment of the invention, the hardness parameter (HRC35) of the titanium alloy material and the characteristic parameters of the cutting tool material (Hard alloy HRC65, wear resistance coefficient 0.8, thermal expansion coefficient 12×10⁻⁶) are obtained. -6 / ℃), construct a 2×3 dimension material-tool characteristic matrix (rows: titanium alloy, cemented carbide; columns: hardness, wear resistance, coefficient of thermal expansion). Combining the cutting force peak parameter (average peak 206.7N), fluctuation frequency characteristics (average dominant frequency 123.3Hz), duration distribution (average duration 0.015s) and characteristic matrix, calculate the ultrasonic vibration energy input parameter according to the formula: Energy = (average peak × duration) / (material hardness / tool hardness) = (206.7 × 0.015) / (35 / 65) ≈ 0.56J; calculate the amplitude threshold range: Amplitude = (average dominant frequency × tool radius) / (1000 × material hardness), tool radius 5mm, amplitude range = (123.3 × 5) / (1000 × 35) ≈ 0.0176-0.02mm (considering ±5% error), generate the basic parameters of energy and amplitude.

[0047] Furthermore, the initial amplitude parameters of ultrasonic vibration are calculated and generated based on the energy input parameters and amplitude threshold range, combined with the path interval distribution data in the preset processing trajectory planning data. In this embodiment of the invention, based on the energy input parameter (0.56J) and the amplitude threshold range (0.0176-0.02mm), combined with the path interval distribution data in the preset processing trajectory planning data (cylindrical surface 2mm, hemispherical surface 1mm, transition region 1-2mm), the initial amplitude of ultrasonic vibration is calculated according to the formula: initial amplitude of ultrasonic vibration = amplitude threshold × (path interval / average path interval). The average path interval is 1.5mm: initial amplitude of cylindrical surface = 0.0188 × (2 / 1.5) ≈ 0.025mm; hemispherical surface = 0.0188 × (1 / 1.5) ≈ 0.0125mm; transition region = 0.0188 × (1.5 / 1.5) ≈ 0.0188mm. Energy matching is verified: the energy corresponding to the amplitude of each region is close to 0.56J (error < 3%), generating the initial amplitude parameters of ultrasonic vibration to ensure that the amplitude and path interval are compatible.

[0048] Furthermore, based on the matching analysis between the fluctuation frequency characteristics and duration distribution and the real-time spindle speed dynamic adjustment parameters, the initial adjustment range corresponding to the ultrasonic vibration is determined, and the initial frequency parameters of ultrasonic vibration are calculated by combining the surface curvature change parameters. At the same time, the initial amplitude parameters of ultrasonic vibration are integrated with the initial amplitude parameters of ultrasonic vibration to generate high-frequency vibration parameters.

[0049] In this embodiment of the invention, a matching analysis was performed based on the fluctuation frequency characteristics (average main frequency 123.3Hz) and the real-time spindle speed dynamic adjustment parameters (cylindrical surface 334r / min, hemispherical surface 371r / min, transition region 510r / min). The analysis showed a positive correlation between speed and frequency, thus determining the initial adjustment range for ultrasonic vibration (120-150Hz). This was combined with the surface curvature variation parameters (cylindrical surface 0mm). -1 0.02mm hemispherical surface -1 Transition zone 0.005-0.02mm -1 The initial frequency is calculated using the formula: Initial adjustment interval median × (1 + curvature value): Cylindrical surface frequency = 135 × (1 + 0) = 135 Hz; Hemispherical surface = 135 × (1 + 0.02) = 137.7 Hz; Transition region = 135 × (1 + 0.0125) ≈ 136.69 Hz. The initial frequency and initial amplitude (cylindrical surface 0.025 mm, hemispherical surface 0.0125 mm, transition region 0.0188 mm) are integrated to generate high-frequency vibration parameters, including the frequency and amplitude combination of each processing area.

[0050] Furthermore, step S3 includes the following steps: Step S31: The real-time contour data of the surface processing area is continuously collected by the real-time contour monitoring unit, and the real-time contour data is registered and compared with the CAD data of the surface design model to calculate the deviation value between the actual position and the theoretical position of each sampling point, and generate a surface processing contour deviation distribution map. In this embodiment of the invention, the laser scanning probe of the real-time contour monitoring unit (scanning line speed 50 mm / s, point cloud density 100 points / mm) is used. 2 The system continuously collects real-time contour data of the surface processing area, generating a point cloud dataset containing 500,000 points (coordinate accuracy 0.003mm) every 0.5 seconds. The real-time point cloud data is then registered with the CAD data of the surface design model (hemispherical radius 50mm, cylindrical diameter 80mm) using an iterative nearest-point algorithm, with the registration error controlled within 0.002mm. For each sampling point, the deviation between the actual and theoretical positions is calculated. For example, the actual deviation of the hemispherical sampling point (104,75,127) is 0.012mm, the cylindrical sampling point (100,75,82) has a deviation of 0.008mm, and the transition area sampling point (102,77,106) has a deviation of 0.015mm. These deviations are color-coded (≤0.01mm blue, 0.01-0.015mm yellow, >0.015mm red) to generate a surface processing contour deviation distribution map, visually displaying the deviation distribution.

[0051] Step S32: Extract the maximum deviation value, average deviation value, and deviation concentration area parameters based on the surface machining profile deviation distribution map, and comprehensively calculate the surface machining profile deviation value; In this embodiment of the invention, key parameters are extracted based on the surface machining contour deviation distribution map: the maximum deviation value is 0.018 mm at the sampling point (103, 78, 105) in the transition area; the average deviation value is calculated by summing and averaging the deviation values ​​of 500,000 sampling points, resulting in 0.011 mm; the deviation concentration area parameter is the transition area (Z100-110 mm, X100-105 mm, Y75-80 mm), where the deviation value is all > 0.012 mm, accounting for 15% of the total machining area. The deviation value is calculated using the formula: Surface machining contour deviation value = (maximum deviation value × 0.4) + (average deviation value × 0.3) + (percentage of deviation concentration area × 0.3), resulting in deviation value = (0.018 × 0.4) + (0.011 × 0.3) + (0.15 × 0.3) = 0.0072 + 0.0033 + 0.045 = 0.0555, which quantitatively reflects the overall machining contour deviation degree.

[0052] Step S33: Collect real-time temperature data in the curved surface machining area through the temperature sensor array corresponding to the cooling and lubrication unit and generate temperature field distribution data of the machining area; In this embodiment of the invention, a temperature sensor array (20 sensors, spaced 10 mm apart, measuring range -20-300℃, accuracy ±0.5℃) of the cooling and lubrication unit collects real-time temperature data of the curved surface machining area at a sampling frequency of 1 Hz. After 10 minutes of machining, the sensor data shows: the temperature in the cylindrical area is 45-55℃, the temperature in the hemispherical area is 55-65℃, and the temperature in the transition area is 65-75℃. The highest temperature occurs at the tool contact point in the transition area (78℃). The temperature data is correlated with the sensor coordinates, and an interpolation algorithm is used to generate temperature field distribution data of the machining area, presented in the form of isotherms (four isotherms at 45℃, 55℃, 65℃, and 75℃), clearly indicating the temperature gradient changes in different areas.

[0053] Step S34: Based on the temperature field distribution data of the processing area, analyze the location parameters, temperature gradient change rate and duration parameters of each temperature zone, and calculate the thermal deformation influence coefficient in combination with the thermal expansion coefficient parameter of the titanium alloy material. In this embodiment of the invention, temperature zones are divided based on the temperature field distribution data of the processing area: a low-temperature zone (45-55℃, cylindrical surface), a medium-temperature zone (55-65℃, hemispherical surface), and a high-temperature zone (65-75℃, transition zone). Positional parameters for each zone are extracted (Z80-90mm for the low-temperature zone, Z90-100mm for the medium-temperature zone, and Z100-110mm for the high-temperature zone). The temperature gradient change rate is calculated as follows: 5℃ / 10mm = 0.5℃ / mm for the low-temperature zone, 10℃ / 10mm = 1℃ / mm for the medium-temperature zone, and 10℃ / 10mm = 1℃ / mm for the high-temperature zone. The duration parameter is the length of time each zone maintains its current temperature (8min for the low-temperature zone, 10min for the medium-temperature zone, and 12min for the high-temperature zone). This is combined with the coefficient of thermal expansion of titanium alloy, which is 10.8 × 10⁻⁶. -6 / ℃, calculated using the formula: Thermal deformation influence coefficient = Temperature × Coefficient of thermal expansion × Duration: Low temperature zone = 50 × 10.8 × 10 -6 ×8≈0.00432, Medium temperature zone = 60×10.8×10 -6 ×10≈0.00648, High-temperature zone = 70×10.8×10 -6 ×12≈0.00907, quantifying the degree of influence of thermal deformation on processing.

[0054] Step S35: Based on the surface machining contour deviation value and thermal deformation influence coefficient, perform dynamic compensation calculation on the sequence of collaborative control parameters for the machining process to generate adaptive machining parameters corresponding to the spindle speed correction, feed rate compensation value and ultrasonic vibration frequency adjustment.

[0055] In this embodiment of the invention, dynamic compensation is performed on the sequence of collaborative control parameters for the machining process based on the surface machining contour deviation value (0.0555) and the thermal deformation influence coefficient (maximum 0.00907 in the high-temperature zone). Spindle speed correction = original speed × (1 - deviation value × thermal deformation coefficient): original speed in the transition region is 510 r / min, correction = 510 × (1 - 0.0555 × 0.00907) ≈ 509.7 r / min; feed rate compensation = original feed rate × (1 - deviation value × thermal deformation coefficient): original feed rate in the transition region is 0.061 mm / r, compensation value ≈ 0.0609 mm / r; ultrasonic vibration frequency adjustment = original frequency × (1 + deviation value × thermal deformation coefficient): original frequency in the transition region is 136.69 Hz, adjustment ≈ 136.77 Hz. Adaptive machining parameters are generated by integrating these parameters to ensure dynamic adjustment based on contour deviation and thermal deformation, maintaining machining accuracy.

[0056] Furthermore, step S35 includes the following steps: Step S351: By setting the surface machining profile deviation threshold and the thermal deformation influence coefficient threshold, the generated surface machining profile deviation value and thermal deformation influence coefficient are compared with the corresponding thresholds respectively. When the surface machining profile deviation value exceeds the deviation threshold or the thermal deformation influence coefficient exceeds the influence coefficient threshold, the machining parameter correction mechanism is activated. In this embodiment of the invention, the surface machining contour deviation threshold is set to 0.05 (determined based on design accuracy requirements), and the hot deformation influence coefficient threshold is set to 0.008 (determined based on the critical value of hot deformation of titanium alloy). The surface machining contour deviation value of 0.0555 generated earlier is compared with the deviation threshold of 0.05; 0.0555 > 0.05. The hot deformation influence coefficient in the high-temperature region is compared with the influence coefficient threshold of 0.008; 0.00907 > 0.008. Both indicators exceed the threshold, thus initiating the machining parameter correction mechanism and triggering subsequent deviation compensation and temperature compensation processes to ensure that the machining accuracy does not exceed the allowable range.

[0057] Step S352: Based on the machining parameter correction mechanism, perform vector decomposition on the surface machining contour deviation value to obtain the deviation component along the surface normal and the deviation component along the machining trajectory tangent direction; calculate the adjustment amount corresponding to the tool feed depth based on the deviation component along the surface normal, and combine the deviation component along the machining trajectory tangent direction to determine the trajectory correction parameters to comprehensively generate the contour deviation compensation amount. In this embodiment of the invention, the surface machining contour deviation value of 0.0555 is vector-decomposed based on a machining parameter correction mechanism. The surface normal is the direction perpendicular to the machined surface (e.g., the normal of the transition region makes an angle of 30° with the Z-axis), and the machining trajectory tangent direction is the direction along the tool path (the tangent of the transition region makes an angle of 45° with the X-axis). Through vector projection calculation: the deviation component along the normal direction = 0.0555 × cos30° ≈ 0.048 mm, and the deviation component along the tangent direction = 0.0555 × sin30° ≈ 0.0278 mm. Based on the normal deviation component, the tool feed depth adjustment amount = 0.048 mm is calculated (the feed depth needs to be reduced to lower the normal deviation); combined with the tangent deviation component, the trajectory correction parameter (offset of 0.0278 mm along the tangent direction) is determined. The two are integrated to generate the contour deviation compensation amount (feed depth - 0.048 mm, trajectory offset + 0.0278 mm), accurately correcting the contour deviation.

[0058] Step S353: Analyze the temperature distribution characteristics of the surface processing area according to the thermal deformation influence coefficient and determine the position parameters and temperature values ​​corresponding to the area with the highest temperature; based on the position parameters and temperature values ​​corresponding to the area with the highest temperature, combined with the flow parameters and injection angle parameters of the cooling and lubrication unit, calculate the cooling intensity parameters required for temperature reduction, and convert the cooling intensity parameters into the corresponding temperature compensation coefficients. In this embodiment of the invention, by analyzing the temperature distribution characteristics based on the thermal deformation influence coefficient of 0.00907, the region with the highest temperature is determined to be the transition region (Z100-110mm, X100-105mm, Y75-80mm), with location parameters corresponding to coordinates (103, 78, 105) and a temperature of 78℃. The current flow rate of the cooling and lubrication unit is 10L / min, and the spray angle is 30° (angle with the processed surface). The cooling intensity parameter is calculated using the formula: Cooling Intensity Parameter = (Target Temperature Reduction × Material Specific Heat Capacity × Density × Processed Area Volume) / (Cooling Time × Coolant Specific Heat Capacity × Flow Rate). The target temperature is reduced to 70℃ (8℃ reduction). The specific heat capacity of titanium alloy is 523 J / (kg). ℃), density 4510kg / m³ 3 The processing area volume is 125mm. 3 Cooling time 1 minute, coolant specific heat capacity 4200 J / (kg) (℃), the cooling intensity parameter is calculated to be 12L / min. The cooling intensity parameter is converted into a temperature compensation coefficient: Temperature compensation coefficient = (new flow rate - original flow rate) / original flow rate × thermal deformation influence coefficient = (12-10) / 10 × 0.00907 ≈ 0.00181, which quantifies the compensation effect of cooling regulation on thermal deformation.

[0059] Step S354: Based on the contour deviation compensation amount and temperature compensation coefficient, perform dynamic compensation calculation on the sequence of collaborative control parameters for the machining process to calculate the spindle speed correction amount, feed rate compensation value and ultrasonic vibration frequency adjustment amount, and generate adaptive machining parameters.

[0060] In this embodiment of the invention, the sequence of collaborative control parameters for the machining process is dynamically compensated based on the contour deviation compensation amount (feed depth -0.048mm, trajectory offset +0.0278mm) and the temperature compensation coefficient 0.00181. Spindle speed correction amount = original speed 510r / min × (1 - contour deviation component × temperature compensation coefficient) = 510 × (1 - 0.048 × 0.00181) ≈ 509.95r / min; Feed rate compensation value = original feed rate 0.061mm / r × (1 - normal deviation component × temperature compensation coefficient) = 0.061 × (1 - 0.048 × 0.00181) ≈ 0.06095mm / r; Ultrasonic vibration frequency adjustment amount = original frequency 136.69Hz × (1 + tangent deviation component × temperature compensation coefficient) = 136.69 × (1 + 0.0278 × 0.00181) ≈ 136.75Hz. By integrating three compensation parameters, adaptive machining parameters are generated to ensure that the spindle, feed, and vibration parameters are synchronously adapted to deviations and temperature changes, thus maintaining machining accuracy.

[0061] Furthermore, the present invention also provides a system for machining curved surfaces of titanium alloy materials, such as... Figure 4 As shown, the device includes a computer-readable storage medium 1003, a processor 1001, a communication interface 1002, and a computer program stored on the computer-readable storage medium 1003 and executable on the processor 1001. The processor 1001, communication interface 1002, and computer-readable storage medium 1003 can be connected via a bus or other means. The communication interface 1002 is used to receive and send data. When the processor 1001 executes the computer program, it implements the titanium alloy material curved surface machining method described above.

[0062] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for machining curved surfaces of titanium alloy materials, applied to a titanium alloy curved surface machining device, wherein the titanium alloy curved surface machining device comprises a five-axis linkage machining unit, an ultrasonic vibration auxiliary unit, a real-time contour monitoring unit, and a cooling and lubrication unit, characterized in that, The method for processing curved surfaces of titanium alloy materials includes the following steps: Step S1: Obtain the initial blank parameters, surface design model data and preset machining trajectory planning data corresponding to the titanium alloy material through the five-axis linkage machining unit, collect the initial contact state data between the tool and the titanium alloy material using the ultrasonic vibration auxiliary unit, and combine the reference contour scanning data generated by the real-time contour monitoring unit to construct the initial parameter set for titanium alloy surface machining. Step S2: Based on the initial parameter set for machining titanium alloy curved surfaces, real-time dynamic adjustment parameters for spindle speed and feed rate are generated through a five-axis linkage machining unit. At the same time, high-frequency vibration parameters are output using an ultrasonic vibration auxiliary unit. The real-time dynamic adjustment parameters for spindle speed and feed rate and the high-frequency vibration parameters are then fused to generate a sequence of collaborative control parameters for the machining process. Step S3: Continuously collect real-time contour data of the surface machining area through the real-time contour monitoring unit, and combine it with the temperature field distribution data of the machining area fed back by the cooling and lubrication unit to calculate the surface machining contour deviation value and thermal deformation influence coefficient; Based on the surface machining contour deviation value and thermal deformation influence coefficient, perform dynamic compensation calculation on the machining process collaborative control parameter sequence to generate adaptive machining parameters including spindle speed correction, feed rate compensation value and ultrasonic vibration frequency adjustment. Step S4: Drive the five-axis linkage machining unit, ultrasonic vibration auxiliary unit and cooling and lubrication unit to operate in coordination according to the adaptive machining parameters to generate a precision machined surface of titanium alloy material. Then, use the real-time contour monitoring unit to detect the accuracy of the precision machined surface of titanium alloy material and generate a surface machining quality assessment report.

2. The method for processing curved surfaces of titanium alloy materials according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the three-dimensional dimensional parameters, material uniformity distribution data, and surface roughness parameters of the initial titanium alloy blank through the laser measurement component corresponding to the five-axis linkage machining unit, so as to generate the initial blank parameters; Step S12: Obtain the CAD model of the surface design corresponding to the initial billet of titanium alloy material based on the initial billet parameters, and extract the corresponding surface curvature change parameters, feature contour line data and accuracy tolerance requirements parameters according to the CAD model of the surface design to generate surface design model data; Step S13: Plan the initial machining trajectory based on the surface design model data, and obtain the trajectory node coordinate data, tool attitude angle parameters and path interval distribution data based on the initial machining trajectory to generate preset machining trajectory planning data; Step S14: Use the force sensor corresponding to the ultrasonic vibration auxiliary unit to collect the initial cutting force data and contact position coordinate parameters at the moment the tool comes into contact with the titanium alloy material, and generate initial contact state data; Step S15: The initial titanium alloy blank is scanned across the entire surface area by the laser scanning probe corresponding to the real-time contour monitoring unit to obtain the reference contour point cloud data. Based on the reference contour point cloud data and the preset processing trajectory planning data, coordinate matching is performed to generate the reference contour scanning data. Step S16: The initial blank parameters, surface design model data, preset machining trajectory planning data, initial contact state data and reference contour scanning data are fused to construct the initial parameter set for titanium alloy surface machining.

3. The method for processing curved surfaces of titanium alloy materials according to claim 2, characterized in that, Step S2 includes the following steps: Step S21: Extract the corresponding surface curvature change parameters and preset machining trajectory planning data from the initial parameter set of titanium alloy surface machining, and calculate the curvature radius change rate and contour complexity coefficient corresponding to different machining areas based on the surface curvature change parameters and preset machining trajectory planning data combined with the reference contour scanning data. Step S22: Based on the rate of change of curvature radius and contour complexity coefficient corresponding to different machining areas, and combined with the surface design model data, the five-axis linkage machining unit is used to generate the corresponding real-time spindle speed and feed rate dynamic adjustment parameters. Step S23: Using the ultrasonic vibration auxiliary unit, the initial amplitude and frequency parameters of the ultrasonic vibration are calculated based on the initial contact state data and the hardness parameters of the titanium alloy material and the characteristic parameters of the tool material, so as to generate high-frequency vibration parameters. Step S24: Perform time series synchronous analysis on the real-time spindle speed, feed rate dynamic adjustment parameters and high-frequency vibration parameters, output the real-time spindle speed change curve, feed rate dynamic change sequence and high-frequency vibration parameter optimization sequence, and fuse them to generate a machining process collaborative control parameter sequence.

4. The method for processing curved surfaces of titanium alloy materials according to claim 3, characterized in that, Step S21, which involves calculating the rate of change of curvature radius and contour complexity coefficient for different processing areas based on surface curvature variation parameters, preset processing trajectory planning data, and reference contour scanning data, includes the following steps: By extracting principal curvature values, curvature direction angles, and Gaussian curvature values ​​corresponding to each sampling point from the surface curvature variation parameters, and constructing a surface curvature feature matrix; and reading the corresponding trajectory node coordinate data and path interval distribution data from the preset processing trajectory planning data, a three-dimensional coordinate sequence of the processing trajectory and a trajectory density distribution map are generated. Spatial registration is performed between the reference contour point cloud data in the reference contour scanning data and the three-dimensional coordinate sequence of the processing trajectory to calculate the spatial distance deviation between the trajectory nodes and the reference contour points, and generate trajectory-contour matching deviation parameters; based on the principal curvature values ​​in the surface curvature feature matrix and the trajectory-contour matching deviation parameters, a differential geometry algorithm is used to calculate the initial value of the curvature radius corresponding to each trajectory node. The initial values ​​of the radius of curvature of adjacent trajectory nodes are differentially calculated, and the trajectory step size parameter in the path interval distribution data is combined to generate the gradient parameter of the radius of curvature change. The direction vector field of the direction of curvature change is constructed based on the gradient parameter of the radius of curvature change and the curvature direction angle, and the rate of change of radius of curvature corresponding to different processing areas is analyzed based on the direction vector field of the direction of curvature change. Gaussian curvature values ​​are extracted from the surface curvature feature matrix, and the convex-concave transition region of the surface is identified by analyzing the sign change corresponding to the Gaussian curvature values, generating region boundary feature parameters; the processing region is segmented based on the region boundary feature parameters, and the standard deviation of the trajectory-contour matching deviation parameter in each sub-region is calculated to generate the contour deviation dispersion coefficient. The initial contour complexity coefficients for different processing areas are obtained by weighting the gradient parameter of the radius of curvature change, the rate of change of the radius of curvature, and the dispersion coefficient of the contour deviation. The initial contour complexity coefficients are then spatially weighted and corrected using the trajectory density distribution map to obtain the contour complexity coefficients for different processing areas.

5. The method for processing curved surfaces of titanium alloy materials according to claim 4, characterized in that, The method of identifying the convex-concave transition region of a surface by analyzing the sign change corresponding to the Gaussian curvature value includes the following steps: Based on the Gaussian curvature values ​​and the corresponding three-dimensional coordinate parameters of each sampling point in the surface design model, a coordinate-Gaussian curvature association dataset is constructed. The three-dimensional coordinate parameters include X-axis coordinate values, Y-axis coordinate values, and Z-axis coordinate values, and the Gaussian curvature values ​​include positive curvature values, negative curvature values, and zero curvature values. Based on the coordinate-Gaussian curvature association dataset, the sampling points are divided into grids along the Z-axis coordinate direction in ascending order of X-axis coordinate values ​​and ascending order of Y-axis coordinate values, generating sampling grid matrices for each surface. Each grid cell contains at least 3 adjacent sampling points and their corresponding Gaussian curvature values. For each surface sampling grid matrix, the Gaussian curvature sign of adjacent sampling points is determined. The number of positive curvature sampling points, negative curvature sampling points, and zero curvature sampling points in each grid cell are counted to generate grid cell curvature sign distribution parameters. Based on the grid cell curvature sign distribution parameters, the sign change coefficient of each grid cell is calculated, which is the ratio of the number of pairs of adjacent sampling points with different curvature signs to the total number of pairs of adjacent sampling points. If the sign change coefficient is greater than a preset threshold, the grid cell is marked as a potential convex-concave transition grid, and a potential transition grid label dataset is generated. The three-dimensional coordinate range of potential convex-concave transition meshes is extracted from the potential transition mesh marker dataset. Combined with the feature contour data of the surface design model, it is determined whether the potential convex-concave transition mesh coincides with the surface feature boundary in the design. If they coincide, the mesh is removed, and a filtered transition mesh dataset is generated. For each transition mesh in the filtered transition mesh dataset, the Gaussian curvature values ​​of all sampling points inside it are extracted. The Gaussian curvature difference between adjacent sampling points is calculated to generate a curvature difference sequence. At the same time, the sampling point coordinate difference corresponding to each difference is recorded. Based on the curvature difference sequence and the corresponding sampling point coordinate difference, the curvature change gradient is calculated to generate the curvature change gradient matrix for each transition grid. According to the curvature change gradient matrix, the transition sampling point with the largest curvature change gradient in each transition grid is identified. At the same time, the number and distribution density of transition sampling points in each transition grid are counted to generate the distribution parameters of transition feature points. Spatial connectivity analysis is performed on the transition feature points within adjacent transition grids. The least squares method is used to fit the spatial curve of the transition feature points. If the radius of curvature of the fitted curve is within a preset range and is continuous without discontinuities, the curve is marked as the preliminary convex-concave transition boundary line, and a preliminary boundary line dataset is generated. The preliminary boundary line dataset is compared with the CAD data of the surface design model, and the spatial deviation between the preliminary boundary line and the design theoretical boundary line is calculated. If the deviation is less than a preset accuracy threshold, the preliminary boundary line is retained; otherwise, the fitting parameters of the transition feature points are corrected based on the deviation value, and the corrected transition boundary line is regenerated. The modified transition boundary lines are extracted, including the starting coordinates, ending coordinates, length, curvature variation range, and angle with adjacent surfaces. Simultaneously, the area of ​​the convex-concave transition region corresponding to each boundary line is calculated to generate a basic parameter set for the transition boundary. Combining this basic parameter set with the previously generated curvature variation gradient matrix, the boundary region stability coefficient is calculated. Specifically, this coefficient is the ratio of the length to the area of ​​the corresponding convex-concave transition region, and the product of the curvature variation range and the average curvature variation gradient within the region. The boundary region stability coefficient is then fused with the basic parameter set to generate the region boundary feature parameters.

6. The method for processing curved surfaces of titanium alloy materials according to claim 3, characterized in that, Step S22 includes the following steps: Based on the rate of change of curvature radius corresponding to different processing areas and combined with the accuracy tolerance requirements parameters in the surface design model data, including shape tolerance, position tolerance and surface roughness tolerance, a curvature-tolerance correlation matrix is ​​constructed, in which each element corresponds to the matching relationship between the rate of change of curvature radius of a single processing area and the accuracy tolerance of that area, generating the accuracy requirement mapping parameters for the processing area. The complexity level of different processing areas is determined based on the contour complexity coefficient, where contour complexity coefficient ≤ 0.3 is low complexity, 0.3 < contour complexity coefficient ≤ 0.7 is medium complexity, and contour complexity coefficient > 0.7 is high complexity. At the same time, the feature contour line data in the surface design model, including spheres, parabolic surfaces and irregular free surfaces, are combined to generate processing area complexity-feature correlation parameters. The tool database of the five-axis linkage machining unit is called to obtain the material property parameters of the tool currently being machined, including hardness, wear resistance coefficient and thermal expansion coefficient, and structural parameters, including tool radius, cutting edge angle and number of cutting edges. Combined with the accuracy requirement mapping parameters of the machining area, the tool-accuracy adaptation coefficient is calculated. The complexity-feature correlation parameters of the machining area and the tool-precision adaptation coefficient are input into the initial calculation model of the spindle speed. With the goal of minimizing tool wear rate in low complexity area and optimizing machining stability in high complexity area, the initial values ​​of spindle speed in different machining areas are calculated. At the same time, the initial values ​​of the speed are corrected based on the rate of change of curvature radius, and real-time dynamic adjustment parameters of spindle speed are generated. Based on the real-time spindle speed dynamic adjustment parameters and the mapping parameters of the machining area accuracy requirements, the feed rate constraint threshold is calculated, specifically the maximum feed rate upper limit that meets the accuracy tolerance requirements. The feed rate adjustment weight is determined by combining the contour complexity coefficient, and feed rate constraint parameters are generated. The servo system performance parameters of the five-axis linkage machining unit are called, and combined with the feed rate constraint parameters, the machining trajectory deviation under different feed rates is simulated by the kinematic simulation algorithm. The feed rate range with trajectory deviation less than the accuracy tolerance requirements is selected, and feed rate dynamic adjustment parameters are generated.

7. The method for processing curved surfaces of titanium alloy materials according to claim 6, characterized in that, Step S23 includes the following steps: The ultrasonic vibration-assisted unit is used to perform time-domain and frequency-domain analysis on the initial cutting force data in the initial contact state data to extract the peak cutting force parameters, fluctuation frequency characteristics and duration distribution; The hardness parameters of titanium alloy and the material properties of cutting tool are obtained and a material-tool property matrix is ​​constructed. At the same time, the energy input parameters and amplitude threshold range required for ultrasonic vibration are calculated based on the peak cutting force parameters, fluctuation frequency characteristics and duration distribution and the material-tool property matrix. The initial amplitude parameters of ultrasonic vibration are calculated and generated based on the energy input parameters and amplitude threshold range, combined with the path interval distribution data in the preset processing trajectory planning data. Based on the matching analysis between the fluctuation frequency characteristics and duration distribution and the real-time spindle speed dynamic adjustment parameters, the initial adjustment range corresponding to ultrasonic vibration is determined. The initial frequency parameters of ultrasonic vibration are calculated by combining the surface curvature change parameters, and then integrated with the initial amplitude parameters of ultrasonic vibration to generate high-frequency vibration parameters.

8. The method for processing curved surfaces of titanium alloy materials according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: The real-time contour data of the surface processing area is continuously collected by the real-time contour monitoring unit, and the real-time contour data is registered and compared with the CAD data of the surface design model to calculate the deviation value between the actual position and the theoretical position of each sampling point, and generate a surface processing contour deviation distribution map. Step S32: Extract the maximum deviation value, average deviation value, and deviation concentration area parameters based on the surface machining profile deviation distribution map, and comprehensively calculate the surface machining profile deviation value; Step S33: Collect real-time temperature data in the curved surface machining area through the temperature sensor array corresponding to the cooling and lubrication unit and generate temperature field distribution data of the machining area; Step S34: Based on the temperature field distribution data of the processing area, analyze the location parameters, temperature gradient change rate and duration parameters of each temperature zone, and calculate the thermal deformation influence coefficient in combination with the thermal expansion coefficient parameter of the titanium alloy material. Step S35: Based on the surface machining contour deviation value and thermal deformation influence coefficient, perform dynamic compensation calculation on the sequence of collaborative control parameters for the machining process to generate adaptive machining parameters corresponding to the spindle speed correction, feed rate compensation value and ultrasonic vibration frequency adjustment.

9. The method for processing curved surfaces of titanium alloy materials according to claim 8, characterized in that, Step S35 includes the following steps: Step S351: By setting the surface machining profile deviation threshold and the thermal deformation influence coefficient threshold, the generated surface machining profile deviation value and thermal deformation influence coefficient are compared with the corresponding thresholds respectively. When the surface machining profile deviation value exceeds the deviation threshold or the thermal deformation influence coefficient exceeds the influence coefficient threshold, the machining parameter correction mechanism is activated. Step S352: Based on the machining parameter correction mechanism, perform vector decomposition on the surface machining contour deviation value to obtain the deviation component along the surface normal and the deviation component along the machining trajectory tangent direction; calculate the adjustment amount corresponding to the tool feed depth based on the deviation component along the surface normal, and combine the deviation component along the machining trajectory tangent direction to determine the trajectory correction parameters to comprehensively generate the contour deviation compensation amount. Step S353: Analyze the temperature distribution characteristics of the surface processing area according to the thermal deformation influence coefficient and determine the position parameters and temperature values ​​corresponding to the area with the highest temperature; based on the position parameters and temperature values ​​corresponding to the area with the highest temperature, combined with the flow parameters and injection angle parameters of the cooling and lubrication unit, calculate the cooling intensity parameters required for temperature reduction, and convert the cooling intensity parameters into the corresponding temperature compensation coefficients. Step S354: Based on the contour deviation compensation amount and temperature compensation coefficient, perform dynamic compensation calculation on the sequence of collaborative control parameters for the machining process to calculate the spindle speed correction amount, feed rate compensation value and ultrasonic vibration frequency adjustment amount, and generate adaptive machining parameters.

10. A system for machining curved surfaces of titanium alloy materials, characterized in that, The invention includes a computer-readable storage medium, a processor, a communication interface, and a computer program stored on the computer-readable storage medium and executable on the processor, characterized in that, when the processor executes the computer program, it implements the method for machining curved surfaces of titanium alloy materials as described in any one of claims 1-9.

Images