A composite material cross-domain milling ultrasonic amplitude self-adaptive cooperative control method

Abstract

This invention relates to the field of ceramic matrix composite material processing technology, and discloses an ultrasonic amplitude adaptive collaborative control method for cross-domain milling of composite materials. This method first constructs an integrated hardware and software collaborative control system comprising a machine tool, a force measurement platform, an ultrasonic generation and execution system, and a host computer computing center. An electrical signal-amplitude mapping is established through offline no-load calibration. A safe amplitude upper limit threshold table is constructed using depth of cut and contact angle as independent variables. During machining, the CNC code is read, and the instantaneous depth of cut and contact angle are analyzed based on Boolean interference. Characteristic transition zones are identified based on the rate of change of depth of cut and the rate of change of contact angle, achieving millisecond-level feedforward amplitude control. Simultaneously, cutting force signals are acquired in real time. When the actual force exceeds the threshold, a nonlinear exponential decay function is used to calculate a correction coefficient, smoothly reducing the amplitude command a second time and outputting it in a closed loop. This invention integrates pre-trajectory prediction and post-force feedback, effectively suppressing edge chipping in multi-feature cross-domain milling of ceramic matrix composite materials.

Description

Technical Field

[0001] This invention relates to the field of ceramic matrix composite material processing technology, specifically to an ultrasonic amplitude adaptive collaborative control method for cross-domain milling of composite materials. Background Technology

[0002] As aero-engines gradually upgrade towards higher thrust-to-weight ratios and higher turbine inlet temperatures, silicon carbide fiber-reinforced silicon carbide ceramic matrix composites, with their excellent properties of low density, high temperature resistance, and oxidation resistance, have become a core replacement material for key hot-end components such as turbine guide vanes and integral turbine disks in aero-engines. However, ceramic matrix composites possess the material characteristics of high hardness, high brittleness, strong anisotropy, and heterogeneity, making them extremely difficult to machine. Furthermore, the hot-end components of aero-engines have complex structures, with individual components integrating various features such as planar end walls, narrow deep grooves, and complex free-form surfaces. During the milling process across these features, the cutting conditions change dynamically and drastically, easily causing machining damage such as fiber pull-out, matrix cracking, interlaminar tearing, and edge chipping, severely restricting the yield rate of components and their mass application in engineering.

[0003] Currently, ultrasonic vibration-assisted milling is an effective means to improve the machining quality of ceramic matrix composites, but existing machining technologies have significant drawbacks. Traditional ultrasonic-assisted milling generally adopts a constant amplitude machining mode, which has a geometric machining blind zone. Under the condition of cross-domain variable cutting depth machining of planes, deep grooves, and curved surfaces, the machining energy and cutting conditions are prone to mismatch, resulting in either poor machining quality of planes or brittle fracture in complex feature areas. At the same time, existing adaptive ultrasonic machining technologies mostly adopt a post-feedback control mode of force measurement and acoustic emission sensors, which has an inherent physical response lag. It is a passive remedy after material damage occurs and cannot achieve preventive damage prevention. In addition, the pre-prediction control method relying solely on CNC toolpaths cannot adapt to the machining disturbances caused by uneven fiber weaving, local material inhomogeneity, and tool wear in ceramic matrix composites, resulting in poor system robustness.

[0004] Therefore, in view of the shortcomings of the prior art, the present invention provides an ultrasonic amplitude adaptive collaborative control method that combines "pre-trajectory prediction and post-state feedback". Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an ultrasonic amplitude adaptive collaborative control method for cross-domain milling of composite materials. This method has the advantages of accurate working condition prediction, no adjustment lag, strong processing stability, high component yield, wide engineering adaptability, and a balance between processing quality and efficiency. It solves the problems of constant amplitude processing energy mismatch, post-feedback adjustment lag, and weak anti-disturbance ability of pure trajectory prediction in the existing cross-domain milling process of ceramic matrix composite materials, which lead to component edge chipping, poor surface quality, low form and position accuracy, and insufficient batch processing stability.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution: an ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials, comprising the following steps:

[0009] Step 1: System Composition and Integration: Construct an integrated hardware and software collaborative control system that includes a machine tool, a force measurement platform, an ultrasonic generation and execution system, and a host computer computing center. The system consists of an execution and measurement unit, a data acquisition and calibration unit, and a computing and control module.

[0010] Step 2, Offline Testing and Critical Damage Threshold Library Construction: Establish the electrical signal-amplitude mapping relationship through no-load calibration, and build a safe amplitude upper limit threshold table with cutting depth and contact angle as independent variables based on the material experience library;

[0011] Step 3, Feedforward Analysis: Read the CNC code, calculate the instantaneous cutting depth and contact angle through Boolean interferometry, identify the characteristic transition zone based on the rate of change of cutting depth and the rate of change of contact angle, and realize millisecond-level feedforward amplitude control;

[0012] Step 4, Feedback Compensation: Real-time acquisition of cutting force signals; when the actual force exceeds the threshold, a nonlinear exponential decay function is used to calculate the correction coefficient, and the amplitude command is smoothly reduced twice and output in a closed loop.

[0013] Preferably, the execution and measurement unit includes an ultrasonic generator, an ultrasonic scalpel handle / spindle, a laser Doppler vibrometer, and a force meter.

[0014] Preferably, the data acquisition and calibration unit includes different processing areas such as the workpiece to be processed, planar milling / deep groove milling / curved surface milling, and a calibration unit for establishing a critical damage threshold table.

[0015] Preferably, the calculation and control module integrates a material and experience knowledge base, a critical damage threshold table, a Boolean interference calculation module, a variable shear depth gradient calculation module, a feature region judgment module, a feedback compensation module, and an amplitude command synthesizer.

[0016] Preferably, in step two, offline testing and critical damage threshold database construction are performed.

[0017] S2.1 No-load calibration and amplitude mapping: Before machining, the ultrasonic vibrating tool holder / spindle is calibrated under no-load using a laser Doppler vibrometer. The actual mechanical amplitude generated at the tool tip by the ultrasonic generator under different output electrical signals is measured to establish the relationship between the ultrasonic generator output electrical signal and the actual tool tip amplitude. The mapping relationship clarifies the adjustable range of the system amplitude. ;

[0018] S2.2, Critical Damage Threshold Database Construction: Combining relevant machining standards and actual cutting experience, a critical damage threshold table is established, including plane, deep groove, and curved surface features. This table is based on the cutting depth. Contact angle with the tool Using laser vibrometers and microscopic detection methods as independent variables, the upper limit threshold of safe ultrasonic amplitude corresponding to different ranges was determined. .

[0019] Preferably, the feedforward parsing in step three is as follows:

[0020] S3.1 Toolpath Data Reading and Geometric Analysis: Read CNC machining code and extract adjacent tool position points and tool axis vector data;

[0021] S3.2 Based on the geometric displacement relationship between adjacent tool positions, the interference physical equation between the tool sweep profile and the part model is constructed using the Boolean interference calculation module, and the instantaneous cutting depth at time t is calculated. and the tool-workpiece contact angle ;

[0022] S3.3 Determination of Variable Cut Depth Gradient and Feature Transition Zone: Based on the analysis results of S3.1-S3.2, the cut depth change rate is calculated using the variable cut depth gradient calculation module. and contact angle change rate .

[0023] Preferably, the variable cut depth gradient calculation module in S3.3 calculates:

[0024] (1) Rate of change of cutting depth :

[0025]

[0026] In the formula, Instantaneous cutting depth For time, when the rate of change of cutting depth When the set threshold is exceeded, it indicates that the tool is about to enter a sudden working condition such as a deep groove or steep slope;

[0027] (2) Rate of change of contact angle :

[0028]

[0029] In the formula, The tool-workpiece contact angle. For time, when the rate of change of contact angle A sharp increase indicates that the tool is transitioning from a planar region to a curved surface with high curvature or a narrow, deep groove.

[0030] Preferably, the feature transition region in S3.3 is determined by the feature region determination module:

[0031] (a) Planar feature region: when the rate of change of cutting depth and contact angle change rate When the temperature falls below a set stability threshold, it is identified as a planar feature region, and the system issues a near-term warning. The amplitude;

[0032] (b) Groove / Surface Feature Region: When the cutting depth changes... or contact angle change rate When the speed increases significantly beyond the set threshold, the system predicts that the tool is about to enter the deep groove cutting zone or the dangerous transition zone with large curvature of the curved surface, and simultaneously extracts the absolute value of the predicted target contact angle at the upcoming position. After performing secondary fine-tuning of features, the system immediately consults the critical damage threshold table and matches the corresponding upper limit of safe amplitude. And before the tool actually reaches that dangerous position In time, the amplitude attenuation command is issued in advance by the amplitude command synthesizer to constrain the amplitude to the safe amplitude upper limit. Inside.

[0033] Preferably, the feedback compensation process in step four is as follows:

[0034] S4.1 Real-time cutting force monitoring: During the machining process, dynamic cutting force signals are collected in real time using a force gauge. And feed it back to the feedback compensation module;

[0035] S4.2, Threshold-based Secondary Adjustment: After the predictive control in step three issues the amplitude command, if the actual cutting force still exceeds the system's set safety force threshold due to the influence of non-homogeneous hard points inside the CMC material or tool wear, the real-time feedback compensation module is triggered. A non-linear exponential decay function is used to calculate the feedback correction coefficient. The current amplitude command is adjusted and reduced a second time.

[0036] S4.3, Closed-loop output: The system will output the final calculated value. The amplitude command synthesizer converts the signal into an analog or digital signal, which is then used to control the dynamic output of the ultrasonic generator in real time, thus completing closed-loop fine-tuning.

[0037] Preferably, in step S4.2, the feedback correction coefficient is calculated. :

[0038]

[0039] In the formula, The dynamic cutting force is collected in real time by the force gauge. The system has a preset safety force threshold. The nonlinear attenuation coefficient is... It is a natural exponential function, when the actual cutting force Exceeding the safety threshold At that time, the current amplitude command is adjusted down a second time, resulting in:

[0040]

[0041] In the formula, The upper limit threshold of the safe amplitude is obtained by looking up the table in step three. For feedback correction coefficient, This refers to the actual amplitude that is ultimately output to the ultrasonic generator.

[0042] Compared with the prior art, the present invention provides an ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials, which has the following beneficial effects:

[0043] 1. This invention uses pre-judgment control of the depth of cut and contact angle changes. When the tool enters a harsh transition zone such as a deep groove, the system actively avoids the impact of high-frequency excess energy on the brittle boundary of CMC. This reduces the peak cutting force in the entry / exit area by about 40% compared to traditional constant amplitude machining. The edge contour of the S-shaped deep groove remains highly clear. Even under the most severe working conditions, the chipping size is still strictly controlled within 45μm, which greatly improves the pass rate of the system for extreme feature parts.

[0044] 2. The "partitioning for shape preservation and quality preservation" strategy implemented by the system of this invention can effectively avoid the efficiency loss caused by blindly reducing the amplitude. It fully releases the processing advantages of conventional large amplitude in the open planar area, so that the three-dimensional surface roughness Ra after processing is stably maintained in the high smoothness range of 0.6μm~1.5μm. SEM morphology confirms that the SiC fibers on the surface are cut evenly, with no obvious macroscopic fiber pull-out or matrix extension microcracks, and finally achieves efficient removal of the ductile domain.

[0045] 3. This invention is designed for wavy surfaces with continuously variable curvature characteristics. By identifying the contact angle in advance and transitioning with a moderate amplitude, it avoids the "tool deflection" phenomenon and machining chatter caused by a sudden increase in cutting force. It significantly reduces the milling profile accuracy error of the curved surface feature from the traditional ±0.3mm and stabilizes it within ±0.1mm. The final surface texture is smooth and free of vibration marks.

[0046] 4. This invention eliminates the reliance on finite element simulation of massive microscopic materials, enabling pre-planning through basic toolpath geometry analysis. Simultaneously, the superimposed high-frequency real-time cutting force closed-loop feedback function effectively shields the occasional excitation caused by the non-homogeneous characteristics inside CMC components. Based on mature force gauges and industrial bus designs such as EtherCAT, the system has a low computational burden and is easily upgraded and modified on existing five-axis high-end CNC machine tools, possessing strong practicality and promotional value. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating the offline testing and critical damage threshold library construction process of this invention.

[0048] Figure 2 This is a diagram illustrating the online feature trajectory analysis and pre-emptive control framework of the present invention.

[0049] Figure 3 This is a flowchart of the closed-loop compensation and execution process for post-event feedback in this invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Please see Figures 1-3 An adaptive and cooperative control method for ultrasonic amplitude in cross-domain milling of composite materials includes the following steps:

[0052] Step 1: System Composition and Integration: Construct an integrated hardware and software collaborative control system that includes a machine tool, a force measurement platform, an ultrasonic generation and execution system, and a host computer computing center. The system consists of an execution and measurement unit, a data acquisition and calibration unit, and a computing and control module.

[0053] Specifically, the execution and measurement unit includes an ultrasonic generator, an ultrasonic scalpel handle / spindle, a laser Doppler vibrometer, and a force gauge;

[0054] The ultrasonic generator receives amplitude commands from the amplitude command synthesizer and outputs corresponding high-frequency electrical signals to drive the ultrasonic tool holder / spindle to vibrate.

[0055] The ultrasonic tool holder / spindle converts the electrical signal output by the ultrasonic generator into high-frequency mechanical vibration at the tool tip, thereby achieving ultrasonic-assisted milling.

[0056] Laser Doppler vibration meters are used for offline no-load calibration of ultrasonic tool holders / spindles, establishing a precise mapping relationship between the output electrical signal of the ultrasonic generator and the actual amplitude at the tool tip, and clarifying the adjustable range of the system amplitude. ;

[0057] The force gauge is integrated into the machine tool worktable to collect dynamic cutting force signals in real time during the machining process. This provides raw data for subsequent feedback and compensation.

[0058] Specifically, the data acquisition and calibration unit includes different feature processing areas such as the workpiece to be processed, planar milling / deep groove milling / curved surface milling, as well as a calibration unit for establishing a critical damage threshold table;

[0059] The workpiece to be processed and the planar / deep groove / curved surface processing areas represent typical characteristic structures of hot-end components of aero-engines (open end walls, blade passages / tenons, blade profiles), and are the processing objects of the control method of this invention;

[0060] The calibration unit is used for preliminary basic cutting experiments. Combined with the material and experience knowledge base, it establishes a system based on depth of cut. Using a critical damage threshold table with the contact angle θ as the independent variable and the absence of macroscopic fragmentation as the criterion, the upper limit of safe amplitude under different working conditions is determined. .

[0061] Specifically, the computation and control module serves as the host computer's computation center, internally integrating a material and experience knowledge base, a critical damage threshold table, a Boolean interference calculation module, a variable shear depth gradient calculation module, a feature region judgment module, a feedback compensation module, and an amplitude command synthesizer.

[0062] The Materials & Experience Knowledge Base is responsible for storing the mechanical properties of CMC materials (such as fracture toughness KIC), previous process experimental data and processing experience, providing data support for threshold calibration and feature judgment.

[0063] The critical damage threshold table, as the core data table, can map ( , (to the corresponding safe amplitude limit) This provides a real-time table lookup function for feedforward control.

[0064] The Boolean interference calculation module constructs the interference relationship between the tool sweep profile and the part model based on adjacent tool position points and tool axis vectors, and calculates the instantaneous depth of cut. Tool-workpiece contact angle ;

[0065] The variable cut depth gradient calculation module and the feature region judgment module are responsible for calculating the cut depth change rate. and contact angle change rate It compares the current state with a set threshold to determine whether it is in a flat, stable region or a deep groove / curved surface transition region.

[0066] The feedback compensation module receives the real-time cutting force collected by the force measuring instrument. When the actual cutting force exceeds the safety force threshold, a nonlinear exponential decay function is used to calculate the feedback correction coefficient. The output is sent to the amplitude command synthesizer;

[0067] Amplitude command synthesizer integrates feedforward control (based on threshold table lookup) ) and feedback compensation (correction coefficient) The command calculates the final amplitude. The signal is converted into an analog or digital communication signal (such as EtherCAT) and sent to the ultrasonic generator.

[0068] This system architecture provides the hardware and data foundation for the subsequent three-level collaborative control of "offline database construction - trajectory feedforward - online feedback".

[0069] The advantages are: by building an integrated hardware and software collaborative architecture of offline calibration, trajectory calculation, real-time acquisition, and closed-loop control, the system systematically divides the execution measurement, data calibration, and calculation control units, which are clearly hierarchical and highly coupled; it can effectively get rid of the limitations of the single control unit in traditional ultrasonic processing, is compatible with existing five-axis machine tools and EtherCAT communication protocol, has strong external modification capabilities and high hardware reuse rate; at the same time, it provides a stable hardware carrier and data transmission foundation for the dual closed-loop collaborative control of feedforward prediction and post-event feedback, ultimately ensuring the synchronization of high-frequency calculation, signal acquisition, and amplitude dynamic adjustment.

[0070] Specifically, in step two, offline testing and the establishment of a critical damage threshold database are conducted.

[0071] S2.1 No-load calibration and amplitude mapping: Before machining, the ultrasonic vibrating tool holder / spindle is calibrated under no-load conditions using a laser Doppler vibrometer. The actual mechanical amplitude generated at the tool tip by the ultrasonic generator under different output electrical signals (voltage / power) is measured to establish the relationship between the ultrasonic generator output electrical signal (voltage / power) and the actual tool tip amplitude. The mapping relationship clarifies the adjustable range of the system amplitude. ;

[0072] S2.2, Critical Damage Threshold Database Construction: Combining relevant machining standards and actual cutting experience (based on the "Material & Experience Knowledge Base"), a critical damage threshold table is established, including plane, deep groove, and curved surface features. This table is based on the cutting depth. Contact angle with the tool Using the absence of macroscopic edge chipping or matrix cracking as the standard, the upper limit threshold of safe ultrasonic amplitude for different ranges was determined by laser vibrometer and microscopic detection methods. .

[0073] The advantages are: by using laser vibration measurement under no-load calibration, precise quantitative matching between ultrasonic electrical signals and the actual amplitude of the cutting tool is achieved, eliminating system transmission errors and ensuring amplitude control accuracy; by establishing a characteristic critical damage threshold library based on material mechanical properties and previous experimental data, using edge non-chipping and matrix non-cracking as quantitative indicators, differentiated safety boundaries are defined for planar / deep groove / curved surfaces; by establishing material damage critical criteria based on measured calibration data without complex finite element simulation, the threshold library can also be iteratively updated, thus enabling the system to adapt to different braided CMC composite materials, and its control basis is more in line with actual processing conditions.

[0074] Specifically, in step three, the feedforward parsing:

[0075] S3.1 Toolpath Data Reading and Geometric Analysis: Read CNC machining code (G code or APT file) and extract adjacent tool position points and tool axis vector data;

[0076] S3.2 Based on the geometric displacement relationship between adjacent tool positions, the interference physical equation between the tool sweep profile and the part model is constructed using the Boolean interference calculation module, and the instantaneous cutting depth at time t is calculated. and the tool-workpiece contact angle ;

[0077] S3.3 Determination of Variable Cut Depth Gradient and Feature Transition Zone: Based on the above results, the cut depth change rate is calculated using the variable cut depth gradient calculation module. and contact angle change rate .

[0078] Specifically, the variable tangent depth gradient calculation module in S3.3 calculates:

[0079] (1) Rate of change of cutting depth (This represents the rate of change of cutting depth over time, in mm / s):

[0080]

[0081] In the formula, This represents the instantaneous cutting depth, in mm. Time, in seconds. Used to quantify the drastic dynamic changes in the cutting layer thickness during the tool's entry into the workpiece, when When the set threshold is exceeded, it indicates that the tool is about to enter a sudden working condition such as a deep groove or steep slope. The ultrasonic amplitude needs to be attenuated in advance to prevent energy overload.

[0082] (2) Rate of change of contact angle (This represents the rate of change of the contact angle over time, expressed in rad / s or ° / s):

[0083]

[0084] In the formula, The tool-workpiece contact angle is expressed in radians (rad) or degrees (°). Time, in seconds. Used to characterize the dynamic evolution rate of the geometric meshing relationship between the tool and the workpiece. When the amplitude increases sharply, it indicates that the tool is transitioning from a planar area to a curved area with large curvature or a narrow deep groove. The instantaneous contact conditions change drastically. At this time, the ultrasonic amplitude must be reduced in advance to avoid brittle fracture caused by high-frequency impact.

[0085] Specifically, in S3.3, the feature transition region is determined by the feature region judgment module. The specific judgment process is as follows:

[0086] (a) Planar Feature Area (Smooth Cutting): When and When the amplitude is below the set stability threshold, it is determined to be a planar feature region, and the system sends out a larger normal amplitude (close to the threshold). By implementing a "planar quality preservation" strategy, the advantages of high-frequency "ironing and chip breaking" of ultrasound are fully utilized to obtain high surface quality;

[0087] (b) Deep groove / curved surface feature area (abrupt transition): when or When the speed increases significantly beyond the set threshold, the system predicts that the tool is about to enter the deep groove cutting zone or the dangerous transition zone with large curvature of the curved surface. At this time, the system simultaneously extracts the absolute value of the predicted target contact angle at the upcoming position. After performing secondary fine-tuning of features, the system immediately consults the critical damage threshold table and matches the corresponding upper limit of safe amplitude. And before the tool actually reaches that dangerous position The time (milliseconds) is controlled by issuing amplitude attenuation commands in advance through the amplitude command synthesizer, thus constraining the amplitude to a certain value. Internally, it achieves proactive and predictive control of "deep trench stress reduction and curved surface anti-collapse".

[0088] The advantages are: by relying on NC code toolpath geometry analysis + Boolean interference model, the instantaneous depth of cut and contact angle are accurately calculated, based on the rate of change of depth of cut. With contact angle change rate To dynamically predict indicators and achieve early identification of cross-domain characteristic transition zones; by adding a secondary discrimination mechanism for the target contact angle, misjudgment of a single parameter is avoided, and differentiated energy strategies for planar quality preservation, deep groove stress reduction, and curved surface anti-collapse are implemented in different zones; by adopting a millisecond-level advance pre-adjustment mode, the drawbacks of geometric blind spots and energy mismatch in traditional constant amplitude processing are completely solved, enabling the system to actively constrain ultrasonic impact energy before the arrival of harsh working conditions, and ultimately suppress fiber pull-out, matrix cracking, and edge chipping defects from the source.

[0089] Specifically, the feedback compensation process in step four:

[0090] S4.1 Real-time cutting force monitoring: During the machining process, dynamic cutting force signals are collected in real time using a force gauge. And feed it back to the feedback compensation module;

[0091] S4.2, Threshold-based secondary reduction: After the predictive control in step three issues the amplitude command, the actual cutting force ( ) is affected by the non-homogeneous hard points inside the CMC material or tool wear. If the force still exceeds the system's set safety threshold, the real-time feedback compensation module is triggered. This module uses a non-linear exponential decay function to calculate the feedback correction coefficient. The current amplitude command is smoothly and without a step decrease.

[0092] S4.3, Closed-loop output: The system will output the final calculated value. The amplitude command is converted into analog or digital communication signals (such as EtherCAT bus) by an amplitude command synthesizer, which controls the dynamic output of the ultrasonic generator in real time and completes closed-loop fine-tuning.

[0093] Specifically, in S4.2, the feedback correction coefficient is calculated. :

[0094]

[0095] In the formula, This refers to the dynamic cutting force collected in real time by the force gauge, expressed in Newton-meters (N). The system's preset safety force threshold, in N. This is the nonlinear attenuation coefficient, in N. -1 , is a positive real number. It is a natural exponential function, when the actual cutting force Exceeding the safety threshold At that time, the natural exponential function makes It decays continuously and without abrupt changes starting from 1, and the decay rate is... This control method simulates the nonlinear characteristics of material damage accumulation, thus avoiding the impact of sudden switching on / off changes on the machining system. Next, the current amplitude command is smoothly and without a step drop, resulting in a secondary reduction.

[0096]

[0097] In the formula, The upper limit threshold of the safe amplitude obtained from the table lookup in step three, in μm. This is the feedback correction coefficient (dimensionless). The final output amplitude to the ultrasonic generator, in μm, is the safe amplitude obtained through feedforward prediction. With feedback correction coefficient Multiplication achieves coordinated adjustment of "predictive benchmark + feedback compensation", due to It is an exponentially continuous function. As the cutting force decreases smoothly under overload, it avoids the mechanical impact caused by traditional binary (fully on / fully off) control, ensuring that the high-frequency vibration energy always converges within the safe range of the material's fracture toughness.

[0098] The advantages are: by using real-time online monitoring based on cutting force, it compensates for the shortcomings of pure trajectory feedforward in dealing with sudden disturbances such as material inhomogeneity, tool wear, and local hard spots; and by using a nonlinear exponential decay function to calculate the feedback correction coefficient. The amplitude is continuously and smoothly adjustable, without abrupt changes or switching shocks, protecting the ultrasonic tool holder and machine tool transmission structure; through The collaborative operation enables the system to achieve secondary precise convergence based on the feedforward reference amplitude, ensuring that the ultrasonic energy is always locked within the safe range of fracture toughness of CMC material. The entire four-step process forms a closed loop of proactive prevention and dynamic correction, which ultimately greatly improves the stability and yield of the system in the batch processing of complex components.

[0099] Through the closed-loop architecture of Step 1 (system integration) → Step 2 (offline library construction) → Step 3 (feedforward prediction) → Step 4 (feedback compensation), high-frequency coordination from microscopic material mechanical response to macroscopic control electrical signal is realized.

[0100] The system of the present invention is applied to the embodiments as follows:

[0101] Example

[0102] Using Cf / SiC ceramic matrix composites prepared by the CVI process as the processing object, the goal is to machine planar / S-groove / wavy curved surface features with different dimensional parameters on their surface using an ultrasonic vibration-assisted milling method. The specific implementation steps using the system of this invention are as follows:

[0103] S1. Sample preparation: The SiCf / SiC ceramic matrix composite material is pre-made into a cuboid sample with dimensions of 40mm×40mm×30mm by wire cutting process, and the upper surface is precision ground and polished.

[0104] S2. Machining System Setup: The sample is fixed to the KISTLER 9119AA1 piezoelectric force gauge using a special precision fixture, and the force gauge is installed on the worktable of a three-axis CNC machine tool. An ultrasonic vibration device is integrated at the spindle end of the machine tool to build a complete closed-loop machining system of "machine tool-ultrasonic tool holder-tool-workpiece-force gauge-ultrasonic controller".

[0105] S3. Establishing the mapping relationship: Before machining, the milling tool is calibrated under no-load using an MVF-10 fiber laser Doppler vibrometer, and the actual amplitude A of the tool tip is measured by frequency sweep. This establishes the relationship between the output electrical signal of the ultrasonic generator and the amplitude at the tool tip, thereby clarifying the adjustable range of the system amplitude. Based on milling machining standards and actual cutting experience, a critical damage threshold table was established to determine the upper limit of safe ultrasonic amplitude for different ranges. ;

[0106] S4. Trajectory Analysis and Pre-control: The host computer's feature trajectory analysis module pre-analyzes the G-code of the model to be processed and extracts adjacent tool position data; based on the geometric displacement relationship of adjacent tool positions, it constructs a Boolean interference calculation formula for the tool sweep profile and the part model; and calculates the instantaneous cutting depth at time t. and the tool contact angle ;

[0107] S5. Milling Feature Machining: During the actual machining process, calculate the depth of cut variation rate based on the results obtained in S4. and contact angle change rate To determine the feature regions that are about to be processed:

[0108] like ∧ < The machining operation is determined to be planar milling. A face milling cutter (D6×R0.2×2.5×45L×D6-MF) is used, and the machining parameters are set as follows: spindle speed 15000 r / min, feed rate 50 mm / min, depth of cut 0.04 mm, ultrasonic frequency 40 kHz. According to S3, the adjustable range of the tool amplitude is […]. The system sends out a larger conventional amplitude, i.e. To achieve "flat surface quality preservation";

[0109] like ∧ ≥ The system anticipates that the cutting tool is about to enter slot milling or surface milling; at this moment, the system simultaneously extracts the absolute value of the predicted target contact angle that is about to reach that position. And set a full-load critical angle. =150° for secondary feature discrimination:

[0110] like The machining operation is determined to be slot milling. A slot milling cutter (D6×R0.2-2.5H-D6) is used, and the machining parameters are set as follows: spindle speed 16000 r / min, feed rate 50 mm / min, depth of cut 0.05 mm, ultrasonic frequency 40 kHz. According to S3, the adjustment range of the tool amplitude is […]. The system issues an anti-impact safety amplitude signal in advance before the cutting tool arrives. To suppress the rapidly increasing high-frequency cutting load and achieve "deep groove stress reduction and anti-collapse";

[0111] like The machining was determined to be surface milling; a graphite ball end mill (D2R1) was used, and the machining parameters were set as follows: spindle speed 14000 r / min, feed rate 40 mm / min, depth of cut 0.05 mm, ultrasonic frequency 40 kHz. Based on S3, the adjustable range of the tool amplitude is […]. The system sends out medium amplitude signals in advance before the tool arrives. To balance local anti-chipping and surface forming quality, achieving "curved surface shape preservation and anti-chipping";

[0112] S6. Milling Process Compensation: In S5, a force gauge mounted on the machine tool worktable is used to collect three-dimensional cutting force signals in real time at a sampling frequency of 1kHz, and the composite cutting force signal is extracted. The high-frequency machine tool vibration and electromagnetic noise interference are eliminated through a low-pass filter to obtain the effective average cutting force. The effective actual cutting force collected The safety force threshold set with respect to the current feature interval Perform real-time comparison:

[0113] like If not, compensation will not be triggered, and the correction coefficient will be adjusted. ;

[0114] like This triggers the real-time feedback compensation module, which then performs a secondary reduction on the current amplitude command. The nonlinear exponential decay function is used to calculate the feedback correction coefficient. The system will update the final amplitude. The signal is transmitted via EtherCAT bus to the ultrasonic-assisted system, which adjusts the output power in real time to ensure a smooth and instantaneous drop in the tool tip amplitude until the actual cutting force is reached. Falling back to The following forms a stable closed-loop adaptive control;

[0115] S7. Milling completed: Repeat steps S5 and S6 until the preset plane, S-groove, and wavy surface are all milled and formed, then all machining steps are completed and machining stops.

[0116] Comparative Example (Control Group)

[0117] The comparative example used the same Cf / SiC ceramic matrix composite sample, the same machine tool, the same ultrasonic vibration-assisted milling system, and the same planar / S-groove / wavy surface machining trajectory as the example. The difference was that the comparative example used a constant ultrasonic amplitude mode throughout the entire process, without any feedforward prediction based on trajectory analysis or feedback compensation based on cutting force. Specifically, the comparative example used a uniform amplitude of 4μm for the planar region, deep groove region, and curved surface region. Other machining parameters (spindle speed, feed rate, depth of cut, ultrasonic frequency, etc.) remained consistent with the example. After machining, the same measurement methods (tool scanner, blue light scanner, SEM, MATLAB reconstruction) were used to record data such as chipped edge size, surface roughness, and surface contour accuracy, which served as a benchmark for comparison with the system of this invention.

[0118] Analysis revealed the following results compared to the control group (using constant ultrasonic amplitude assisted milling process): macroscopic dimensional measurements of the machined specimens were performed, combined with microscopic morphology observation using a tool scanner, blue light scanner, and scanning electron microscope (SEM), and the surface three-dimensional point cloud data was reconstructed using MATLAB.

[0119] (1) Compared with the control group, this system fully utilizes the "ironing" and "chip breaking" effects of high-frequency ultrasound in the planar region. Three-dimensional morphology measurement shows that the surface roughness Ra of the planar region is stable between 0.6μm and 1.5μm, and the surface smoothness is extremely high. Its SEM image shows that the SiC fiber is cut evenly, and there are no obvious peeling and extensional microcracks in the matrix, thus realizing the removal of the ductile domain of the material.

[0120] (2) Unlike the control group, which had severe macroscopic chipping at the groove edge, the cutting peak force at the moment of cutting in / out was reduced by more than 40% after using the system of the present invention. After observation, the edge contour of the S-shaped deep groove was clear and the chipping defect was strictly controlled within 45μm.

[0121] (3) The results of the contour measuring instrument show that the surface milling accuracy error has been reduced from ±0.3mm in the control group to within ±0.1mm, and the surface texture transition is smooth.

[0122] In summary, the ultrasonic amplitude adaptive collaborative control method combining pre-trajectory prediction and post-state feedback provided by this invention identifies cross-domain characteristic transition zones (planes, deep grooves, curved surfaces) in real time by analyzing CNC toolpath geometric data. It can pre-attenuate the amplitude to within the material fracture toughness safety threshold and superimpose nonlinear exponential closed-loop fine adjustment based on cutting force. This fundamentally solves the edge chipping problem caused by constant amplitude energy mismatch and single feedback lag, and achieves efficient, low-damage, and high-yield machining of complex CMC components.

[0123] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive and coordinated ultrasonic amplitude control in cross-domain milling of composite materials, characterized in that, Includes the following steps: Step 1: System Composition and Integration: Construct an integrated hardware and software collaborative control system that includes a machine tool, a force measurement platform, an ultrasonic generation and execution system, and a host computer computing center. The system consists of an execution and measurement unit, a data acquisition and calibration unit, and a computing and control module. Step 2, Offline Testing and Critical Damage Threshold Library Construction: Establish the electrical signal-amplitude mapping relationship through no-load calibration, and build a safe amplitude upper limit threshold table with cutting depth and contact angle as independent variables based on the material experience library; Step 3, Feedforward Analysis: Read the CNC code, calculate the instantaneous cutting depth and contact angle through Boolean interferometry, identify the characteristic transition zone based on the rate of change of cutting depth and the rate of change of contact angle, and realize millisecond-level feedforward amplitude control; Step 4, Feedback Compensation: Real-time acquisition of cutting force signals; when the actual force exceeds the threshold, a nonlinear exponential decay function is used to calculate the correction coefficient, and the amplitude command is smoothly reduced twice and output in a closed loop.

2. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, The execution and measurement unit includes an ultrasonic generator, an ultrasonic tool holder / spindle, a laser Doppler vibrometer, and a force meter.

3. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, The data acquisition and calibration unit includes different processing areas such as the workpiece to be processed, planar milling / deep groove milling / curved surface milling, as well as a calibration unit for establishing a critical damage threshold table.

4. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, The computation and control module integrates a material and experience knowledge base, a critical damage threshold table, a Boolean interference calculation module, a variable shear depth gradient calculation module, a feature region judgment module, a feedback compensation module, and an amplitude command synthesizer.

5. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, In step two, offline testing and critical damage threshold database construction are performed. S2.1 No-load calibration and amplitude mapping: Before machining, the ultrasonic vibrating tool holder / spindle is calibrated under no-load using a laser Doppler vibrometer. The actual mechanical amplitude generated at the tool tip by the ultrasonic generator under different output electrical signals is measured to establish the relationship between the ultrasonic generator output electrical signal and the actual tool tip amplitude. The mapping relationship clarifies the adjustable range of the system amplitude. ; S2.2, Critical Damage Threshold Database Construction: Combining relevant machining standards and actual cutting experience, a critical damage threshold table is established, including plane, deep groove, and curved surface features. This table is based on the cutting depth. Contact angle with the tool Using laser vibrometers and microscopic detection methods as independent variables, the upper limit threshold of safe ultrasonic amplitude corresponding to different ranges was determined. .

6. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, The feedforward parsing in step three: S3.1 Toolpath Data Reading and Geometric Analysis: Read CNC machining code and extract adjacent tool position points and tool axis vector data; S3.2 Based on the geometric displacement relationship between adjacent tool positions, the interference physical equation between the tool sweep profile and the part model is constructed using the Boolean interference calculation module, and the instantaneous cutting depth at time t is calculated. and the tool-workpiece contact angle ; S3.3 Determination of Variable Cut Depth Gradient and Feature Transition Zone: Based on the analysis results of S3.1-S3.2, the cut depth change rate is calculated using the variable cut depth gradient calculation module. and contact angle change rate .

7. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 6, characterized in that, The variable cut depth gradient calculation module in S3.3 calculates: (1) Rate of change of cutting depth : ； In the formula, Instantaneous cutting depth For time, when the rate of change of cutting depth When the set threshold is exceeded, it indicates that the tool is about to enter a sudden working condition of deep groove or steep slope; (2) Rate of change of contact angle : ； In the formula, The tool-workpiece contact angle. For time, when the rate of change of contact angle A sharp increase indicates that the tool is transitioning from a planar region to a curved surface with high curvature or a narrow, deep groove.

8. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 6, characterized in that, The feature transition region in S3.3 is determined by the feature region judgment module: (a) Planar feature region: when the rate of change of cutting depth and contact angle change rate When the temperature drops below a set stability threshold, it is identified as a planar feature region, and the system issues a near-term warning. The amplitude; (b) Groove / Surface Feature Region: When the cutting depth changes... or contact angle change rate When the speed increases significantly beyond the set threshold, the system predicts that the tool is about to enter the deep groove cutting zone or the dangerous transition zone with large curvature of the curved surface, and simultaneously extracts the absolute value of the predicted target contact angle at the upcoming position. After performing secondary fine-tuning of features, the system immediately consults the critical damage threshold table and matches the corresponding upper limit of safe amplitude. And before the tool actually reaches that dangerous position In time, the amplitude attenuation command is issued in advance by the amplitude command synthesizer to constrain the amplitude to the safe amplitude upper limit. Inside.

9. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 1, characterized in that, The feedback compensation process in step four: S4.1 Real-time cutting force monitoring: During the machining process, dynamic cutting force signals are collected in real time using a force gauge. And feed it back to the feedback compensation module; S4.2, Threshold-based Secondary Adjustment: After the predictive control in step three issues the amplitude command, if the actual cutting force still exceeds the system's set safety force threshold due to the influence of non-homogeneous hard points inside the CMC material or tool wear, the real-time feedback compensation module is triggered. A non-linear exponential decay function is used to calculate the feedback correction coefficient. The current amplitude command is adjusted and reduced a second time. S4.3, Closed-loop output: The system will output the final calculated value. The amplitude command synthesizer converts the signal into an analog or digital signal, which is then used to control the dynamic output of the ultrasonic generator in real time, thus completing closed-loop fine-tuning.

10. The ultrasonic amplitude adaptive cooperative control method for cross-domain milling of composite materials according to claim 9, characterized in that, The feedback correction coefficient is calculated in S4.

2. : ； In the formula, The dynamic cutting force is collected in real time by the force gauge. The system has a preset safety force threshold. The nonlinear attenuation coefficient is... It is a natural exponential function, when the actual cutting force Exceeding the safety threshold At that time, the current amplitude command is adjusted down a second time, resulting in: ； In the formula, The upper limit threshold of the safe amplitude is obtained by looking up the table in step three. For feedback correction coefficient, This refers to the actual amplitude that is ultimately output to the ultrasonic generator.

Images