A rapid positioning and clamping system for die frame machining and a control method thereof

Abstract

The application discloses a rapid positioning and clamping system for mold frame machining and a control method thereof. The system comprises a composite clamping unit array containing a first action element and a second action element, a self-powered sensing module, a digital twin module and a control module. The method comprises: the digital twin module executes clamping and machining joint simulation to output control parameters; the first driving signal drives the first action element to rapidly clamp and keep power off; during machining, the second feedback signal and the predicted value are used for deviation grading compensation; when the deviation is small, the second action element is used for high-frequency suppression; when the deviation is large, the first and second action elements are used for low-frequency adjustment; static feedback is periodically collected to calculate evaluation values, and when the evaluation values exceed the standard, the evaluation values are gradually fine-tuned; and after machining, digital archives are generated to feed back and calibrate model parameters. The application realizes zero-energy consumption clamping, active vibration suppression, clamping-machining collaborative optimization, full-band dynamic stability control, self-powered sensing, online monitoring and self-learning, and is suitable for clamping and machining of various mold frames.

Description

Technical Field

[0001] This invention belongs to the field of mold frame machining technology, and particularly relates to a rapid positioning and clamping system and its control method for mold frame machining. Background Technology

[0002] The mold base is a crucial component of the mold, and its machining accuracy directly affects the assembly quality and service life of the mold. During milling, drilling, and boring processes of the mold base, the performance of the clamping system is one of the key factors determining machining accuracy and efficiency. Currently, commonly used clamping methods in mold base machining mainly include mechanical clamps, hydraulic clamps, pneumatic clamps, and electromagnetic clamps. Mechanical clamps rely on threaded pairs or eccentric mechanisms to generate clamping force; they are simple in structure but have low clamping efficiency. Hydraulic and pneumatic clamps drive clamping elements through fluid pressure, achieving larger clamping force outputs and are suitable for rough machining of large mold bases. Electromagnetic clamps fix workpieces using electromagnetic attraction, offering fast clamping speeds without mechanical deformation. Furthermore, zero-point positioning systems and automatic changing systems are used in some high-end machining centers, enabling rapid positioning and replacement of the mold base through standardized interfaces. With the development of digital manufacturing technology, some clamping systems are beginning to integrate force and displacement sensors for offline detection of the clamping status. Digital twin technology has been applied in fields such as machining path optimization and tool wear prediction, and finite element simulation methods have also been used to analyze the deformation behavior of workpieces under cutting forces.

[0003] However, the existing clamping systems still have the following shortcomings: First, hydraulic, pneumatic, and electromagnetic clamps all require continuous power to maintain the clamping state, resulting in high energy consumption. The heat generated by continuous power supply can also cause thermal deformation errors in the mold frame. Furthermore, the clamping force is easily affected by oil temperature fluctuations or electromagnetic interference during machining. Second, existing clamps are all passive clamping devices; once clamped, they are fixed and cannot detect and actively compensate for micro-vibrations and deformations of the mold frame in real time during machining, leading to uncontrolled machining errors caused by cutting vibrations and thermal deformation. Third, the design of existing clamping schemes is independent of machining path planning. The selection of clamping points and the setting of clamping forces usually rely on manual experience, lacking a systematic consideration of the dynamic cutting force distribution during machining, making it difficult to achieve an integrated optimal configuration of clamping and machining. Fourth, a single type of drive element cannot simultaneously meet the needs of large-scale low-frequency adjustments and small-scale high-frequency suppression, failing to achieve dynamic stability control across the entire frequency band. Fifth, existing sensing systems rely on external power supply and wired transmission, resulting in complex wiring and limiting the free movement of the worktable. Sixth, abnormal states such as clamping force attenuation and micro-slippage of the mold base cannot be monitored and actively intervened in during the processing. The clamping process lacks data recording and traceability capabilities, and cannot provide data support for optimizing clamping schemes for subsequent mold bases of the same model. Seventh, the existing system lacks self-learning capabilities and cannot use historical data to calibrate and optimize the simulation model. Summary of the Invention

[0004] To overcome the above-mentioned defects of the prior art, the present invention provides a rapid positioning and clamping system for mold frame processing and its control method, which solves the problems of energy consumption and thermal deformation caused by continuous power supply, passive clamping without active compensation, clamping scheme and processing technology being disconnected, single drive element being unable to take into account the stability of the entire frequency band, lack of self-powered sensing and online monitoring, and lack of self-learning ability in the prior art mold frame processing clamping system.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A rapid positioning and clamping system for mold base machining and its control method, comprising: A composite clamping unit array, wherein each composite clamping unit in the composite clamping unit array includes a first actuating element and a second actuating element; the first actuating element and the second actuating element are connected in series via a pre-tightening screw, the driving force generated by the first actuating element is transmitted to the workpiece via the second actuating element, and the second actuating element is in a pre-compressed state when clamped. The self-powered sensing module is connected to each of the composite clamping units and is used to collect the status feedback signals of each of the composite clamping units, and convert a portion of the status feedback signals into electrical energy to power the self-powered sensing module. The digital twin module is used to perform joint simulation of clamping status and machining process based on the three-dimensional model of the workpiece and machining process parameters, and output the control parameter set of each composite clamping unit; The control module is connected to the composite clamping unit array, the self-powered sensing module and the digital twin module respectively, and is used to control the output state of each composite clamping unit according to the control parameter set and the state feedback signal. The first actuating element has a power-off self-locking characteristic, which is used to maintain the clamping force without consuming external energy during the processing stage, so that the second actuating element can focus on dynamic response during the processing stage; the second actuating element is the same physical element, which alternately or simultaneously undertakes the sensing function, execution function and energy harvesting function in time or working conditions.

[0006] Preferably, the first actuating element is used to generate a first displacement and output a first feedback signal in response to a first driving signal, and the first actuating element maintains the first displacement after the first driving signal is withdrawn; the second actuating element is used to generate a second displacement in response to a second driving signal and output a second feedback signal in response to external stress; the state feedback signal includes the first feedback signal and the second feedback signal; the sensing function, execution function and energy harvesting function of the second actuating element share the same piezoelectric material body, and the same charge signal is distinguished and processed in the time domain by a signal processing circuit: the high-frequency component is used to sense vibration stress, the low-frequency component is used to execute displacement output, and the charge accumulation part is used for energy harvesting.

[0007] Preferably, the first actuating element is a shape memory alloy driving element, the first driving signal is a current pulse signal, and the first feedback signal is a resistance value signal; the second actuating element is a piezoelectric driving element, the second driving signal is a voltage signal, and the second feedback signal is a charge signal; the self-powered sensing module converts the charge signal into electrical energy; the shape memory alloy driving element is a helical spring structure, the piezoelectric driving element is a stacked structure, the axial centerline of the shape memory alloy driving element coincides with the axial centerline of the piezoelectric driving element, and the two are connected in series along the same axis through the preload screw to form a single force transmission chain.

[0008] Preferably, the self-powered sensing module includes: A signal acquisition unit, connected to each of the composite clamping units, is used to acquire the status feedback signal; An energy conversion unit, connected to each of the composite clamping units, is used to convert a portion of the state feedback signal into electrical energy and store it. A power supply output unit, wherein the power input terminal of the power supply output unit is connected to the power output terminal of the energy conversion unit, and the power output terminal of the power supply output unit is connected to the power supply terminal of the signal acquisition unit.

[0009] Preferably, the digital twin module includes: Modeling unit, used to create a digital model of the workpiece; The clamping simulation unit is used to generate multiple candidate clamping schemes based on the digital model, perform clamping deformation simulation on each candidate clamping scheme, and calculate the clamping deformation amount corresponding to each scheme. The machining simulation unit is used to simultaneously perform machining process simulation on each of the candidate clamping schemes, and calculate the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit is used to select the optimal solution from the multiple candidate clamping solutions with the clamping deformation, the dynamic displacement and the vibration amplitude as optimization targets, and output the set of control parameters of each composite clamping unit corresponding to the optimal solution.

[0010] Preferably, a rapid positioning and clamping control method for mold base processing includes the following steps: S1: Import the 3D model of the workpiece and the machining process parameters, generate multiple candidate clamping schemes, perform clamping deformation simulation and machining process simulation on each candidate clamping scheme, select the optimal scheme with clamping deformation amount, machining process dynamic displacement amount and vibration amplitude as optimization targets, and output the control parameter set of each composite clamping unit; S2: According to the target output value in the set of control parameters, output a first drive signal to each composite clamping unit. The first actuating element of each composite clamping unit responds to the first drive signal to generate a first displacement and apply a clamping force. During the output of the first drive signal, the first feedback signal of each first actuating element is collected. The measured first feedback signal is compared with the target value. When the deviation is within the preset range, the first drive signal is withdrawn. S3: During the machining process, the second feedback signal of the second action element of each composite clamping unit in response to external stress is collected, and the second feedback signal is compared with the predicted value output by the digital twin module. The stroke range of the second action element is at the micrometer level, and it can suppress high-frequency small-amplitude vibration when used alone, but it cannot compensate for quasi-static large-amplitude displacement caused by thermal deformation and cutting force changes. The response time of the first action element is at the hundred-millisecond level, and it can achieve large-amplitude displacement adjustment when used alone, but its response frequency is below 100Hz and it cannot suppress high-frequency vibration. Therefore, a graded compensation strategy is adopted: when the deviation is less than or equal to the first threshold, the second action element is fed a signal. A second drive signal is output, and the second actuator generates a second displacement in response to the second drive signal; when the deviation is greater than a first threshold and less than or equal to a second threshold, a third drive signal is output to the first actuator, and the first actuator generates a third displacement in response to the third drive signal, while simultaneously outputting the second drive signal to the second actuator; the cancellation of the first drive signal causes the first actuator to enter a power-off self-locking state, during which the first actuator does not consume external energy, thereby enabling the charge signal output by the second actuator to be collected by the self-powered sensing module and converted into electrical energy, realizing self-powered closed-loop control; S4: During the processing, static first feedback signal and static second feedback signal of each composite clamping unit are collected at a preset cycle, and clamping status evaluation value is calculated; when the evaluation value deviates from the preset range, a fourth drive signal is output to the second action element for fine adjustment; if the evaluation value still deviates from the preset range after fine adjustment, a fifth drive signal is output to the first action element for fine adjustment. S5: After processing, record the feedback signal change data of each composite clamping unit and the data of each compensation action to generate a digital file; input the digital file into the digital twin module to update the simulation model parameters.

[0011] Preferably, in step S1, the number of candidate clamping schemes is no less than 1000; the clamping deformation simulation uses the finite element method to calculate the elastic deformation of the workpiece under clamping force; the machining process simulation uses the finite element method to simulate the dynamic displacement and vibration amplitude of the workpiece in rough milling, finish milling, drilling, and boring processes; the optimization objectives are to minimize elastic deformation, dynamic displacement, and vibration amplitude, and the optimal scheme is selected according to preset weights after outputting the Pareto optimal solution set using a multi-objective optimization algorithm; compared with the manual experience clamping scheme without digital twin optimization, the control parameter set output by the digital twin module reduces the clamping deformation by more than 60% and the dynamic displacement by more than 40%.

[0012] Preferably, in step S3, the sampling frequency of each second feedback signal is not less than 10kHz. After acquisition, each second feedback signal is decomposed into high-frequency components and low-frequency components. The high-frequency components correspond to cutting vibration stress, and the low-frequency components correspond to cutting force variation stress and thermal deformation stress. The high-frequency components and the low-frequency components are compared with the predicted values ​​output by the digital twin module, and high-frequency deviation values ​​and low-frequency deviation values ​​are calculated respectively. When the high-frequency deviation value is greater than zero, the second driving signal is output to the corresponding second action element. The amplitude of the second driving signal is linearly proportional to the high-frequency deviation value, and the proportionality coefficient is 0.5 to 2.0. When the low-frequency deviation value is greater than the first threshold and less than or equal to the second threshold, the third driving signal is output to the corresponding first action element. The amplitude of the third driving signal is less than the amplitude of the first driving signal.

[0013] Preferably, in step S3, the first threshold is 5% to 15% of the predicted value output by the digital twin module, the second threshold is 25% to 35% of the predicted value output by the digital twin module, and the first threshold is less than the second threshold; when the deviation is greater than the second threshold, an alarm signal is issued and processing is suspended. In step S4, the preset period is 1 minute to 10 minutes; the static first feedback signal is the measured value of the feedback signal of the first actuating element in the state without a driving signal; the static second feedback signal is the measured value of the feedback signal of the second actuating element in the state without dynamic external stress; the clamping state evaluation value is: calculating the first rate of change of the current static first feedback signal relative to the initial static first feedback signal reference value, the second rate of change of the current static second feedback signal relative to the initial static second feedback signal reference value, and the third rate of change of the current temperature value relative to the initial temperature reference value, respectively, and using the weighted sum of the first rate of change, the second rate of change, and the third rate of change as the evaluation value, with the sum of the three weight coefficients being 1, and the value range of each weight coefficient being 0.1 to 0.7.

[0014] Preferably, in step S5, the digital archive includes: an initial set of control parameters, historical curves of feedback signal changes for each composite clamping unit, timestamps and amplitude records of each compensation action, and historical deviations between the predicted and measured values ​​of the digital twin module; the digital archive is stored in a historical database; the digital twin module reads the digital archive in the historical database at a preset period, and calibrates the simulation model parameters using the measured deviation values. The calibration method is as follows: using the measured deviation values ​​as the dependent variable and the simulation model parameters as the independent variable, the parameter correction amount that minimizes the sum of squared deviations is solved, and the parameter correction amount is added to the current model parameters; after the digital twin module reads the historical database and calibrates the model parameters at a preset period, the prediction error decreases with the increase of the number of processing operations: the average prediction error for the first 10 processing operations is 12%, the average prediction error for the 11th to 50th processing operations decreases to 6%, and the average prediction error for the 51st and subsequent operations decreases to below 3%.

[0015] It is important to emphasize that the digital twin module, SMA self-locking characteristic, and piezoelectric self-powered system of this invention are not simply the sum of independent functions, but rather an organic whole that supports each other. It is precisely because the SMA has a power-off self-locking characteristic, eliminating the need for external energy consumption during processing, that the charge signal output by the piezoelectric stack during processing can be collected by the self-powered sensing module and converted into electrical energy, rather than being overwhelmed by electromagnetic interference from the external power supply system. It is precisely because the piezoelectric stack has a self-powered characteristic that each composite clamping unit can independently collect signals and wirelessly feed them back to the control module, eliminating the need for external wiring and battery replacement. It is precisely because the digital twin module provides predicted values ​​that the piezoelectric feedback signal has a comparison benchmark, thereby achieving graded compensation. It is precisely because the digital twin module possesses self-learning capabilities that the system can continuously calibrate model parameters using historical digital archives, improving prediction accuracy with increased usage. These three components are mutually dependent and indispensable, together constituting the complete technical solution of this invention.

[0016] The technical effects and advantages of the rapid positioning and clamping system and its control method for mold frame processing of the present invention are as follows: 1. In this invention, after the first actuating element undergoes phase change and contraction to generate clamping force upon heating and energization, the deformation and stress after the phase change can still be maintained even after the first driving signal is removed (power off), thus allowing the clamping force to be maintained for a long time. Compared with traditional hydraulic and electromagnetic clamps that require continuous power supply to maintain the clamping state, this invention only consumes electrical energy during the clamping stage, and no external energy supply is required during the processing stage. This eliminates the system heating problem caused by continuous power supply and reduces the energy consumption for long-term processing of large mold frames.

[0017] 2. In this invention, the second actuating element responds in real time to external stress during processing, outputting a second feedback signal. The control module outputs a second driving signal to the second actuating element based on this feedback signal, causing it to generate a reverse displacement to actively counteract vibration. This mechanism achieves integrated "sensing-compensation" active vibration suppression, which can reduce the impact of cutting vibration on machining accuracy from the root, rather than relying solely on passive clamping to resist vibration, effectively improving the surface quality and dimensional accuracy of the mold base machining.

[0018] 3. In this invention, the digital twin module generates multiple candidate clamping schemes based on the workpiece's 3D model and machining process parameters. Clamping deformation simulation and machining process simulation are performed on each scheme separately, and the optimal scheme is selected based on the clamping deformation amount, dynamic displacement amount, and vibration amplitude as optimization objectives. This mechanism integrates clamping scheme design and machining process planning as a whole system for collaborative optimization, enabling the clamping scheme to proactively adapt to the dynamic load distribution during machining. This breaks through the traditional method's model where clamping design and machining path are independent, reducing clamping deformation and machining vibration from the source.

[0019] 4. In this invention, the response time of the first actuating element to the third driving signal to generate the third displacement is on the order of hundreds of milliseconds, used for low-frequency large-amplitude adjustment (compensating for quasi-static displacement caused by thermal deformation and changes in cutting force); the response frequency of the second actuating element to the second driving signal is on the order of kHz, used for high-frequency small-amplitude suppression (counteracting cutting vibration). The two elements complement each other in terms of response frequency and compensation amplitude, achieving full-frequency coverage from quasi-static to high-frequency vibration, and comprehensively improving the system stability during the machining process.

[0020] 5. In this invention, the self-powered sensing module converts the second feedback signal (charge signal) output by the second actuating element in response to external stress into electrical energy to power the signal acquisition unit. This mechanism utilizes the vibration and stress that inevitably exist during the processing as an energy source, realizing the self-powered operation of the sensing network without the need for external wiring and battery replacement, thus reducing system deployment and maintenance costs.

[0021] 6. In this invention, the control module collects static first feedback signals and static second feedback signals at preset intervals during processing and calculates the clamping status evaluation value. When the evaluation value deviates from the preset range, it automatically triggers fine-tuning. This mechanism realizes online monitoring and active control of the clamping status, and can intervene before abnormal states such as clamping force decay and thermal deformation accumulation develop to an unacceptable level, avoiding the risk of process loss of control in traditional clamping methods where "processing begins immediately after clamping and problems are discovered afterward".

[0022] 7. In this invention, after processing, a digital archive is generated that includes an initial set of control parameters, a history curve of feedback signal changes, compensation action records, and a history of deviations between predicted and measured values. This digital archive transforms the clamping process from a one-time "black box operation" into a transparent process that is traceable, analyzable, and reproducible, providing a complete data foundation for quality traceability and process improvement.

[0023] 8. In this invention, the digital twin module reads digital files from the historical database at preset intervals and uses measured deviation values ​​to calibrate the simulation model parameters. This mechanism enables the prediction accuracy of the digital twin model to continuously improve with the number of uses, realizing the system's self-learning evolution and allowing for continuous optimization of the clamping scheme and compensation strategy during long-term use.

[0024] 9. In this invention, the control module compares the deviation between the measured value and the predicted value with a first threshold and a second threshold respectively. When the deviation is less than or equal to the first threshold, only the second actuating element performs high-frequency compensation. When the deviation is greater than the first threshold but less than or equal to the second threshold, the first and second actuating elements work together to perform low-frequency large-amplitude compensation. This hierarchical judgment mechanism avoids system oscillations caused by triggering large adjustments when the deviation is small, thus improving the rationality of the compensation action and the system stability.

[0025] 10. In this invention, when the deviation exceeds the second threshold, the control module issues an alarm signal and suspends processing. This mechanism can respond promptly to abnormal conditions such as sudden increases in cutting force, tool breakage, and clamping failure, stopping the processing action in the shortest possible time to prevent equipment damage and workpiece scrapping, thus ensuring the safety of the processing process.

[0026] 11. In this invention, the first threshold can work normally within the range of 5% to 15% of the predicted value, the second threshold within the range of 25% to 35% of the predicted value, and the weighting coefficient within the range of 0.1 to 0.7. This provides sufficient adjustment space for parameter configuration under different processing scenarios and enhances the system's adaptability to different mold materials, different processing procedures, and different environmental conditions. Attached Figure Description

[0027] Figure 1 This is a system connection diagram of a rapid positioning and clamping system and its control method for mold frame processing proposed in this invention; Figure 2 This is a schematic diagram of the composite clamping unit structure of a rapid positioning and clamping system and its control method for mold frame processing proposed in this invention; Figure 3 This is a flowchart of a rapid positioning and clamping system and its control method for mold frame processing proposed in this invention. Detailed Implementation

[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0029] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include," "contain," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "includes..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0030] refer to Figure 1-3 This invention provides a rapid positioning and clamping system and its control method for mold frame processing. Its core components include four main parts: a composite clamping unit array, a self-powered sensing module, a digital twin module, and a control module.

[0031] The composite clamping unit is fabricated as follows: NiTi shape memory alloy wire with a diameter of 0.5mm to 1.0mm is wound into a helical spring shape, with an outer diameter of 8mm to 15mm and 5 to 12 turns. After heat treatment to set the shape, the SMA spring is connected in series with the upper and lower sets of piezoelectric ceramic stacks via a pre-tightening screw. The piezoelectric ceramic stacks are made of PZT-5H material, with a single layer thickness of 0.1mm, and a total of 50 to 200 layers are stacked. The stack height is 5mm to 20mm, and the pre-compression of each set is 0.5μm to 5μm. The entire composite clamping unit is fixed in the T-slot of the worktable via a mounting base.

[0032] The energy management unit of the self-powered sensing module adopts the following circuit topology: the charge signal output terminal of each piezoelectric ceramic stack is connected to the rectifier bridge circuit. After rectification, it is smoothed by the filter capacitor, and then outputs a stable 3.3V or 5V DC voltage through the DC-DC boost / buck converter (such as LTC3588 or similar energy harvesting chip) to power the microcontroller (such as STM32L0 series low power MCU) and the wireless transmission module (such as nRF24L01 or BLE module).

[0033] The digital twin module runs on a computer, using ABAQUS or ANSYS as the finite element simulation engine, NSGA-II or MOEA / D as the optimization algorithm, and Python or MATLAB as the programming language.

[0034] The control module is based on an embedded industrial controller (such as a PLC or an ARM-based real-time controller). It receives sensor data collected by each composite clamping unit wirelessly and outputs drive signals to each unit.

[0035] Calibration method for SMA resistance-clamping force mapping relationship: Before the system is put into use, the mapping relationship between the resistance value of the shape memory alloy driving layer and the clamping force needs to be calibrated. The calibration method is as follows: Take a NiTi shape memory alloy spring specimen (wire diameter 0.6 mm, spring outer diameter 10 mm, 8 turns) of the same specifications as in the composite clamping unit and mount it on a universal testing machine. Apply a series of loads from 0 N to 200 N to the specimen. Under each load value, heat from 20 °C to 120 °C at a rate of 0.5 °C / min, and simultaneously record the resistance, temperature, and output force of the specimen. Plot the resistance-temperature curve and the force-temperature curve based on the recorded data.

[0036] The initial temperature at which the resistance value abruptly changes on the resistance-temperature curve corresponds to the martensitic transformation initiation temperature M_s (typically 55℃-65℃), and the temperature at which the resistance value stabilizes corresponds to the austenitic transformation completion temperature A_f (typically 75℃-90℃). Within this temperature range, the rate of change of resistance and the progress of the phase transformation show a monotonic relationship. The force-temperature curve reflects the output force value of the SMA spring at different temperatures.

[0037] Through the above calibration, a mapping relationship F=f(R) between the resistance value R and the output clamping force F is established. Simultaneously, a theoretical resistance-time curve model is established, which takes the amplitude I of the current pulse, the pulse width t_p, and the initial ambient temperature T_0 as input parameters, and outputs a curve showing the change of the theoretical resistance value R_th(t) over time. After calibration, the mapping relationship and curve model are stored in the digital twin module.

[0038] Calibration method for piezoelectric charge signal-stress mapping relationship: The mapping relationship between charge signal and stress on the piezoelectric ceramic stack was calibrated. The piezoelectric ceramic stack was mounted on a mechanical loading device, and a series of compressive stresses from 0 MPa to 50 MPa were applied. The amount of charge output by the piezoelectric stack under each stress value was recorded simultaneously.

[0039] Calibration results show that, within the range of 0 MPa to 50 MPa, the output charge Q of the piezoelectric ceramic stack exhibits a good linear relationship with the applied compressive stress σ, with a proportionality constant (piezoelectric constant) of approximately 2.5 × 10⁻⁶. -10 C / N up to 3.5 × 10 -10 C / N conversion efficiency is approximately 35%-40%.

[0040] Based on the calibration results, a charge signal-stress mapping relationship σ=g(Q) was established. Simultaneously, cutting force spectra for typical mold frame machining (rough milling, finish milling, drilling, and boring) were obtained through cutting experiments. It was determined that the charge signal frequencies generated by cutting vibration stress are mainly concentrated above 100Hz, while the charge signal frequencies generated by cutting force changes and thermal deformation are mainly concentrated below 100Hz. 100Hz was used as the boundary frequency between high-frequency and low-frequency components.

[0041] Specific implementation methods of digital twin co-simulation: The co-simulation of the digital twin module is performed according to the following steps. The following steps S1-1 to S1-5 together constitute the specific implementation of step S1: S1-1 Modeling Stage: Import the CAD 3D model of the mold base, identify the geometric contour of the mold base, the location of guide pin holes, the location of ejector pin holes, the location of the frame opening, and the area to be machined. Set material property parameters, including: elastic modulus E (210 GPa for commonly used mold base materials such as P20 mold steel, and 205 GPa for S50C), Poisson's ratio ν (0.28-0.32), and density ρ (7800-7850 kg / m³). 3 ).

[0042] S1-2 Clamping Scheme Generation: Based on the geometric features of the mold base and the distribution of the area to be processed, candidate clamping points are generated on the bottom and side surfaces of the mold base. Each candidate clamping point corresponds to the position coordinates of a composite clamping unit. By combining different clamping points, no fewer than 1000 candidate clamping schemes are generated.

[0043] S1-3 Clamping Deformation Simulation: For each candidate scheme, a finite element model is established in ABAQUS. Boundary conditions are set: the contact surface between the bottom surface of the mold frame and the worktable is in frictional contact, with a friction coefficient of 0.15-0.25. Target clamping force (range 50N-500N, determined according to the mold frame size and material) is applied at each clamping point. The elastic deformation distribution of the mold frame under the clamping force is calculated, and the maximum deformation f1 is recorded.

[0044] S1-4 Machining Process Simulation: For each clamping scheme, rough milling, finish milling, drilling, and boring operations are simulated simultaneously. Cutting Force Model: The classic empirical formula for milling force, F_c=K_c·a_p·f_z·Z, is adopted, where K_c is the specific cutting force (approximately 1800-2200 N / mm² for P20 material). 2 ), where a_p is the depth of cut (1-3mm for rough milling, 0.1-0.5mm for finish milling), f_z is the feed per tooth (0.05-0.2mm / z), and Z is the number of teeth on the milling cutter. Calculate the displacement response and vibration amplitude of the mold base under dynamic cutting force in each process, and record the maximum dynamic displacement f2 and the maximum vibration amplitude f3.

[0045] S1-5 Multi-Objective Optimization: Using F=[f1,f2,f3] as the multi-objective function, the clamping force at each clamping point (50-500N) and the pre-compression of each piezoelectric ceramic stack (0.5-5μm) are the optimization variables. The allowable clamping stress of the mold material (not exceeding 60% of the material's yield strength) and the maximum allowable compression of each piezoelectric stack are the constraints. The NSGA-II algorithm is used for optimization, with a population size of 200 and 500 iterations, outputting a Pareto optimal solution set. The optimal solution is selected from the Pareto optimal solution set according to preset weights (e.g., deformation weight 0.5, vibration amplitude weight 0.3, processing time weight 0.2).

[0046] Implementation method for clamping and compensation actions of the compound clamping unit: S2 clamping stage: According to the target output value in the control parameter set, a first drive signal is output to each composite clamping unit. The first drive signal is a current pulse signal with a current amplitude of 2A to 8A and a pulse width of 500ms to 2000ms. In response to the first drive signal, the first actuating element (SMA spring) of each composite clamping unit increases its temperature from room temperature (approximately 20℃-25℃) to the austenitic phase transformation completion temperature (approximately 75℃-90℃), undergoes phase transformation contraction, generates a first displacement, and applies clamping force to the mold frame through the preload screw.

[0047] During the applied current pulse, the resistance value signal (first feedback signal) of each first actuating element is acquired at a sampling frequency of not less than 1kHz. The measured resistance value is compared in real time with the theoretical resistance-time curve output by the digital twin module. When the deviation is within a preset range (±5% of the theoretical value), the current is cut off and the first drive signal is canceled. After the current is cut off, the SMA spring naturally cools to room temperature (approximately 40℃-60℃), still maintaining the deformation and stress of the austenitic phase, and the clamping force retention rate can reach over 90%.

[0048] When the deviation between the measured resistance value and the theoretical value exceeds the preset range, the output pulse width of the first drive signal is extended (the single extension is 50ms to 200ms) or the number of outputs is increased (the maximum number of increases is 3 to 5). When the deviation still exceeds the preset range after the number of outputs is increased to the maximum number of increases, an alarm signal is issued.

[0049] S3 Compensation Phase: During the machining process, the charge signals (second feedback signals) output by the second actuating element (piezoelectric ceramic stack) of each composite clamping unit in response to external stress are acquired at a sampling frequency of not less than 10kHz. The acquired charge signals are decomposed into high-frequency components (>100Hz, corresponding to cutting vibration stress) and low-frequency components (≤100Hz, corresponding to cutting force variation stress and thermal deformation stress) through a 100Hz digital filter. The high-frequency and low-frequency components are compared with the predicted charge signals output by the digital twin module, and the high-frequency deviation value and low-frequency deviation value are calculated respectively.

[0050] To verify the superiority of the graded compensation strategy of this invention over a single driving element, three sets of comparative experiments were conducted under the same processing conditions: Comparison Group A: SMA clamping only (no piezoelectric compensation). The flatness of the upper surface of the mold base is 0.12mm, the surface roughness Ra=1.6μm, and high-frequency vibration cannot be suppressed.

[0051] Comparison Group B: Piezoelectric compensation only (without SMA low-frequency adjustment). The piezoelectric stroke is only 2μm, which cannot compensate for the 15μm quasi-static displacement caused by changes in cutting force, resulting in a mold position drift of 0.08mm.

[0052] Comparative Group C: The composite clamping unit of this invention (SMA + piezoelectric hierarchical synergy). Flatness 0.05mm, surface roughness Ra=0.8μm, position drift 0.01mm.

[0053] The comparative results show that the flatness of the present invention is improved by 58% compared with the SMA clamping solution alone, and the positional accuracy is improved by 87.5% compared with the piezoelectric compensation solution alone.

[0054] When the high-frequency deviation value is greater than zero, a second drive signal (voltage signal) is output to the corresponding second actuating element. The second actuating element responds to the second drive signal by generating a second displacement (reverse displacement, with a response frequency in the kHz range and a displacement amount of 0.1 μm to 5 μm). The amplitude of the second drive signal has a linear proportional relationship with the high-frequency deviation value, with a proportionality coefficient of 0.5 to 2.0.

[0055] When the low-frequency deviation value is greater than the first threshold (5%-15% of the predicted value output by the digital twin module) and less than or equal to the second threshold (25%-35% of the predicted value output by the digital twin module), a third drive signal (current pulse signal, pulse width 50ms to 200ms, amplitude 30%-70% of the first drive signal) is output to the corresponding first actuating element. The first actuating element responds to the third drive signal by generating a third displacement (additional contraction or relaxation, response time on the order of hundreds of milliseconds), and simultaneously outputs a second drive signal to the second actuating element. The two work together to adjust the clamping force distribution. When the deviation is greater than the second threshold, an alarm signal is issued and processing is paused.

[0056] To verify the effectiveness of digital twin co-simulation optimization, a comparison was conducted under the same mold frame and processing conditions: Digital twin-less optimization (human experience-based point placement) scheme: clamping deformation 0.12mm, dynamic displacement during processing 0.08mm, vibration amplitude 15μm.

[0057] The digital twin optimization scheme (the solution of this invention) results in: clamping deformation of 0.04 mm (reduction of 67%), dynamic displacement of 0.03 mm (reduction of 62.5%), and vibration amplitude of 6 μm (reduction of 60%). S4 Assessment and Adjustment Phase: During the processing, the static first feedback signal (resistance measurement value of the first actuating element in the state of no driving signal) and the static second feedback signal (charge output value of the second actuating element in the state of no dynamic external stress) of each composite clamping unit are collected at a preset cycle of 1 minute to 10 minutes, and the clamping status evaluation value is calculated.

[0058] The static resistance value R0, static charge output value Q0, and temperature value T0 measured within the first second after all composite clamping units complete clamping and current cutoff in step S2 are used as initial reference values. The first rate of change of the current static first feedback signal relative to the initial static first feedback signal reference value, the second rate of change of the current static second feedback signal relative to the initial static second feedback signal reference value, and the third rate of change of the current temperature value relative to the initial temperature reference value are calculated respectively. The weighted sum of the first, second, and third rates of change is used as the evaluation value E = α·ΔR + β·ΔQ + γ·ΔT, where the sum of the three weighting coefficients is 1, and the value of each weighting coefficient ranges from 0.1 to 0.7.

[0059] When the evaluation value deviates from the preset range (e.g., the absolute value of E is greater than 0.15), a fourth drive signal is output to the second actuator for fine adjustment. If the evaluation value still deviates from the preset range after fine adjustment, a fifth drive signal is output to the first actuator for fine adjustment.

[0060] To verify the rationality of the first threshold range of 5%-15% and the second threshold range of 25%-35%, a comparative experiment was conducted outside the threshold range: When the first threshold is below 5% (e.g., set to 1%), piezoelectric compensation frequently intervenes even with small deviations, increasing the number of piezoelectric stack actions by 300%, increasing system power consumption and not contributing to accuracy improvement; when the first threshold is above 15% (e.g., set to 20%), the piezoelectric compensation that should intervene early fails to be triggered in time, and the accumulation of vibration causes the surface roughness to deteriorate from 0.8μm to 1.4μm.

[0061] When the second threshold is higher than 35% (e.g., set to 45%), under abnormal conditions of sudden increase in cutting force, the system does not trigger an alarm when the deviation reaches 38%, and the continued processing leads to tool breakage and workpiece surface scratches; when the second threshold is lower than 25% (e.g., set to 15%), the alarm is triggered when the cutting force fluctuates normally during normal processing, with a false alarm rate as high as 25%, which seriously affects processing efficiency.

[0062] The above comparison proves that the value ranges of the first threshold of 5%-15% and the second threshold of 25%-35% are critical and not arbitrary.

[0063] S5 data closed-loop stage: After machining, the feedback signal changes of each compound clamping unit and the data of each compensation action are recorded to generate a digital archive. The digital archive includes: the initial control parameter set, the historical curves of the feedback signal changes of each compound clamping unit, the timestamps and amplitude records of each compensation action, and the historical deviation between the predicted and measured values ​​of the digital twin module. The digital archive is stored in a historical database.

[0064] The digital twin module reads digital files from the historical database at preset intervals (e.g., after processing 10 identical mold frames), and calibrates the simulation model parameters using measured deviation values. The calibration method is as follows: using the measured deviation value as the dependent variable and the simulation model parameters (material elastic modulus, contact surface friction coefficient, structural damping coefficient) as independent variables, the parameter correction amount that minimizes the sum of squares of the deviations is solved, and the parameter correction amount is added to the current model parameters.

[0065] To verify the effectiveness of digital twin self-learning calibration, the change in prediction error during the continuous processing of 100 mold frames of the same model was statistically analyzed: Items 1-10: Average prediction error 12%, maximum error 18%; Items 11-50: Average prediction error 6%, maximum error 10%; Items 51-100: Average prediction error 2.8%, maximum error 5%.

[0066] The prediction error continued to decrease with the number of processing cycles, indicating that the digital twin model achieved self-learning evolution through calibration with historical data.

[0067] Example 1 Objective: To verify the clamping efficiency, machining accuracy, and energy-saving effect of the system of the present invention in rough milling of large mold frames.

[0068] Implementation System: The rapid positioning and clamping system for mold frame processing described in this invention is employed. The composite clamping unit array comprises 12 composite clamping units. The first actuating element of each composite clamping unit is a NiTi shape memory alloy drive (0.6mm wire diameter, 10mm spring outer diameter, 8 turns), and the second actuating element is a PZT-5H piezoelectric ceramic stack (0.1mm single-layer thickness, 100 stack layers, 10mm stack height). The self-powered sensing module includes a signal acquisition unit, an energy conversion unit (rectifier bridge circuit + filter capacitor + LTC3588 DC-DC converter), and a power output unit (outputting 3.3V DC voltage). The digital twin module runs the ABAQUS simulation engine and NSGA-II optimization algorithm on a computer. The control module is an ARM-based embedded real-time controller that communicates with each composite clamping unit via an nRF24L01 wireless module.

[0069] The mold base to be processed is an automotive body panel mold base, made of P20 mold steel, with external dimensions of 1200mm × 800mm × 400mm and a weight of approximately 1200kg. The machining process is rough milling of the upper surface, with a cutting depth a_p = 2.5mm, a milling cutter diameter of φ80mm, a spindle speed of 1500rpm, and a feed rate of 600mm / min.

[0070] Implementation steps: S1: Import the 3D model of the mold base and machining parameters (rough milling of the upper surface, cutting depth 2.5mm, milling cutter diameter 80mm, spindle speed 1500rpm, feed rate 600mm / min) into the digital twin module. The modeling unit establishes a digital model containing the geometric contour and the area to be machined. The clamping simulation unit generates 2000 candidate clamping schemes based on the digital model, performs finite element simulation of clamping deformation on each candidate scheme, and calculates the clamping deformation amount corresponding to each scheme. The machining simulation unit simultaneously performs machining process simulation on each candidate scheme, and calculates the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit uses the clamping deformation amount, dynamic displacement amount, and vibration amplitude as optimization targets and outputs the optimal scheme: Arrange 12 composite clamping units on the worktable, of which 8 are distributed on the four sides of the bottom surface of the mold base (2 on each side), and 4 are distributed on the four corners of the bottom surface of the mold base. The target clamping force of each unit is 350N for the corner units and 250N for the four side units. The pre-compression amount of each piezoelectric stack is set to 2μm.

[0071] S2: The control module outputs a first drive signal (current pulse, amplitude 5A, pulse width 1200ms) to each composite clamping unit according to the target output value (350N for the four corner units, 250N for the four side units) in the control parameter set. The first actuating element (SMA spring) of each composite clamping unit responds to the first drive signal, undergoing phase change contraction after its temperature rises to 82℃, generating a first displacement, and applying clamping force to the mold frame through the preload screw. During the output of the first drive signal, the resistance value signal (first feedback signal) of each first actuating element is acquired at a sampling frequency of 1kHz. The measured resistance value is compared with the theoretical value output by the digital twin module. If the deviation is within ±3% of the theoretical value (within the preset range ±5%), the current is cut off, and the first drive signal is canceled. The total clamping time is approximately 8 seconds.

[0072] S3: After rough milling begins, the charge signal (second feedback signal) output by the second actuating element (piezoelectric ceramic stack) of each composite clamping unit in response to external stress is acquired at a sampling frequency of 10kHz. The second feedback signal is compared with the predicted value output by the digital twin module. During machining, the maximum deviation of the high-frequency component is 6% of the predicted value, which is within the first threshold (8% of the predicted value). A second drive signal (voltage signal) is output to the corresponding second actuating element, and the second actuating element responds to the second drive signal by generating a reverse displacement (maximum 1.2μm) to actively suppress high-frequency vibration. The maximum deviation of the low-frequency component is 12% of the predicted value, which is greater than the first threshold (8%) and less than the second threshold (30% of the predicted value). A third drive signal (current pulse, amplitude 3A, pulse width 100ms) is output to the corresponding first actuating element. The first actuating element responds to the third drive signal by generating additional contraction, and simultaneously outputs the second drive signal to the second actuating element. Both work together to adjust the clamping force distribution. No alarm is triggered during the entire machining process, which lasts for approximately 45 minutes.

[0073] S4: During the processing, the static resistance value (resistance measurement value of the first actuating element in the absence of a drive signal) and static charge output value (charge output value of the second actuating element in the absence of dynamic external stress) of each composite clamping unit are collected every 5 minutes to calculate the clamping status evaluation value. The maximum evaluation value at each time point is 0.09, which does not deviate from the preset range (E≤0.15) and does not trigger adjustment.

[0074] S5: After machining, record the feedback signal changes of each composite clamping unit and the data of each compensation action to generate a digital file. Input the digital file into the digital twin module to update the simulation model parameters.

[0075] Implementation Results: In this embodiment, during the 45-minute rough milling process, the system consumed approximately 0.005 kWh of energy only during the clamping stage, maintaining clamping with zero energy consumption during the machining phase. After machining, the flatness of the upper surface of the mold base was measured to be 0.05 mm, and the surface roughness Ra = 0.8 μm. The system's digital profile shows that the SMA resistance value drifted by a maximum of 2.3% of its initial value during machining, indicating stable clamping. This embodiment verifies the effectiveness and energy-saving advantages of this invention in rough milling of heavy-tonnage mold bases.

[0076] Example 2 Purpose of implementation: To verify the vibration suppression effect and surface quality improvement effect of the system of the present invention in precision milling of mold frames.

[0077] Implementation System: The rapid positioning and clamping system for mold frame processing described in this invention is employed. The composite clamping unit array comprises six composite clamping units. The first actuating element of each composite clamping unit is a NiTi shape memory alloy drive (0.5mm wire diameter, 8mm spring outer diameter, 6 turns), and the second actuating element is a PZT-5H piezoelectric ceramic stack (0.1mm single-layer thickness, 50 stacked layers, 5mm stack height). The self-powered sensing module, digital twin module, and control module are configured the same as in Embodiment 1.

[0078] The mold base to be processed is a laptop shell mold base, made of S50C, with dimensions of 400mm × 300mm × 150mm and a weight of approximately 60kg. The machining process is precision milling of the parting surface, with a cutting depth a_p = 0.2mm, a milling cutter diameter of φ20mm, a spindle speed of 8000rpm, a feed rate of 300mm / min, and a required surface roughness Ra ≤ 0.8μm.

[0079] Implementation steps: S1: Import the 3D model of the mold base and machining parameters (fine milling of the parting surface, cutting depth 0.2mm, milling cutter diameter 20mm, spindle speed 8000rpm, feed rate 300mm / min) into the digital twin module. The modeling unit establishes a digital model containing the geometric contour and the area to be machined. The clamping simulation unit generates 1500 candidate clamping schemes based on the digital model, performs finite element simulation of clamping deformation on each candidate scheme, and calculates the clamping deformation amount corresponding to each scheme. The machining simulation unit simultaneously performs machining process simulation on each candidate scheme, and calculates the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit uses the clamping deformation amount, dynamic displacement amount, and vibration amplitude as optimization targets and outputs the optimal scheme: arrange 6 composite clamping units on the worktable, evenly distributed around the bottom surface of the mold base, with a target clamping force of 150N for each unit and a pre-compression amount of 1μm for each piezoelectric stack.

[0080] S2: The control module outputs a first drive signal (current pulse, amplitude 3A, pulse width 800ms) to each composite clamping unit according to the target output value (150N per unit) in the control parameter set. The first actuating element (SMA spring) of each composite clamping unit responds to the first drive signal, undergoing phase change contraction after its temperature rises to 78℃, generating a first displacement, and applying clamping force to the mold frame through the preload screw. During the output of the first drive signal, the resistance value signal (first feedback signal) of each first actuating element is acquired at a sampling frequency of 1kHz. The measured resistance value is compared with the theoretical value output by the digital twin module. If the deviation is within ±2% of the theoretical value (within the preset range ±5%), the current is cut off, and the first drive signal is canceled.

[0081] S3: After the finish milling begins, the charge signal (second feedback signal) output by the second actuating element (piezoelectric ceramic stack) of each composite clamping unit in response to external stress is acquired at a sampling frequency of 10kHz. The second feedback signal is compared with the predicted value output by the digital twin module. Due to the small cutting force of finish milling, the maximum deviation of the high-frequency component is 3% of the predicted value, which is within the first threshold (5% of the predicted value). A second drive signal (voltage signal) is output to the corresponding second actuating element, and the second actuating element responds to the second drive signal by generating a reverse displacement (maximum 0.3μm) to actively suppress high-frequency vibration. The maximum deviation of the low-frequency component is 6% of the predicted value, which is greater than the first threshold (5%) and less than the second threshold (25% of the predicted value). A third drive signal (current pulse, amplitude 2A, pulse width 80ms) is output to the corresponding first actuating element. The first actuating element responds to the third drive signal by generating additional contraction, and at the same time outputs the second drive signal to the second actuating element. The two work together to adjust the clamping force distribution. The machining lasts for about 20 minutes without triggering an alarm.

[0082] S4: During processing, the static resistance and static charge output values ​​of each composite clamping unit are collected every 3 minutes to calculate the clamping status evaluation value. The maximum evaluation value at each time point is 0.08, which does not deviate from the preset range (E≤0.15) and does not trigger adjustment.

[0083] S5: After machining, record the feedback signal changes of each composite clamping unit and the data of each compensation action to generate a digital file. Input the digital file into the digital twin module to update the simulation model parameters.

[0084] Implementation Results: In this embodiment, the piezoelectric ceramic stack achieved active suppression of high-frequency vibration during precision milling. After machining, the flatness of the parting surface was measured to be 0.015 mm, and the surface roughness Ra = 0.6 μm, meeting the mold requirements (Ra ≤ 0.8 μm). The system's digital archive showed a maximum evaluation value E of 0.08, far below the alarm threshold of 0.15. This embodiment verifies the vibration suppression and surface quality improvement effects of the present invention in precision machining.

[0085] Example 3 Objective: To verify the impact load suppression effect and hole position accuracy improvement effect of the system of the present invention in the drilling / boring of mold frames.

[0086] Implementation System: The rapid positioning and clamping system for mold frame processing described in this invention is employed. The composite clamping unit array comprises 8 composite clamping units. The first actuating element of each composite clamping unit is a NiTi shape memory alloy drive (wire diameter 0.55mm, spring outer diameter 9mm, 7 turns), and the second actuating element is a PZT-5H piezoelectric ceramic stack (single layer thickness 0.1mm, 80 stack layers, stack height 8mm). The self-powered sensing module, digital twin module, and control module are configured the same as in Embodiment 1.

[0087] The mold base to be processed is an injection mold base, made of P20, with external dimensions of 600mm×500mm×250mm and a weight of approximately 200kg. The processing steps are drilling guide post holes (φ32mm, depth 150mm) and boring (φ40mm, accuracy H7).

[0088] Implementation steps: S1: Import the 3D model of the mold base and machining parameters (drilling guide post holes φ32mm depth 150mm, boring φ40mm precision H7) into the digital twin module. The modeling unit establishes a digital model including geometric contours, guide post hole positions, and the area to be machined. The clamping simulation unit generates 1800 candidate clamping schemes based on the digital model, performs clamping deformation finite element simulation on each candidate scheme, and calculates the clamping deformation amount corresponding to each scheme. The machining simulation unit simultaneously performs machining process simulation on each candidate scheme, and calculates the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit uses clamping deformation, dynamic displacement, and vibration amplitude as optimization targets and outputs the optimal scheme: arrange 8 composite clamping units on the worktable, with 4 distributed at the four corners of the mold base bottom surface and 4 distributed around the guide post holes. The target clamping force of each unit is 300N for the four corner units and 200N for the hole perimeter units. The pre-compression amount of each piezoelectric stack is set to 1.5μm.

[0089] S2: The control module outputs a first drive signal (current pulse, amplitude 4A, pulse width 1000ms) to each composite clamping unit according to the target output value (300N for the four corner units, 200N for the hole perimeter units) in the control parameter set. The first actuating element (SMA spring) of each composite clamping unit responds to the first drive signal, undergoing phase change contraction after its temperature rises to 80℃, generating a first displacement, and applying clamping force to the mold frame through the preload screw. During the output of the first drive signal, the resistance value signal (first feedback signal) of each first actuating element is acquired at a sampling frequency of 1kHz. The measured resistance value is compared with the theoretical value output by the digital twin module. If the deviation is within ±4% of the theoretical value (within the preset range ±5%), the current is cut off, and the first drive signal is canceled.

[0090] S3: After drilling begins, the charge signal (second feedback signal) output by the second actuating element (piezoelectric ceramic stack) of each composite clamping unit in response to external stress is acquired at a sampling frequency of 10kHz. The second feedback signal is compared with the predicted value output by the digital twin module. During drilling, the axial force of drilling is approximately 800N, generating a large low-frequency impact. The maximum deviation of the low-frequency component is 18% of the predicted value, which is greater than the first threshold (10%) but less than the second threshold (30% of the predicted value). A third drive signal (current pulse, amplitude 3A, pulse width 120ms) is output to the corresponding first actuating element. The first actuating element responds to the third drive signal by generating additional contraction, and simultaneously outputs a second drive signal to the second actuating element. Both work together to adjust the clamping force distribution. During boring, the cutting force is relatively small but requires extremely high hole position accuracy. The piezoelectric stack actively suppresses micro-vibration with a high-frequency response. No alarm is triggered during the machining process.

[0091] S4: During processing, the static resistance and static charge output values ​​of each composite clamping unit are collected every 4 minutes to calculate the clamping status evaluation value. The maximum evaluation value at each time point is 0.11, which does not deviate from the preset range (E≤0.15) and does not trigger adjustment.

[0092] S5: After machining, record the feedback signal changes of each composite clamping unit and the data of each compensation action to generate a digital file. Input the digital file into the digital twin module to update the simulation model parameters.

[0093] Implementation Results: In this embodiment, the synergistic effect of SMA and piezoelectricity effectively suppressed low-frequency deformation caused by drilling impact during drilling. After machining, the guide post hole position accuracy was measured to be φ0.02mm, and the hole diameter accuracy met the H7 requirement. The system's digital archive recorded the timing and amplitude of each compensation action. This embodiment verifies the compensation capability and hole position accuracy improvement effect of the present invention under impact load conditions.

[0094] Example 4 Purpose of implementation: To verify the thermal deformation sensing and active compensation effect of the system of the present invention under high temperature environment.

[0095] Implementation System: The rapid positioning and clamping system for mold frame processing described in this invention is employed. The composite clamping unit array comprises 10 composite clamping units. The first actuating element of each composite clamping unit is a NiTi shape memory alloy drive (wire diameter 0.65mm, spring outer diameter 12mm, 10 turns), and the second actuating element is a PZT-5H piezoelectric ceramic stack (single layer thickness 0.1mm, 150 stacked layers, stack height 15mm). The self-powered sensing module includes temperature sensors (arranged on the surface of each composite clamping unit). The digital twin module incorporates the coefficient of thermal expansion (11.5 × 10⁻⁶ for H13 steel) in the simulation. -6 / ℃). The other modules are the same as in Example 1.

[0096] The mold base to be processed is a large die-casting mold base, made of H13 hot work die steel, with external dimensions of 1000mm×700mm×350mm and a weight of approximately 800kg. The processing environment is a summer workshop temperature of approximately 35℃, and the mold base will experience a significant temperature rise (approximately 15℃-20℃) due to cutting heat during the processing.

[0097] Implementation steps: S1: Import the 3D model of the mold base and its machining parameters (including an ambient temperature of 35℃) into the digital twin module. The modeling unit establishes geometric contours and material properties (including a thermal expansion coefficient of 11.5 × 10⁻⁶). -6The digital model includes the area to be processed (at / ℃). The clamping simulation unit generates 2000 candidate clamping schemes based on the digital model, performs finite element simulation of clamping deformation (including thermal deformation calculation) on each candidate scheme, and calculates the clamping deformation amount corresponding to each scheme. The machining simulation unit simultaneously performs machining process simulation on each candidate scheme, calculating the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit uses clamping deformation, dynamic displacement, and vibration amplitude as optimization targets and outputs the optimal scheme: 10 composite clamping units are arranged on the worktable, each unit has a target clamping force of 300N, and the pre-compression amount of each piezoelectric stack is set to 2μm.

[0098] S2: The control module outputs a first drive signal (current pulse, amplitude 4.5A, pulse width 1100ms) to each composite clamping unit according to the target output value (300N per unit) in the control parameter set. The first actuating element (SMA spring) of each composite clamping unit responds to the first drive signal, undergoing phase change contraction after its temperature rises to 85℃, generating a first displacement, and applying clamping force to the mold frame through the preload screw. During the output of the first drive signal, the resistance value signal (first feedback signal) of each first actuating element is acquired at a sampling frequency of 1kHz. The measured resistance value is compared with the theoretical value output by the digital twin module. If the deviation is within ±3% of the theoretical value (within the preset range ±5%), the current is cut off, and the first drive signal is canceled.

[0099] S3: After processing begins, the charge signal (second feedback signal) output by the second actuating element (piezoelectric ceramic stack) of each composite clamping unit in response to external stress is acquired at a sampling frequency of 10kHz. The second feedback signal is compared with the predicted value output by the digital twin module. In the early stage of processing, both high-frequency and low-frequency components are within the threshold, and the piezoelectric stack performs conventional vibration suppression.

[0100] S4: After approximately 30 minutes of processing, the temperature sensor detected that the mold frame temperature rose from 35℃ to 52℃. Every 5 minutes, the system collects the static resistance, static charge output, and temperature values ​​of each composite clamping unit to calculate the clamping status evaluation value. In this embodiment, the weighting coefficients are α=0.5, β=0.3, and γ=0.2. This weighting is a balanced configuration: the resistance signal (α) directly characterizes the SMA clamping force retention and is the core indicator of the clamping status, hence its highest weight; the charge signal (β) reflects the clamping contact state, and is second; the temperature signal (γ) reflects the heat accumulation effect and has the lowest weight. The sum of the three weighting coefficients is 1, which meets the normalization requirement of the evaluation value formula.

[0101] At this point, ΔT = (52-35) / 35 = 0.486, ΔR = 0.042, ΔQ = 0.038, and E = 0.5 × 0.042 + 0.3 × 0.038 + 0.2 × 0.486 = 0.123, which is within the preset range (E ≤ 0.15). The system continues processing.

[0102] After approximately 60 minutes of processing, the temperature sensor detected that the mold base temperature had risen to 58℃. At this point, ΔT = 0.657, ΔR = 0.055, and ΔQ = 0.048. E = 0.5 × 0.055 + 0.3 × 0.048 + 0.2 × 0.657 = 0.168, the evaluated value deviates from the preset range (E > 0.15). The system outputs a fourth drive signal (voltage signal) to the second actuating element of the three composite clamping units corresponding to the thermal deformation direction for fine-tuning. After fine-tuning, the evaluated value is 0.16, still deviating from the preset range. The system outputs a fifth drive signal (current pulse, amplitude 3.5A, pulse width 150ms) to the first actuating element of these three units for fine-tuning. The first actuating element responds to the fifth drive signal by generating additional contraction, adjusting the clamping force distribution to compensate for thermal deformation. After adjustment, the evaluated value drops to 0.11, returning to the preset range.

[0103] S5: After machining, record the feedback signal changes of each composite clamping unit and the data of each compensation action to generate a digital profile. The digital profile includes the temperature change history, the evaluation value change curve, and the thermal compensation action record. Input the digital profile into the digital twin module to update the thermal expansion compensation parameters in the simulation model.

[0104] Implementation Results: In this embodiment, under high-temperature conditions, the system senses thermal deformation through temperature sensors, automatically calculates and evaluates values, and triggers SMA collaborative fine-tuning when values ​​exceed limits. After processing, the mold flatness was measured at 0.08 mm, indicating effective control of thermal deformation. A digital archive fully records the evolution and compensation process of thermal deformation. This embodiment verifies the invention's ability to sense and actively compensate for thermal deformation under high-temperature conditions.

[0105] Example 5 Purpose of implementation: To verify the rationality of the parameter ranges of the system of the present invention (first threshold 5%-15%, second threshold 25%-35%, weight coefficient 0.1-0.7), and to prove that the system can work normally under different parameter combinations.

[0106] Implementation System: The rapid positioning and clamping system for mold frame processing described in this invention is employed. The composite clamping unit array comprises 8 composite clamping units. The first actuating element of each composite clamping unit is a NiTi shape memory alloy drive (wire diameter 0.6mm, spring outer diameter 10mm, 8 turns), and the second actuating element is a PZT-5H piezoelectric ceramic stack (single layer thickness 0.1mm, 100 stacked layers, stack height 10mm). The self-powered sensing module, digital twin module, and control module are configured the same as in Embodiment 1.

[0107] The mold base to be processed is made of P20 material, with external dimensions of 600mm×500mm×250mm and a weight of approximately 200kg. The processing procedure is planar milling, with the same cutting parameters.

[0108] Implementation steps: Three sets of different parameters were used for comparative verification: Group A: The first threshold is 5% of the predicted value output by the digital twin module, the second threshold is 25% of the predicted value, and the weighting coefficients are α=0.7, β=0.1, and γ=0.2.

[0109] S1: Import the 3D model of the mold base and machining parameters into the digital twin module. The modeling unit establishes the digital model. The clamping simulation unit generates candidate clamping schemes and performs clamping deformation simulation. The machining simulation unit performs machining process simulation. The optimization unit outputs the optimal scheme: arrange 8 composite clamping units on the worktable, with a target clamping force of 250N for each unit and a pre-compression amount of 1.5μm for each piezoelectric stack.

[0110] S2: The control module outputs a first drive signal (current pulse, amplitude 4A, pulse width 1000ms) to each composite clamping unit according to the target output value. Each first actuating element responds to the first drive signal, generates a first displacement, and applies a clamping force. During the output of the first drive signal, a first feedback signal is acquired, and the measured value is compared with the target value. If the deviation is within ±5% of the preset range, the first drive signal is revoked.

[0111] S3: During processing, a second feedback signal is acquired and compared with the predicted value output by the digital twin module. Due to the low first threshold (5%), piezoelectric compensation is triggered frequently. When the high-frequency deviation value is greater than zero, a second drive signal is output to the second actuator. The amplitude of the second drive signal is linearly proportional to the high-frequency deviation value (proportional coefficient 1.0). When the low-frequency deviation value is greater than the first threshold and less than or equal to the second threshold, a third drive signal is output to the first actuator, and simultaneously, a second drive signal is output to the second actuator.

[0112] S4: During processing, static first feedback signal and static second feedback signal are collected every 5 minutes, and the evaluation value E = 0.7·ΔR + 0.1·ΔQ + 0.2·ΔT is calculated. The evaluation value fluctuates slightly and no adjustment is triggered.

[0113] S5: After processing, generate a digital file and input it into the digital twin module.

[0114] Group B: The first threshold is 10% of the predicted value output by the digital twin module, and the second threshold is 30% of the predicted value. The weighting coefficients are α=0.5, β=0.3, and γ=0.2. The implementation steps are the same as Group A, but the piezoelectric compensation trigger frequency in S3 is moderate, and the evaluation value calculation formula in S4 is E=0.5·ΔR+0.3·ΔQ+0.2·ΔT.

[0115] Group C: The first threshold is 15% of the predicted value output by the digital twin module, and the second threshold is 35% of the predicted value. The weighting coefficients are α=0.1, β=0.7, and γ=0.2. The implementation steps are the same as in Group A, but the piezoelectric compensation trigger frequency is lower in S3, and the SMA compensation intervention threshold is higher. In S4, the evaluation value calculation formula is E=0.1·ΔR+0.7·ΔQ+0.2·ΔT. The evaluation value fluctuates greatly but is still within a controllable range.

[0116] Implementation results: In group A, due to the low first threshold (5%), piezoelectric compensation is frequently triggered, the piezoelectric stack has a large power consumption but a good vibration suppression effect, and the surface roughness Ra=0.55μm.

[0117] In Group B, all indicators are balanced, the surface roughness Ra=0.62μm, and the system energy consumption is moderate.

[0118] In group C, because the weighting coefficient is biased towards the charge signal (β=0.7), the system is more sensitive to changes in clamping force. The evaluation value fluctuates greatly but is still within a controllable range. The surface roughness of the machined surface is Ra=0.70μm.

[0119] The machining accuracy of all three sets of parameters meets the requirements for mold base machining (Ra≤0.8μm), indicating that the parameters of this invention can work normally within the first threshold range of 5%-15%, the second threshold range of 25%-35%, and the weighting coefficient range of 0.1-0.7. This embodiment verifies the rationality and wide adaptability of the parameter ranges of this invention.

[0120] Supplementary verification, alarm trigger verification: To verify the effectiveness of the alarm function (which issues an alarm signal and suspends processing when the deviation exceeds the second threshold), an abnormal working condition simulation was added to the processing of Example 1.

[0121] In Example 1, after approximately 20 minutes of rough milling, a severely worn end mill was manually replaced (flank wear VB = 0.8 mm, normal VB ≤ 0.3 mm), and machining continued. The worn tool caused a sudden increase in cutting force, and the actual measured depth of cut increased from the set 2.5 mm to approximately 4.0 mm (due to the instantaneous increase in cutting force caused by tool chipping).

[0122] The piezoelectric ceramic stack detected a low-frequency deviation value that instantaneously reached 38% of the predicted value, exceeding the second threshold (30% of the predicted value). The system issued an alarm signal and paused processing within 200ms after the deviation exceeded the second threshold. Simultaneously, the SMA and piezoelectric unit maintained their current states to prevent displacement of the mold holder during the emergency stop. Upon inspection, the operator found severe tool wear. After replacing the tool and restarting processing, the system automatically restored its previous clamping and compensation states and continued processing.

[0123] This supplementary verification shows that under abnormal operating conditions, the system's deviation judgment and alarm functions can be triggered normally, and the pause processing function is effective, verifying the feasibility of the technical feature in the claim that "an alarm signal is issued and processing is paused when the deviation is greater than the second threshold".

[0124] Comparative Example 1 A conventional hydraulic clamp was used to clamp and rough mill a mold frame (P20 mold steel, 1200mm×800mm×400mm) of the same specifications as in Example 1. The hydraulic clamp was provided with clamping force by four hydraulic cylinders, each with a clamping force of 500N, and a continuous oil supply was required to maintain the clamping state.

[0125] During machining, the hydraulic system operated continuously, and the increased oil temperature caused fluctuations in clamping force (fluctuation range ±8%). Rough milling took 45 minutes, consuming approximately 2.2 kWh of hydraulic system energy. Because the hydraulic fixture provides rigid clamping, it cannot sense or compensate for vibrations and thermal deformations during machining. The flatness of the mold base's upper surface was 0.15 mm, and the surface roughness Ra = 1.8 μm. After machining, inspection revealed approximately 0.10 mm of clamping deformation springback in the mold base, requiring secondary finishing correction.

[0126] In contrast, under the same processing conditions, Embodiment 1 of the present invention achieves zero-energy clamping (consuming approximately 0.005 kWh of electrical energy only during the clamping stage), a flatness of 0.05 mm (an improvement of 67%), and a surface roughness Ra = 0.8 μm (an improvement of 56%), without requiring secondary correction.

[0127] Comparative Example 2 An electromagnetic clamp was used to precision mill a mold frame (S50C, 400mm×300mm×150mm) of the same specifications as in Example 2. The electromagnetic clamp fixes the mold frame by electromagnetic attraction and needs to be continuously powered to maintain the attraction (power consumption of about 0.8kW).

[0128] The precision milling process takes 20 minutes, and the electromagnetic fixture consumes approximately 0.27 kWh of energy. Because the electromagnetic fixture lacks vibration sensing and active compensation capabilities, cutting vibrations are directly transmitted to the mold frame during machining. The parting surface flatness is 0.045 mm, and the surface roughness Ra = 1.2 μm. Since Ra ≤ 0.8 μm does not meet the mold requirements, polishing is necessary.

[0129] In Embodiment 2 of the present invention, under the same conditions: zero energy consumption to maintain clamping, flatness of 0.015mm, and surface roughness Ra=0.6μm, directly meeting the mold requirements.

[0130] Comparative Example 3 A simple SMA fixture containing only SMA clamping units (without piezoelectric ceramic stacks, self-powered sensing modules, or digital twin modules) was used to drill holes in a mold frame of the same specifications as in Example 3.

[0131] The simple SMA clamping fixture is arranged based on operator experience, with one SMA clamp at each of the four corners of the mold base bottom, for a total of four. The clamping force is adjusted by the operator to approximately 300N by feel. The SMA springs contract to generate clamping force after being heated by electricity, and maintain clamping after power is cut off. However, due to the lack of piezoelectric sensing and active compensation capabilities, drilling impacts and vibrations cannot be detected during machining. During drilling, the axial force of drilling causes a slight displacement of the mold base (approximately 0.05mm), resulting in a guide post hole position deviation of φ0.08mm, exceeding the tolerance requirement (φ0.03mm). Simultaneously, due to the lack of digital twin optimization and reliance on manual experience for the clamping point layout, the mold base clamping deformation is relatively large (0.12mm).

[0132] Under the same conditions, Embodiment 3 of the present invention has a guide post hole position accuracy of φ0.02mm and a clamping deformation of 0.03mm, all of which are better than those of Comparative Example 3.

[0133] Based on the above embodiments and comparative examples, the present invention has the following outstanding technical effects compared with the prior art: Zero-energy clamping and holding: In Examples 1 to 5, the first actuating element (SMA spring) retains more than 90% of the clamping force even after the first drive signal is removed after the phase change clamping is completed. In Comparative Example 1, the hydraulic clamp continuously consumes 2.2 kWh of energy, and in Comparative Example 2, the electromagnetic clamp continuously consumes 0.27 kWh of energy. However, the embodiments of the present invention consume virtually no energy after clamping is completed, resulting in significant energy-saving effects.

[0134] Active vibration suppression: In Examples 1 to 5, the second actuating element (piezoelectric ceramic stack) realizes active vibration suppression by sensing external stress, outputting a second feedback signal, and generating a second displacement in response to the second driving signal. The quality of the processed surface is improved by more than 50% compared with Comparative Examples 1-3.

[0135] Clamping-machining collaborative optimization: In Examples 1 to 5, the digital twin module establishes a digital model through the modeling unit, performs clamping deformation simulation through the clamping simulation unit, performs machining process simulation through the machining simulation unit, and selects the optimal solution with clamping deformation, dynamic displacement and vibration amplitude as optimization targets. The clamping deformation is reduced by more than 60% compared with Comparative Example 3 (human experience-based point placement).

[0136] Full-band dynamic stability: In Examples 1 to 5, the first actuating element (response time in the hundreds of milliseconds, used for large-amplitude adjustment at low frequencies) and the second actuating element (response frequency in the kHz level, used for small-amplitude suppression at high frequencies) complement each other in terms of response frequency, covering the entire frequency band from quasi-static to high-frequency vibration.

[0137] Self-powered sensing: In Examples 1 to 5, the second feedback signal (charge signal) output by the second actuating element under pressure and vibration is converted into electrical energy by the energy conversion unit of the self-powered sensing module, which then powers the signal acquisition unit, realizing wireless and battery-free deployment. The comparative examples all require external power.

[0138] Clamping status is traceable and self-learning: In Examples 1 to 5, after machining is completed, the system generates a digital archive containing an initial set of control parameters, historical curves of feedback signal changes, compensation action records, and historical deviations between predicted and measured values, which is stored in a historical database. The digital twin module reads the historical database and calibrates the simulation model parameters according to a preset cycle. In the comparative example, the clamping status cannot be traced and analyzed.

[0139] Alarm and safety assurance: Supplementary verification shows that when the deviation exceeds the second threshold, the system can issue an alarm signal and suspend processing within 200ms, effectively ensuring the safety of equipment and workpieces under abnormal operating conditions.

[0140] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.

[0141] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A rapid positioning and clamping system for mold base processing, characterized in that, include: A composite clamping unit array, wherein each composite clamping unit in the composite clamping unit array includes a first actuating element and a second actuating element; the first actuating element and the second actuating element are connected in series via a pre-tightening screw, the driving force generated by the first actuating element is transmitted to the workpiece via the second actuating element, and the second actuating element is in a pre-compressed state when clamped. The self-powered sensing module is connected to each of the composite clamping units and is used to collect the status feedback signals of each of the composite clamping units, and convert a portion of the status feedback signals into electrical energy to power the self-powered sensing module. The digital twin module is used to perform joint simulation of clamping status and machining process based on the three-dimensional model of the workpiece and machining process parameters, and output the control parameter set of each composite clamping unit; The control module is connected to the composite clamping unit array, the self-powered sensing module and the digital twin module respectively, and is used to control the output state of each composite clamping unit according to the control parameter set and the state feedback signal. The first actuating element has a power-off self-locking characteristic, which is used to maintain the clamping force without consuming external energy during the processing stage, so that the second actuating element can focus on dynamic response during the processing stage; the second actuating element is the same physical element, which alternately or simultaneously undertakes the sensing function, execution function and energy harvesting function in time or working conditions.

2. The quick positioning and clamping system for mold base processing as described in claim 1, characterized in that, The first actuating element is used to generate a first displacement and output a first feedback signal in response to a first driving signal, and the first actuating element maintains the first displacement after the first driving signal is withdrawn; the second actuating element is used to generate a second displacement in response to a second driving signal and output a second feedback signal in response to external stress; the state feedback signal includes the first feedback signal and the second feedback signal; the sensing function, execution function and energy harvesting function of the second actuating element share the same piezoelectric material body, and the same charge signal is distinguished and processed in the time domain by a signal processing circuit: the high-frequency component is used to sense vibration stress, the low-frequency component is used to execute displacement output, and the charge accumulation part is used for energy harvesting.

3. The quick positioning and clamping system for mold frame processing as described in claim 1, characterized in that, The first actuating element is a shape memory alloy driving element, the first driving signal is a current pulse signal, and the first feedback signal is a resistance value signal; the second actuating element is a piezoelectric driving element, the second driving signal is a voltage signal, and the second feedback signal is a charge signal; the self-powered sensing module converts the charge signal into electrical energy; the shape memory alloy driving element is a helical spring structure, the piezoelectric driving element is a stacked structure, the axial centerline of the shape memory alloy driving element coincides with the axial centerline of the piezoelectric driving element, and the two are connected in series along the same axis through the preload screw to form a single force transmission chain.

4. The quick positioning and clamping system for mold frame processing as described in claim 1, characterized in that, The self-powered sensing module includes: A signal acquisition unit, connected to each of the composite clamping units, is used to acquire the status feedback signal; An energy conversion unit, connected to each of the composite clamping units, is used to convert a portion of the state feedback signal into electrical energy and store it. A power supply output unit, wherein the power input terminal of the power supply output unit is connected to the power output terminal of the energy conversion unit, and the power output terminal of the power supply output unit is connected to the power supply terminal of the signal acquisition unit.

5. The quick positioning and clamping system for mold base processing as described in claim 1, characterized in that, The digital twin module includes: Modeling unit, used to create a digital model of the workpiece; The clamping simulation unit is used to generate multiple candidate clamping schemes based on the digital model, perform clamping deformation simulation on each candidate clamping scheme, and calculate the clamping deformation amount corresponding to each scheme. The machining simulation unit is used to simultaneously perform machining process simulation on each of the candidate clamping schemes, and calculate the dynamic displacement and vibration amplitude during the machining process corresponding to each scheme. The optimization unit is used to select the optimal solution from the multiple candidate clamping solutions with the clamping deformation, the dynamic displacement and the vibration amplitude as optimization targets, and output the set of control parameters of each composite clamping unit corresponding to the optimal solution.

6. A rapid positioning and clamping control method for mold base processing, applied to the system described in any one of claims 1-5, characterized in that, Includes the following steps: S1: Import the 3D model of the workpiece and the machining process parameters, generate multiple candidate clamping schemes, perform clamping deformation simulation and machining process simulation on each candidate clamping scheme, select the optimal scheme with clamping deformation amount, machining process dynamic displacement amount and vibration amplitude as optimization targets, and output the control parameter set of each composite clamping unit; S2: According to the target output value in the set of control parameters, output a first drive signal to each composite clamping unit, and the first actuating element of each composite clamping unit responds to the first drive signal to generate a first displacement and apply a clamping force. During the output of the first drive signal, the first feedback signal of each first action element is collected, the measured first feedback signal is compared with the target value, and the first drive signal is canceled when the deviation is within a preset range. S3: During the machining process, the second feedback signal of the second action element of each composite clamping unit in response to external stress is collected, and the second feedback signal is compared with the predicted value output by the digital twin module. The stroke range of the second action element is at the micrometer level. When used alone, it can suppress high-frequency small-amplitude vibrations but cannot compensate for quasi-static large-amplitude displacements caused by thermal deformation and cutting force changes. The response time of the first action element is at the hundred-millisecond level. When used alone, it can achieve large-amplitude displacement adjustment but the response frequency is below 100Hz and cannot suppress high-frequency vibrations. Therefore, a graded compensation strategy is adopted: when the deviation is less than or equal to the first threshold, a second drive signal is output to the second action element, and the second action element generates a second displacement in response to the second drive signal. When the deviation is greater than the first threshold and less than or equal to the second threshold, a third drive signal is output to the first action element, and the first action element generates a third displacement in response to the third drive signal, while simultaneously outputting the second drive signal to the second action element. The cancellation of the first driving signal causes the first actuating element to enter a power-off self-locking state. During the processing stage, the first actuating element does not consume external energy, so that the charge signal output by the second actuating element can be collected by the self-powered sensing module and converted into electrical energy, thereby realizing self-powered closed-loop control. S4: During the processing, static first feedback signal and static second feedback signal of each composite clamping unit are collected at a preset cycle, and clamping status evaluation value is calculated; when the evaluation value deviates from the preset range, a fourth drive signal is output to the second action element for fine adjustment; if the evaluation value still deviates from the preset range after fine adjustment, a fifth drive signal is output to the first action element for fine adjustment. S5: After processing, record the feedback signal change data of each composite clamping unit and the data of each compensation action to generate a digital file; input the digital file into the digital twin module to update the simulation model parameters.

7. The rapid positioning and clamping control method for mold frame processing as described in claim 6, characterized in that, In step S1, the number of candidate clamping schemes is no less than 1000; the clamping deformation simulation uses the finite element method to calculate the elastic deformation of the workpiece under clamping force. The machining process simulation uses the finite element method to simulate the dynamic displacement and vibration amplitude of the workpiece during rough milling, finish milling, drilling, and boring processes. The optimization objectives are to minimize elastic deformation, dynamic displacement, and vibration amplitude. After outputting the Pareto optimal solution set using a multi-objective optimization algorithm, the optimal solution is selected according to preset weights. Compared with the manual experience clamping scheme without digital twin optimization, the control parameter set output by the digital twin module reduces clamping deformation by more than 60% and dynamic displacement by more than 40%.

8. The rapid positioning and clamping control method for mold frame processing as described in claim 6, characterized in that, In step S3, the sampling frequency of each second feedback signal is not less than 10kHz. After acquisition, each second feedback signal is decomposed into high-frequency components and low-frequency components. The high-frequency components correspond to cutting vibration stress, and the low-frequency components correspond to cutting force variation stress and thermal deformation stress. The high-frequency components and the low-frequency components are compared with the predicted values ​​output by the digital twin module, and the high-frequency deviation value and the low-frequency deviation value are calculated respectively. When the high-frequency deviation value is greater than zero, the second driving signal is output to the corresponding second actuator. The amplitude of the second driving signal is linearly proportional to the high-frequency deviation value, and the proportionality coefficient is 0.5 to 2.

0. When the low-frequency deviation value is greater than the first threshold and less than or equal to the second threshold, the third driving signal is output to the corresponding first actuator. The amplitude of the third driving signal is less than the amplitude of the first driving signal.

9. The rapid positioning and clamping control method for mold base processing as described in claim 6, characterized in that, In step S3, the first threshold is 5% to 15% of the predicted value output by the digital twin module, the second threshold is 25% to 35% of the predicted value output by the digital twin module, and the first threshold is less than the second threshold; when the deviation is greater than the second threshold, an alarm signal is issued and processing is suspended. In step S4, the preset period is 1 minute to 10 minutes; the static first feedback signal is the measured value of the feedback signal of the first actuating element in the state without a driving signal; the static second feedback signal is the measured value of the feedback signal of the second actuating element in the state without dynamic external stress; the clamping state evaluation value is: calculating the first rate of change of the current static first feedback signal relative to the initial static first feedback signal reference value, the second rate of change of the current static second feedback signal relative to the initial static second feedback signal reference value, and the third rate of change of the current temperature value relative to the initial temperature reference value, respectively, and using the weighted sum of the first rate of change, the second rate of change, and the third rate of change as the evaluation value, with the sum of the three weight coefficients being 1, and the value range of each weight coefficient being 0.1 to 0.

7.

10. The rapid positioning and clamping control method for mold base processing as described in claim 6, characterized in that, In step S5, the digital archive includes: an initial set of control parameters, historical curves of feedback signal changes for each composite clamping unit, timestamps and amplitude records of each compensation action, and historical deviations between the predicted and measured values ​​of the digital twin module. The digital archive is stored in a historical database. The digital twin module reads the digital archive from the historical database at a preset period and calibrates the simulation model parameters using the measured deviation values. The calibration method is as follows: using the measured deviation values ​​as the dependent variable and the simulation model parameters as the independent variable, the parameter correction amount that minimizes the sum of squared deviations is calculated, and the parameter correction amount is added to the current model parameters. After the digital twin module reads the historical database and calibrates the model parameters at a preset period, the prediction error decreases with the number of processing operations: the average prediction error for the first 10 processing operations is 12%, the average prediction error for the 11th to 50th processing operations decreases to 6%, and the average prediction error for the 51st and subsequent operations decreases to below 3%.

Images