Cutting process parameter design method, device, equipment and storage medium

Abstract

The present application relates to the technical field of cutting process parameter design, and particularly relates to a cutting process parameter design method, device, equipment and storage medium, the method firstly acquires a plurality of to-be-tested laser powers, calls a pre-trained target temperature prediction model to respectively predict the plurality of to-be-tested laser powers, obtains a plurality of predicted temperatures, performs a simulation operation based on the plurality of predicted temperatures to construct a first training sample set, constructs an initial cutting parameter prediction model, performs multi-round iterative training on the initial cutting parameter prediction model by using the first training sample set to obtain a target cutting parameter prediction model, acquires a plurality of to-be-tested temperatures, calls the target cutting parameter prediction model to respectively predict each to-be-tested temperature, obtains a plurality of initial cutting process data groups, adopts an SAC algorithm to screen in the plurality of initial cutting process data groups, and obtains an optimized cutting process data group, aiming to realize efficient prediction and optimization of critical cutting depth and cutting force.

Description

Technical Field

[0001] This invention relates to the field of cutting process parameter design technology, and in particular to a cutting process parameter design method, apparatus, equipment and storage medium. Background Technology

[0002] Monocrystalline silicon is a key material in inkjet-printed OLED (organic light-emitting diode) manufacturing, used for the printhead micro-nozzle plate, precision alignment components, and optical inspection windows. It possesses high hardness, a wide bandgap, and excellent optical properties. However, its inherent brittleness easily leads to brittle fracture and subsurface damage in traditional ultra-precision machining, severely impacting component lifespan and printing accuracy. To achieve high surface quality and shape accuracy in the plastic domain, laser-assisted machining and single-point diamond turning techniques have been introduced into this field. These techniques reduce the brittleness of silicon materials through localized laser thermal softening, combined with nanoscale precision cutting to suppress crack propagation and increase the critical depth of cut, thus meeting the reliability requirements of printing equipment for silicon-based components in high-frequency, high-precision environments.

[0003] However, existing technologies still have many shortcomings. For example, the influence of temperature on the brittle-plastic transition mechanism lacks systematic quantification; the nanoscale processing mechanism is difficult to observe directly through experiments; experimental temperature measurements are easily affected by process parameters, resulting in large errors in temperature field distribution; and the nonlinear relationship between laser power and temperature is not adequately modeled. In terms of numerical simulation, traditional finite element methods are difficult to simulate the large deformation and fracture processes of nanoscale cutting due to mesh distortion. While smoothed particle dynamics methods can solve the mesh distortion problem, they do not systematically analyze the influence of temperature on the brittle-plastic transition and cannot provide sufficient support for process optimization. Existing related patents either rely on high-cost external monitoring equipment, employ limited simulation methods, or depend on experimental trial and error to optimize parameters. These not only have limited applicability but also suffer from complex optimization processes and low efficiency, making it difficult to cope with the optimization needs of multivariable nonlinear process parameters and meet the high-precision processing requirements of key components in inkjet printing equipment. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the present invention aims to provide a cutting process parameter design method, device, equipment and storage medium, which aims to achieve efficient prediction and optimization of critical cutting depth and cutting force, and improve the critical cutting depth and surface quality of single crystal silicon devices under laser-assisted conditions.

[0005] The first aspect of this invention provides a method for designing cutting process parameters, comprising: acquiring multiple laser powers to be measured; calling a pre-trained target temperature prediction model to predict the multiple laser powers to be measured respectively, thereby obtaining multiple predicted temperatures; performing simulation operations based on the multiple predicted temperatures to construct a first training sample set; constructing an initial cutting parameter prediction model; using the first training sample set to perform multiple rounds of iterative training on the initial cutting parameter prediction model to obtain a target cutting parameter prediction model; acquiring multiple temperatures to be measured; calling the target cutting parameter prediction model to predict each temperature to be measured respectively, thereby obtaining multiple initial cutting process data sets, each initial cutting process data set including the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured; and using the SAC (Soft Actor-Critic) algorithm to filter the multiple initial cutting process data sets to obtain an optimized cutting process data set.

[0006] Optionally, in a first implementation of the first aspect of the present invention, before acquiring multiple laser powers to be measured and calling a pre-trained target temperature prediction model to predict the multiple laser powers to be measured respectively to obtain multiple predicted temperatures, the method further includes: constructing an initial simulation model; acquiring multiple laser power samples, performing simulation operations on each laser power sample based on the initial simulation model to obtain simulation results, and optimizing the initial simulation model based on the simulation results to obtain a target simulation model; acquiring multiple laser powers to be simulated, performing simulation operations on each laser power to be simulated based on the target simulation model to construct a second training sample set; constructing an initial temperature prediction model, and performing multiple rounds of iterative training on the initial temperature prediction model based on the second training sample set to obtain a target temperature prediction model.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the step of performing simulation operations on each of the laser power samples based on the initial simulation model to obtain simulation results, and optimizing the initial simulation model based on the simulation results to obtain a target simulation model, includes: presetting multiple initial correction coefficient candidate values ​​based on the initial simulation model; for each of the initial correction coefficient candidate values, calling the initial simulation model to perform simulation operations on each of the laser power samples to obtain simulation feature width data corresponding to each of the laser power samples; obtaining experimental feature width data corresponding to each of the laser power samples; and optimizing the initial simulation model based on the multiple initial correction coefficient candidate values, the multiple simulation feature width data, and the multiple experimental feature width data to obtain the target simulation model.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the step of optimizing the initial simulation model based on multiple initial correction coefficient candidate values, multiple simulation feature width data, and multiple experimental feature width data to obtain a target simulation model includes: for each initial correction coefficient candidate value, calculating the relative error between the simulation feature width data and the experimental feature width data corresponding to each laser power sample to obtain multiple relative error rates; obtaining a preset relative error threshold, and comparing each relative error rate with the relative error threshold; when there exists a situation where the relative error rate of all laser power samples under the initial correction coefficient candidate value is less than or equal to the relative error threshold, using the initial correction coefficient candidate value as the target correction coefficient value; and optimizing the initial simulation model based on the target correction coefficient value to obtain the target simulation model.

[0009] Optionally, in a fourth implementation of the first aspect of the present invention, the step of performing simulation operations based on multiple predicted temperatures to construct a first training sample set includes: constructing an SPH simulation (smooth particle hydrodynamic simulation) model for simulation; performing simulation operations on each predicted temperature based on the SPH simulation model to obtain cutting force data corresponding to each predicted temperature; constructing an initial relationship curve based on multiple predicted temperatures and multiple cutting force data; performing smoothing and noise reduction processing on the initial relationship curve using a Gaussian convolution algorithm to obtain a target relationship curve; obtaining the brittle-plastic transition depth corresponding to each predicted temperature based on the target relationship curve; and constructing the first training sample set based on multiple predicted temperatures, multiple cutting force data, and multiple brittle-plastic transition depths.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the step of using the SAC algorithm to filter multiple initial cutting process data sets to obtain an optimized cutting process data set includes: constructing an initial agent using the SAC algorithm; obtaining a predefined reward function; performing multiple rounds of iterative updates on the network parameters of the initial agent based on the reward function and multiple initial cutting process data sets; when a preset iteration stopping condition is met, using the updated agent as the target agent; and performing a filtering operation on multiple initial cutting process data sets based on the target agent to obtain the optimized cutting process data set.

[0011] Optionally, in the sixth implementation of the first aspect of the present invention, before obtaining the predefined reward function, the method further includes: determining multiple initial index parameters based on the initial cutting process data set; performing dimensional unification processing on the multiple index parameters using a normalization algorithm to obtain multiple normalized index parameters; obtaining a preset process priority requirement rule; assigning weights to each normalized index parameter based on the process priority requirement rule to obtain a weight coefficient corresponding to each normalized index parameter; and defining the reward function based on the multiple normalized index parameters and the multiple weight coefficients.

[0012] A second aspect of the present invention provides a cutting process parameter design apparatus, comprising: a temperature prediction module for acquiring multiple laser powers to be measured, and calling a pre-trained target temperature prediction model to predict the multiple laser powers to be measured respectively, thereby obtaining multiple predicted temperatures; a simulation module for performing simulation operations based on the multiple predicted temperatures to construct a first training sample set; a training module for constructing an initial cutting parameter prediction model, and using the first training sample set to perform multiple rounds of iterative training on the initial cutting parameter prediction model to obtain a target cutting parameter prediction model; a cutting parameter prediction module for acquiring multiple temperatures to be measured, and calling the target cutting parameter prediction model to predict each temperature to be measured respectively, thereby obtaining multiple initial cutting process data sets, each initial cutting process data set including the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured; and an optimization module for using the SAC algorithm to filter the multiple initial cutting process data sets to obtain an optimized cutting process data set.

[0013] A third aspect of the present invention provides a cutting process parameter design device, the cutting process parameter design device comprising: a memory and at least one processor, the memory storing instructions; at least one processor calling the instructions in the memory to cause the cutting process parameter design device to perform each step of the cutting process parameter design method described in any of the preceding claims.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the cutting process parameter design method described in any of the preceding claims.

[0015] In the technical solution of this invention, firstly, multiple laser powers to be measured are acquired, and a pre-trained target temperature prediction model is called to predict the multiple laser powers to be measured, resulting in multiple predicted temperatures. Simulation operations are performed based on the multiple predicted temperatures to construct a first training sample set. Then, an initial cutting parameter prediction model is constructed, and the initial cutting parameter prediction model is iteratively trained using the first training sample set to obtain a target cutting parameter prediction model. Next, multiple temperatures to be measured are acquired, and the target cutting parameter prediction model is called to predict each temperature to be measured, resulting in multiple initial cutting process data sets. Each cutting process data set includes the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured. Finally, the SAC algorithm is used to filter the multiple initial cutting process data sets to obtain an optimized cutting process data set. This method aims to achieve efficient prediction and optimization of critical cutting depth and cutting force, improving the critical cutting depth and surface quality of single-crystal silicon components under laser-assisted conditions. This method effectively overcomes the problems of brittle fracture and subsurface damage, enhances processing stability, and can meet the reliability and lifespan requirements of inkjet printing equipment for key silicon-based components in high-frequency and high-precision environments. Attached Figure Description

[0016] Figure 1 A logic flowchart of a cutting process parameter design method provided in an embodiment of the present invention;

[0017] Figure 2 A schematic diagram of the relationship between relative error rate and correction coefficient provided in an embodiment of the present invention;

[0018] Figure 3 This is a schematic diagram of the cutting process parameter design device provided in an embodiment of the present invention;

[0019] Figure 4 A schematic diagram of the cutting process parameter design equipment provided in an embodiment of the present invention. Detailed Implementation

[0020] This invention provides a method, apparatus, device, and storage medium for designing cutting process parameters. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the cutting process parameter design method in this invention includes:

[0022] 101. Obtain multiple laser powers to be measured, and call the pre-trained target temperature prediction model to predict the multiple laser powers to be measured respectively, so as to obtain multiple predicted temperatures;

[0023] In this embodiment, the pre-trained target temperature prediction model is built upon a large amount of laser-assisted processing experimental data and high-precision temperature field simulation data, through multiple rounds of algorithm iteration and parameter optimization. The model training process fully considers the nonlinear mapping relationship between laser power and temperature, as well as the influence of interference factors in the processing environment, resulting in excellent generalization ability and prediction accuracy. During the prediction execution phase, each laser power to be measured is sequentially input into the target temperature prediction model. The model, through its built-in feature extraction and mapping mechanism, quickly outputs the predicted temperature corresponding to each power value, achieving accurate quantitative conversion from laser power to processing temperature. This effectively avoids the problems of large errors, susceptibility to interference, and high equipment costs associated with traditional experimental measurements relying on thermometers and infrared detection, significantly improving the accuracy and reliability of temperature data acquisition.

[0024] 102. Perform simulation operations based on multiple predicted temperatures to construct a first training sample set;

[0025] In this embodiment, simulation operations based on multiple predicted temperatures are performed, with smooth particle dynamics as the core technology. A three-dimensional numerical model adapted to the ultra-precision oblique cutting scenario of single-crystal silicon is constructed using the LS-Dyna explicit dynamics simulation platform. This fully leverages the meshless characteristics of this method to simulate large deformation and fracture processes at the nanoscale, avoiding the mesh distortion limitations of traditional finite element methods. During the simulation, predetermined process parameter specifications are strictly followed, including geometric simplification design of local workpiece regions (such as truncating local workpiece regions to optimize computational efficiency), particle discretization with fixed particle size, tool modeling with specific cutting edge radius and angle parameters, and fixed boundary constraints with full degrees of freedom on the bottom and left sides. Simultaneously, tool movement is controlled according to preset cutting speed and trajectory to ensure that the maximum cutting depth covers the critical range of the brittle-plastic transition. Predicted temperature, as the core control variable, is incorporated into the simulation system in a gradient distribution from 20 to 800℃, forming multiple temperature conditions. The system simulates the cutting response of monocrystalline silicon under different temperature environments, focusing on capturing key physical phenomena such as material plastic flow, crack initiation and propagation, stress distribution, and hydrostatic pressure evolution. Simultaneously, it records the variation of cutting force with cutting depth and processes the original cutting force signal through Gaussian convolution to accurately locate the critical value of the brittle-plastic transition depth at each temperature. Based on each predicted temperature, the simulation output of the brittle-plastic transition depth, and cutting force, and other key process and material response indicators, the simulation data under multiple temperature domains and conditions are integrated and organized according to a unified data standard to form the first training sample set. Leveraging the simulation advantages of the SPH method, the system achieves accurate reproduction of the nanoscale cutting micro-mechanism and dynamic process of the brittle-plastic transition, which are difficult to observe in traditional experiments, effectively overcoming the limitations of physical experiments in microscale observation and multi-temperature domain parameter control. Through the systematic integration and precise processing of simulation data, the first training sample set constructed has the characteristics of high data consistency and complete correlation information, avoiding data fluctuations caused by environmental interference, equipment errors and other factors in physical experiments, and providing high-quality and reliable data support for the training of subsequent cutting force and brittle-plastic transition depth prediction models.

[0026] 103. Construct an initial cutting parameter prediction model, and use the first training sample set to perform multiple rounds of iterative training on the initial cutting parameter prediction model to obtain a target cutting parameter prediction model;

[0027] In this embodiment, when constructing the initial cutting parameter prediction model, a BP neural network is selected as the core architecture of the initial model, taking into account the nonlinear mapping characteristics of process parameters and performance indicators in the ultra-precision machining of single-crystal silicon. The first training sample set integrates the temperature-brittle-plastic transition depth data obtained from SPH cutting simulation and the measured results verified by the oblique cutting experiment. Among them, the measured results verified by the oblique cutting experiment refer to the processing and performance data obtained through actual operation in the dynamic load simulation experiment of continuous cutting depth of single-crystal silicon. This experiment uses a continuous cutting depth from 0 to 500 nm as a variable, and combines laser power to control the temperature of the workpiece deformation zone. It actually observes and records the state evolution of single-crystal silicon during the cutting process, including the actual position of the critical depth of brittle-plastic transition (i.e., the cutting depth threshold from plastic deformation dominance to brittle fracture dominance) under different temperature conditions, the processing state of the tool-workpiece contact area (such as the actual range of plastic flow, the specific process of crack initiation and propagation), and the quality characteristics of the processed surface (such as the presence and distribution of defects). It also includes actual measurement data of the cutting force changing with the cutting depth during the cutting process. The core function of these measured results is to verify and calibrate the simulation data obtained through SPH simulation. By comparing the deviation between the measured data and the simulation results, the parameter configuration of the cutting parameter prediction model is corrected, thereby improving the fit between the model prediction results and the actual processing conditions, and ensuring that the model can accurately reflect the real process response in the ultra-precision machining of single crystal silicon.

[0028] Before training using the first training sample set, the sample data needs to be standardized and preprocessed to eliminate dimensional differences. Then, the samples are input into the initial BP neural network model (backpropagation neural network model) in batches. The loss function is calculated based on the deviation between the predicted value and the true label of the sample. Through multiple iterations, the weights and threshold parameters of the network are adjusted. Simultaneously, oblique experimental data is introduced to verify the model's prediction results after each iteration. The experimental deviation is used to correct the model parameters until the model's prediction accuracy meets the preset error threshold, ultimately obtaining the target cutting parameter prediction model. The trained target cutting parameter prediction model can directly predict laser power to cutting performance indicators, replacing the inefficient traditional process of sequentially conducting simulations and experiments, significantly shortening the parameter prediction cycle.

[0029] 104. Obtain multiple temperatures to be measured, and call the target cutting parameter prediction model to predict each of the temperatures to be measured to obtain multiple initial cutting process data sets. Each initial cutting process data set includes the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured.

[0030] In this embodiment, the acquisition of the temperature to be measured needs to be combined with the temperature response characteristics of the brittle-plastic transition in the ultra-precision machining of single-crystal silicon. Multiple temperature values ​​covering the key process range are selected, typically covering the high-temperature range suitable for processing from room temperature to the plastic domain. These values ​​also include the brittle-plastic transition threshold and the characteristic temperatures corresponding to performance jumps, ensuring the representativeness and process coverage of the measured temperatures. When calling the target cutting parameter prediction model that has been trained and calibrated, each temperature to be measured is input into the model as an input feature. Based on the nonlinear mapping relationship between temperature and cutting process parameters formed during the previous training, the model can quickly output the cutting process parameters corresponding to that temperature, including core parameters such as the critical depth of the brittle-plastic transition and cutting force. Then, each temperature to be measured is integrated with its corresponding cutting process parameters to form an initial cutting process data set.

[0031] 105. The SAC algorithm is used to filter through multiple initial cutting process data sets to obtain an optimized cutting process data set.

[0032] In this embodiment, when using the SAC algorithm to screen the initial cutting process data set, it is necessary to rely on the algorithm's ability to adapt to continuous parameter optimization. Using the measured temperature and corresponding cutting process parameters in each data set as state inputs, and combining its entropy maximization exploration strategy, the algorithm explores potential high-quality process conditions while traversing typical working conditions of existing data sets, avoiding getting stuck in local optima and missing globally better process combinations. Using a predefined reward function, the reward value of each data set is calculated with normalized brittle-plastic transition depth, cutting force, and temperature as core dimensions. With the help of the Actor-Critic architecture collaborative optimization algorithm's decision strategy, this strategy continuously adjusts the selection bias towards the initial data set, gradually focusing on candidate data sets with better reward values. Simultaneously, the Critic module continuously improves the accuracy of its evaluation of the process value of the data set. This iterative process continues until the algorithm's decision strategy converges, i.e., the fluctuation range of the reward value stabilizes within a preset threshold. At this point, the process data set that matches the converged decision strategy is selected from multiple initial cutting process data sets, which is the optimized cutting process data set. The SAC algorithm's decision strategy convergence mechanism effectively avoids the limitations of local optima, accurately locating the globally optimal process combination that balances increased brittle-plastic transition depth, reduced cutting force, and reasonable temperature control. Meanwhile, the iterative optimization of the decision strategy does not rely on repeated trial and error in physical experiments, significantly shortening the data set screening cycle. The resulting optimized cutting process data set can directly match the high surface quality, low energy consumption, and low tool wear requirements of single-crystal silicon ultra-precision machining, improving the stability and economic efficiency of the machining process.

[0033] In this embodiment of the invention, before acquiring multiple laser powers to be measured and calling a pre-trained target temperature prediction model to predict the multiple laser powers to be measured to obtain multiple predicted temperatures, the method further includes: constructing an initial simulation model; acquiring multiple laser power samples, performing simulation operations on each laser power sample based on the initial simulation model to obtain simulation results, and optimizing the initial simulation model based on the simulation results to obtain a target simulation model; acquiring multiple laser powers to be simulated, performing simulation operations on each laser power to be simulated based on the target simulation model to construct a second training sample set; constructing an initial temperature prediction model, and performing multiple rounds of iterative training on the initial temperature prediction model based on the second training sample set to obtain a target temperature prediction model.

[0034] In this embodiment, the initial simulation model is constructed using the COMSOL multiphysics simulation platform to build a heat conduction model for laser heating of single-crystal silicon. The model is simplified and boundary conditions are configured based on the process scenario of single-crystal silicon nano-cutting. Specifically, the laser scanning is equated to a moving heat source, and the scanning speed is set to 1000 mm / min. The diamond tool is ignored in the physical modeling, and its impact on the laser is simplified to energy attenuation. A 60 μm laser spot is used, and the mesh of the main irradiated area is refined. Simultaneously, the silicon wafer model is simplified to 3×3×1.5 mm to balance computational efficiency and simulation accuracy. After obtaining multiple laser power samples covering different energy ranges, heat conduction simulations are performed on each sample based on the initial simulation model to obtain simulation results including temperature field distribution. The initial simulation model is then optimized based on the simulation results to obtain the target simulation model. Subsequently, multiple laser powers to be simulated, covering the energy range suitable for single-crystal silicon processing, were acquired. Parametric simulation of the target simulation model was performed on each laser power with a step size of 0.5W to obtain the stable surface temperature data of the workpiece corresponding to each laser power. At the same time, the data accuracy was improved by interpolation method, and the mapping relationship between laser power and corresponding temperature was integrated into a structured dataset, that is, the second training sample set was constructed. This breakthrough overcomes the problems of insufficient spatial resolution and physical interference faced by direct measurement of temperature in deformation zone in single-crystal silicon nano-cutting. Through model calibration combining experiments and simulations, the temperature field prediction accuracy of the target simulation model was greatly improved, ensuring the reliability and accuracy of the second training sample set.

[0035] When constructing the initial temperature prediction model, a neural network structure with one hidden layer and four nodes was adopted. The activation function of the hidden layer was selected as tansig (hyperbolic tangent activation function), and the activation function of the output layer was selected as purelin (linear transfer function). Based on the second training sample set, the initial model was trained in multiple rounds of iterations, with laser power as input and corresponding temperature as output. By adjusting the network weights and thresholds, the deviation between the predicted value and the actual temperature of the sample was minimized until the model prediction accuracy met the preset requirements. Finally, the target temperature prediction model was obtained, realizing a fast and accurate mapping from laser power to temperature. This replaced the time-consuming process of frequently performing simulations, significantly reduced the cost of acquiring temperature data, and provided high-quality basic temperature data for the subsequent optimization of cutting process parameters.

[0036] In this embodiment of the invention, the step of performing simulation operations on each laser power sample based on the initial simulation model to obtain simulation results, and optimizing the initial simulation model based on the simulation results to obtain a target simulation model, includes: presetting multiple initial correction coefficient candidate values ​​based on the initial simulation model; for each initial correction coefficient candidate value, calling the initial simulation model to perform simulation operations on each laser power sample to obtain simulation feature width data corresponding to each laser power sample; obtaining experimental feature width data corresponding to each laser power sample; and optimizing the initial simulation model based on the multiple initial correction coefficient candidate values, the multiple simulation feature width data, and the multiple experimental feature width data to obtain the target simulation model.

[0037] In this embodiment, when multiple initial correction coefficient candidate values ​​are preset based on the initial simulation model, it is necessary to combine the influence characteristics of energy attenuation during laser transmission on the heating temperature field of monocrystalline silicon, and select a combination of candidate values ​​covering a reasonable range of variation. Typically, the range is set to 0.6 to 1.1 with a step size of 0.1 to fully cover the potential adjustment range of the heat source energy distribution. For each initial correction coefficient candidate value, it is substituted into the Gaussian heat source power density distribution formula of the initial simulation model. Subsequently, a heat conduction simulation operation is performed on each laser power sample to simulate the temperature field evolution process of laser scanning on the monocrystalline silicon surface at the corresponding power. Then, the characteristic width data corresponding to the 3173℃ isotherm of the monocrystalline silicon boiling point is extracted, thereby obtaining the simulation characteristic width data matching each correction coefficient candidate value with each laser power sample. To obtain the experimental feature width data corresponding to each laser power sample, a high-power ablation experiment is required. Using the same power parameters as the laser power samples, a laser scan is performed on the single-crystal silicon surface at a speed of 1000 mm / min. The surface morphology of the ablation zone is observed using a white-light interferometer, and the width of the fully ablated region (i.e., the characteristic region corresponding to the boiling point isotherm of single-crystal silicon) is identified and measured. This yields the experimental feature width data corresponding to each laser power sample. Then, based on multiple simulated feature width data and multiple experimental feature width data, the initial simulation model is optimized to obtain the target simulation model. By traversing multiple sets of candidate correction coefficient values ​​and comparing them with experimental data, the influence of laser transmission energy attenuation on the temperature field simulation results is effectively corrected. This ensures that the output of the target simulation model highly matches the thermal response characteristics in actual processing. The calibrated target simulation model can accurately reproduce the temperature field distribution of single-crystal silicon under different laser powers, providing high-precision simulation support for the efficient acquisition of subsequent temperature data.

[0038] In this embodiment of the invention, optimizing the initial simulation model based on multiple initial correction coefficient candidate values, multiple simulation feature width data, and multiple experimental feature width data to obtain a target simulation model includes: for each initial correction coefficient candidate value, calculating the relative error between the simulation feature width data and the experimental feature width data corresponding to each laser power sample to obtain multiple relative error rates; obtaining a preset relative error threshold, and comparing each relative error rate with the relative error threshold; when there exists a situation where the relative error rate of all laser power samples under the initial correction coefficient candidate value is less than or equal to the relative error threshold, using the initial correction coefficient candidate value as the target correction coefficient value; and optimizing the initial simulation model based on the target correction coefficient value to obtain the target simulation model.

[0039] In this embodiment, for each candidate value of the initial correction coefficient, it is necessary to substitute them one by one into the modified Gaussian heat source power density formula used in the initial simulation model. The modified Gaussian heat source power density formula is the core input model for the simulation of the temperature field of single-crystal silicon laser heating. Its core parameter is the correction coefficient, which directly determines the distribution of laser energy on the surface of single-crystal silicon. Different correction coefficients will change the power density distribution of the heat source, thereby affecting the temperature field characteristics obtained from the simulation. Ultimately, this is reflected in the difference in the simulation feature width data corresponding to the boiling point of single-crystal silicon of 3173℃. Specifically, the calculation expression of the modified Gaussian heat source power density formula is as follows:

[0040] ,

[0041] in, This represents the laser power density at a specific point in space, i.e., the laser energy input per unit area. Its value reflects the concentration of laser energy at that location. This represents a correction factor, a dimensionless parameter used to quantify the energy attenuation effect caused by laser absorption and scattering during laser light transmission through a diamond tool. Its default value is 1 when there is no energy attenuation. The laser spot radius is a characteristic parameter that defines the spatial range of a Gaussian spot. It typically refers to the radius corresponding to the point where the energy drops to 1 / e² of the central peak. This represents the spatial distance between the calculation point and the center of the laser source; it is a parameter characterizing the position of the calculation point. This indicates the input laser power.

[0042] When optimizing the initial simulation model based on multiple simulation feature width data and experimental feature width data, it is necessary to combine the correction logic of the Gaussian heat source model. For each candidate value of the initial correction coefficient, the relative error corresponding to all laser power samples under that candidate value is calculated to obtain multiple relative error rates. The specific calculation expression is as follows:

[0043] ,

[0044] in, This represents the relative error rate. This represents the simulation feature width data. This represents the width data of the experimental feature.

[0045] Subsequently, a preset relative error threshold is invoked. This threshold is typically set to 10%, taking into account the experimental measurement accuracy, the actual adaptability of the simulation model, and the accuracy requirements of subsequent process applications. The relative error rate of all laser power samples corresponding to each initial candidate correction coefficient value is then compared sequentially with this threshold. Only when the relative error rate of all laser power samples for a given initial candidate correction coefficient value is less than or equal to this threshold is the candidate value deemed to effectively match the simulation and experimental data across the entire power range, and thus determined as the target correction coefficient value. For example, ... Figure 2 As shown, the horizontal axis represents the correction coefficient, and the vertical axis represents the relative error rate corresponding to each laser power sample. The three curves correspond to the relative error change trends of the three laser power samples of 21.92W, 29.58W, and 35.54W, respectively. When the correction coefficient... When the coefficient is 0.7, the relative errors between the simulated and measured widths under all three power levels converge within the 10% error band, indicating that the model calibration is complete. Substituting this target correction coefficient value into the Gaussian heat source formula of the initial simulation model completes the calibration of the heat source parameters and model optimization. The adjusted model is the target simulation model.

[0046] In this embodiment of the invention, the step of performing simulation operations based on multiple predicted temperatures to construct a first training sample set includes: constructing an SPH simulation model for simulation; performing simulation operations on each predicted temperature based on the SPH simulation model to obtain cutting force data corresponding to each predicted temperature; constructing an initial relationship curve based on multiple predicted temperatures and multiple cutting force data; smoothing and denoising the initial relationship curve using a Gaussian convolution algorithm to obtain a target relationship curve; obtaining the brittle-plastic transition depth corresponding to each predicted temperature based on the target relationship curve; and constructing the first training sample set based on multiple predicted temperatures, multiple cutting force data, and multiple brittle-plastic transition depths.

[0047] In this embodiment, when performing simulation operations for each predicted temperature based on the SPH simulation model, the predicted temperature needs to be substituted into the model as a variable, a uniform temperature field is set for the corresponding temperature condition, and the cutting behavior of the tool advancing along a preset trajectory during the oblique cutting of single-crystal silicon is simulated. Simultaneously, continuous data on the change of cutting force with cutting depth is recorded, thereby obtaining the cutting force data corresponding to each predicted temperature. This data directly characterizes the dynamic evolution of the material removal mode at different temperatures. Subsequently, an initial relationship curve is constructed based on multiple predicted temperatures and corresponding cutting force data. With temperature as the horizontal axis and cutting force as the vertical axis, each predicted temperature and its corresponding key cutting force value are mapped sequentially as coordinate points to form a temperature-cutting force correlation curve. This visually presents the evolution law of cutting force within different temperature ranges and clearly reflects the negative control characteristics of temperature on cutting force. Considering that the original cutting force data is easily affected by factors such as particle discretization errors and local stress concentration caused by built-up edge during the simulation process, resulting in random noise and local outliers, a Gaussian convolution algorithm is used to smooth and denoise the initial relationship curve. By weakening noise interference and correcting outlier data points through convolution operations, the curve's trend becomes more continuous and stable, ultimately yielding the target relationship curve.

[0048] In this embodiment, the evolution of cutting force with depth of cut is as follows: In the initial stage of cutting, when the depth of cut is low, single-crystal silicon is mainly removed by plastic flow. Workpiece particles form chips through continuous sliding, and the first deformation zone at the tool-workpiece contact surface expands steadily. At this time, the cutting force shows a steady upward trend, and the amplitude and frequency of fluctuations are at a low level. The fitting deviation value after Gaussian convolution processing also remains stable, reflecting the continuity and regularity of material deformation. As the depth of cut continues to increase, when the critical point of brittle-plastic transition is reached, microcracks begin to emerge inside the single-crystal silicon, and the material removal mode gradually transitions from plastic-dominated to brittle-dominated, entering the initial state of brittle-plastic mixed phase transition. The significant feature of this stage is that the amplitude and frequency of cutting force fluctuations increase synchronously. The initiation and micro-propagation of large cracks trigger instantaneous changes in cutting load, leading to increased fluctuations in the force signal. However, since the cracks have not yet formed a continuously expanding network, the material still exhibits some plastic response. The fitting deviation after Gaussian convolution remains low, without significant fluctuations, accurately characterizing the transitional characteristics of this mixed phase transition. When the cutting depth further increases, exceeding the critical range of the brittle-plastic transition, single-crystal silicon enters the brittle-dominated period. At this time, microcracks within the material rapidly interconnect and diffuse to form a crack network. Material removal is dominated by discontinuous fracture separation, and the load transfer during the cutting process becomes rapidly unstable, manifesting as violent oscillations in the cutting force. The Gaussian convolution fitting deviation also increases significantly, reflecting the dominant role of brittle fracture in cutting mechanics. Based on the above evolution law of cutting force with cutting depth, the starting position where the amplitude and frequency of cutting force fluctuations increase synchronously in the initial state of the brittle-plastic mixed phase transition is defined as the brittle-plastic transition depth of single-crystal silicon, based on the target relationship curve. This allows us to obtain the brittle-plastic transition depth corresponding to each predicted temperature. Finally, based on multiple predicted temperatures, corresponding cutting force core data, and accurately extracted brittle-plastic transition depth, a structured dataset was formed by integrating them according to a unified data specification, thus completing the first training sample set.

[0049] In this embodiment of the invention, the step of using the SAC algorithm to filter multiple initial cutting process data sets to obtain an optimized cutting process data set includes: constructing an initial agent using the SAC algorithm; obtaining a predefined reward function; performing multiple rounds of iterative updates on the network parameters of the initial agent based on the reward function and multiple initial cutting process data sets; when a preset iteration stopping condition is met, using the updated agent as the target agent; and performing a filtering operation on multiple initial cutting process data sets based on the target agent to obtain the optimized cutting process data set.

[0050] In this embodiment, when constructing the initial agent using the SAC algorithm, the algorithm's adaptability to continuous parameter optimization is fully utilized to build an Actor-Critic architecture that meets the optimization requirements of single-crystal silicon cutting process parameters. The Actor network is responsible for outputting temperature control actions, generating temperature suggestions within a continuous range based on a preset strategy. The Critic network is responsible for evaluating the comprehensive process value corresponding to the temperature and outputting a quantized Q-value. Simultaneously, an entropy maximization strategy is incorporated to ensure that while exploring known effective temperature ranges, the agent actively expands into unknown temperature domains, fundamentally avoiding the limitations of local optima and laying the foundation for discovering globally optimal process parameters. Based on a predefined reward function and multiple initial cutting process data sets, the initial agent performs multiple rounds of iterative updates. Specifically, the agent selects temperature parameters as control actions from the multiple initial cutting process data sets, and substitutes the brittle-plastic transition depth and cutting force from the initial cutting process data sets into the reward function for calculation to obtain the reward value corresponding to the temperature. Subsequently, the control actions and reward values, as well as other empirical data, are stored in the experience replay pool. The Actor network dynamically adjusts its strategy based on these experiences to optimize the temperature output tendency, while the Critic network continuously corrects the value evaluation model to improve the Q-value prediction accuracy. The two work together to complete the iterative update of the network parameters. When the preset iteration stopping conditions are met, i.e., the reward value fluctuation amplitude is stable within a preset threshold, the average reward value of multiple consecutive iterations does not significantly increase, or the preset number of iterations is reached, the parameter update stops. The agent at this point is the target agent.

[0051] When the target agent filters through multiple initial cutting process data sets, the Actor network of the target agent, based on the optimized strategy, accurately identifies the optimal temperature parameter that can balance the optimization objective from the data set. The set of process data such as the brittle-plastic transition depth and cutting force corresponding to this temperature is the optimized cutting process data set. The optimization objective refers to finding the optimal temperature through intelligent control in the optimization of process parameters for ultra-precision machining of single crystal silicon to maximize the brittle-plastic transition depth and minimize the cutting force and temperature, so that the three achieve the optimal synergistic state under the preset weight of the reward function.

[0052] In this embodiment, before obtaining the predefined reward function, the method further includes: determining multiple initial index parameters based on the initial cutting process data set; performing dimensional unification processing on the multiple index parameters using a normalization algorithm to obtain multiple normalized index parameters; obtaining a preset process priority requirement rule; assigning weights to each normalized index parameter based on the process priority requirement rule to obtain a weight coefficient corresponding to each normalized index parameter; and defining the reward function based on the multiple normalized index parameters and the multiple weight coefficients.

[0053] In this embodiment, when determining multiple initial index parameters based on the initial cutting process data set, it is necessary to closely adhere to the core optimization requirements of single-crystal silicon ultra-precision machining and extract key parameters directly related to process performance from the data set, specifically the brittle-plastic transition depth, cutting force, and temperature. Given the fundamental differences in the physical dimensions of the brittle-plastic transition depth, cutting force, and temperature, direct numerical calculations would lead to an imbalance in the weights of the optimization objectives. Therefore, a normalization algorithm is needed to scale the original values ​​of the three types of parameters to the range of 0 to 1. By eliminating the influence of different physical dimensions, the three types of parameters have an equal basis for numerical comparison, preventing any one index from dominating the reward calculation due to unit differences, and ensuring the fairness of subsequent weight allocation. After obtaining the preset process priority requirement rules, weights are assigned to each normalized index parameter based on these rules. The process priority requirement rules are guiding principles for prioritizing the core objective of "finding the optimal temperature to maximize the brittle-plastic transition depth and minimize cutting force and temperature" in combination with the specific requirements of actual processing scenarios. According to the process priority requirement rules, weight coefficients are assigned to the normalized index parameters corresponding to the brittle-plastic transition depth, cutting force, and temperature, and the constraint condition that the sum of the three types of coefficients is 1 is strictly followed, so that the reward function can accurately match the scenario requirements and guide the agent to evolve towards the preset optimization direction.

[0054] Next, a reward function is defined based on multiple normalized index parameters and their corresponding weighting coefficients. The specific calculation expression of the reward function is as follows:

[0055] ,

[0056] in, R(T) Represents the reward function, T Indicates temperature. Depth(T) Indicates the depth of the brittle-plastic transition. Force(T) Indicates cutting force. w 1 represents the first weighting coefficient corresponding to the depth of the brittle-plastic transition. w 2 represents the second weighting coefficient corresponding to the cutting force. w 3 indicates the third weighting coefficient corresponding to temperature. Indicates the depth of brittle-plastic transition Depth (T) The result after normalization Indicates the cutting force Force(T) The result after normalization Indicates temperature TThe result after normalization. The final defined reward function can transform multi-dimensional optimization objectives into intuitive quantitative signals, providing a clear learning guide for the agent of the SAC algorithm. This ensures that the agent can accurately capture the optimization patterns of process parameters during the iteration process, and ultimately find the global optimal solution that meets the needs of the scenario, thereby improving the intelligent optimization level and practical application value of monocrystalline silicon processing technology.

[0057] The cutting process parameter design method in the embodiments of the present invention has been described above. The cutting process parameter design apparatus in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 3 One embodiment of the cutting process parameter design device in this invention includes:

[0058] Temperature prediction module 301: used to acquire multiple laser powers to be measured, call a pre-trained target temperature prediction model to predict the multiple laser powers to be measured respectively, and obtain multiple predicted temperatures;

[0059] Simulation module 302: used to perform simulation operations based on multiple predicted temperatures to construct a first training sample set;

[0060] Training module 303: used to construct an initial cutting parameter prediction model, and to perform multiple rounds of iterative training on the initial cutting parameter prediction model using the first training sample set to obtain a target cutting parameter prediction model;

[0061] Cutting parameter prediction module 304: acquires multiple temperatures to be measured, calls the target cutting parameter prediction model to predict each of the temperatures to be measured, and obtains multiple initial cutting process data sets, each of the initial cutting process data sets including the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured;

[0062] Optimization module 305: Used to filter multiple initial cutting process data sets using the SAC algorithm to obtain an optimized cutting process data set.

[0063] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0064] above Figure 3 The cutting process parameter design device in this embodiment of the invention is described in detail from the perspective of modular functional entities. The cutting process parameter design equipment in this embodiment of the invention is described in detail from the perspective of hardware processing.

[0065] Figure 4This is a schematic diagram of a cutting process parameter design device 400 provided in an embodiment of the present invention. The cutting process parameter design device 400 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 410 (e.g., one or more processors) and a memory 420, and one or more storage media 430 (e.g., one or more mass storage devices) storing application programs 433 or data 432. The memory 420 and storage media 430 can be temporary or persistent storage. The program stored in the storage media 430 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the cutting process parameter design device 400. Furthermore, the processor 410 may be configured to communicate with the storage media 430 and execute the series of instruction operations in the storage media 430 on the cutting process parameter design device 400 to implement the steps of the cutting process parameter design method provided in the above-described method embodiments.

[0066] The cutting process parameter design device 400 may also include one or more power supplies 440, one or more wired or wireless network interfaces 450, one or more input / output interfaces 460, and / or one or more operating systems 431, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 4 The structure of the cutting process parameter design equipment shown does not constitute a limitation on the cutting process parameter design equipment. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0067] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a cutting process parameter design method.

[0068] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0069] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. 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 method for designing cutting process parameters, characterized in that, include: Construct the initial simulation model; Multiple laser power samples are obtained, and multiple candidate values ​​for initial correction coefficients are preset based on the initial simulation model. For each candidate initial correction coefficient, the initial simulation model is invoked to perform simulation operations on each laser power sample to obtain simulation feature width data corresponding to each laser power sample; experimental feature width data corresponding to each laser power sample is obtained; the initial simulation model is optimized based on multiple candidate initial correction coefficients, multiple simulation feature width data, and multiple experimental feature width data to obtain a target simulation model; multiple laser powers to be simulated are obtained, and simulation operations are performed on each laser power to be simulated based on the target simulation model to construct a second training sample set; An initial temperature prediction model is constructed, and the initial temperature prediction model is iteratively trained multiple times based on the second training sample set to obtain a target temperature prediction model; multiple laser powers to be measured are obtained, and the pre-trained target temperature prediction model is called to predict the multiple laser powers to be measured respectively to obtain multiple predicted temperatures; Simulation operations are performed based on multiple predicted temperatures to construct a first training sample set; An initial cutting parameter prediction model is constructed, and the initial cutting parameter prediction model is iteratively trained multiple times using the first training sample set to obtain a target cutting parameter prediction model. Multiple temperatures to be measured are acquired, and the target cutting parameter prediction model is called to predict each of the temperatures to be measured, thereby obtaining multiple initial cutting process data sets. Each initial cutting process data set includes the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured. The SAC algorithm is used to filter through multiple initial cutting process data sets to obtain an optimized cutting process data set.

2. The cutting process parameter design method according to claim 1, characterized in that, The process of optimizing the initial simulation model based on multiple candidate values ​​of the initial correction coefficients, multiple simulation feature width data, and multiple experimental feature width data to obtain the target simulation model includes: For each candidate value of the initial correction coefficient, the relative error between the simulated feature width data and the experimental feature width data corresponding to each laser power sample is calculated to obtain multiple relative error rates; Obtain a preset relative error threshold, and compare each relative error rate with the relative error threshold respectively; When the relative error rate of all laser power samples with an initial correction coefficient candidate value is less than or equal to the relative error threshold, the initial correction coefficient candidate value is used as the target correction coefficient value. The initial simulation model is optimized based on the target correction coefficient value to obtain the target simulation model.

3. The cutting process parameter design method according to claim 1, characterized in that, The step of performing simulation operations based on multiple predicted temperatures to construct a first training sample set includes: Construct an SPH simulation model for simulation purposes; Based on the SPH simulation model, simulation operations are performed on each of the predicted temperatures to obtain cutting force data corresponding to each of the predicted temperatures. An initial relationship curve is constructed based on multiple predicted temperatures and multiple cutting force data; The initial relationship curve is smoothed and denoised using a Gaussian convolution algorithm to obtain the target relationship curve. Based on the target relationship curve, the brittle-plastic transition depth corresponding to each of the predicted temperatures is obtained; The first training sample set is constructed based on multiple predicted temperatures, multiple cutting force data, and multiple brittle-plastic transition depths.

4. The cutting process parameter design method according to claim 1, characterized in that, The step of using the SAC algorithm to filter through multiple initial cutting process data sets to obtain an optimized cutting process data set includes: The initial agent is constructed using the SAC algorithm; Obtain a predefined reward function, and perform multiple rounds of iterative updates on the network parameters of the initial agent based on the reward function and multiple sets of initial cutting process data. When the preset iteration stopping condition is met, the updated agent is used as the target agent. The target intelligent agent performs a filtering operation on multiple initial cutting process data sets to obtain the optimized cutting process data set.

5. The cutting process parameter design method according to claim 4, characterized in that, Before obtaining the predefined reward function, the following steps are also included: Multiple initial index parameters are determined based on the initial cutting process data set; The normalization algorithm is used to unify the dimensions of the multiple index parameters to obtain multiple normalized index parameters; Obtain a preset process priority requirement rule, and assign weights to each normalized index parameter based on the process priority requirement rule to obtain a weight coefficient corresponding to each normalized index parameter. The reward function is defined based on multiple normalized index parameters and multiple weight coefficients.

6. A cutting process parameter design device, characterized in that, include: Temperature prediction module: used to build the initial simulation model; Multiple laser power samples are obtained, and multiple candidate values ​​for initial correction coefficients are preset based on the initial simulation model. For each candidate initial correction coefficient, the initial simulation model is invoked to perform simulation operations on each laser power sample to obtain simulation feature width data corresponding to each laser power sample; experimental feature width data corresponding to each laser power sample is obtained; the initial simulation model is optimized based on multiple candidate initial correction coefficients, multiple simulation feature width data, and multiple experimental feature width data to obtain a target simulation model; multiple laser powers to be simulated are obtained, and simulation operations are performed on each laser power to be simulated based on the target simulation model to construct a second training sample set; An initial temperature prediction model is constructed, and the initial temperature prediction model is iteratively trained multiple times based on the second training sample set to obtain a target temperature prediction model; multiple laser powers to be measured are obtained, and the pre-trained target temperature prediction model is called to predict the multiple laser powers to be measured respectively to obtain multiple predicted temperatures; Simulation module: used to perform simulation operations based on multiple predicted temperatures to construct a first training sample set; Training module: used to construct an initial cutting parameter prediction model, and to perform multiple rounds of iterative training on the initial cutting parameter prediction model using the first training sample set to obtain a target cutting parameter prediction model; Cutting parameter prediction module: acquires multiple temperatures to be measured, calls the target cutting parameter prediction model to predict each of the temperatures to be measured, and obtains multiple initial cutting process data sets. Each initial cutting process data set includes the temperature to be measured and the cutting process parameters corresponding to the temperature to be measured. Optimization module: Used to filter through multiple initial cutting process data sets using the SAC algorithm to obtain an optimized cutting process data set.

7. A cutting process parameter design device, characterized in that, The cutting process parameter design device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the cutting process parameter design device to perform the steps of the cutting process parameter design method as described in any one of claims 1-5.

8. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement each step of the cutting process parameter design method as described in any one of claims 1-5.

Images