Method and system for updating digital twin model of numerical control machine tool

Abstract

The invention relates to the technical field of numerical control machine tools, in particular to a numerical control machine tool digital twin model updating method and system. The method comprises the following steps: acquiring actual processing technology data; performing cutting process simulation according to the actual machining process data and the digital twin model of the numerical control machine tool to obtain cutting process data; performing tool wear state feature analysis on the cutting process data to obtain tool wear state data; updating a tool model according to the tool wear state data to obtain an updated tool model; according to the updated cutter model, dynamic contact state solving is carried out, and a dynamic contact solving result is obtained; and performing contact key data analysis on a dynamic contact solving result to obtain cutter-workpiece dynamic contact information. According to the method, the modeling precision and the simulation reliability are greatly improved, so that the digital twin model can reflect complex processing conditions more truly, and a reliable decision basis is provided for process optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of CNC machine tools, and in particular to a method and system for updating a digital twin model of a CNC machine tool. Background Art

[0002] Traditional digital twin models of CNC machine tools are mainly based on offline modeling with CAD / CAM software, and the visualization and optimization of the machining process are achieved through simulation. However, most of these models use idealized tool and workpiece geometry models, assuming that the tool is in its initial state, and ignoring the development of tool wear and changes in the dynamic contact state between the tool and the workpiece during the actual machining process. This static model processing method leads to the following disadvantages:

[0003] (1) As the cutting process progresses, the tool will inevitably wear and its geometry and size will change. The static model cannot reflect the dynamic changes of the tool wear state, resulting in large prediction deviations of important parameters such as cutting force and cutting temperature.

[0004] (2) The contact state between the tool and the workpiece changes dynamically in actual machining, affecting the magnitude and direction of the cutting force. Static models usually simplify this dynamic contact relationship and use constant contact conditions, which leads to a decrease in the accuracy of cutting force calculation.

[0005] (3) Due to the lack of consideration of factors such as tool wear and dynamic contact, the static model is difficult to accurately predict key indicators such as tool life and machining surface quality, and cannot provide reliable decision support for actual production.

[0006] (4) There are differences between the static model and the actual processing status, which makes it difficult for the optimization results to achieve the expected results in actual production, affecting the practical value of digital twin technology. Summary of the invention

[0007] Based on this, it is necessary to provide a method and system for updating the digital twin model of a CNC machine tool to solve at least one of the above technical problems.

[0008] To achieve the above purpose, a method for updating a digital twin model of a CNC machine tool comprises the following steps:

[0009] Step S1: Acquire actual processing data; simulate the cutting process according to the actual processing data and the digital twin model of the CNC machine tool to obtain cutting process data;

[0010] Step S2: Analyze the tool wear state characteristics of the cutting process data to obtain tool wear state data;

[0011] Step S3: updating the tool model according to the tool wear state data to obtain an updated tool model; solving the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyzing the contact key data on the dynamic contact solution result to obtain tool-workpiece dynamic contact information; correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient is corrected; calculating the cutting force components according to the cutting process data and the cutting force model after the cutting force coefficient is corrected, and outputting the cutting force to obtain corrected cutting force data;

[0012] Step S4: Use the tool wear status data, tool-workpiece contact information and corrected cutting force data to update the model parameters of the digital twin model to obtain an updated digital twin model to implement the CNC machine tool digital twin model update operation.

[0013] The present invention can integrate the real processing environment and conditions into the digital twin model by acquiring the actual processing technology data in real time and simulating the cutting process, providing a data basis closer to reality for subsequent analysis, and improving the accuracy and reliability of the analysis results. By analyzing the cutting process data, extracting the tool wear characteristics and predicting the tool wear state, the health of the tool can be monitored in real time, providing a decision basis for tool maintenance and replacement, and avoiding the decline in processing quality and production accidents caused by excessive tool wear. According to the predicted tool wear state, the tool model is updated, and dynamic contact analysis and cutting force model correction are performed, which can more accurately simulate the tool-workpiece interaction in the actual cutting process, improve the accuracy of cutting force prediction, and provide a more reliable basis for the subsequent digital twin model parameter update. Using the tool wear state, contact information and corrected cutting force data to update the parameters of the digital twin model, the model can be made to reflect the actual processing process more realistically, improve the prediction accuracy and reliability of the model, and provide more effective support for processing process monitoring, optimization and prediction. Therefore, the present invention provides a method for updating a digital twin model of a numerically controlled machine tool, which is used to solve the disadvantages caused by the static processing method of the model. By acquiring processing data in real time and updating the model, the digital twin model is synchronized with the actual processing status, greatly improving the modeling accuracy and simulation reliability. The impact of tool wear development and tool-workpiece dynamic contact on the cutting process is taken into account, so that the digital twin model can more realistically reflect complex processing conditions and provide a reliable decision-making basis for process optimization.

[0014] Preferably, step S1 comprises the following steps:

[0015] Step S11: extracting a workpiece model and a tool model from the digital twin model of the CNC machine tool, and performing model preprocessing to obtain model data;

[0016] Step S12: Acquire actual processing data; set process parameters according to the actual processing data to obtain process parameters;

[0017] Step S13: using the model data as input data and the process parameters as boundary conditions, finite element analysis technology is used to simulate the cutting process to obtain simulation results;

[0018] Step S14: Analyze key cutting data on the simulation results to obtain cutting process data.

[0019] The present invention extracts the workpiece model and the tool model from the digital twin model of the numerically controlled machine tool, and performs model preprocessing, so as to obtain model data that meets the requirements of the finite element analysis software, and provide a basis for the subsequent cutting process simulation. Model preprocessing operations, such as removing model defects, gap repairing, and model simplification, can improve the model quality, reduce the amount of calculation, and improve the simulation efficiency. By obtaining the actual processing technology data and setting the process parameters, the real processing conditions can be introduced into the cutting process simulation to improve the accuracy and reliability of the simulation results. Setting the process parameters according to the actual processing technology data can better reflect the changing laws of the cutting force, cutting temperature, etc. in the actual processing process. By using the finite element analysis technology to simulate the cutting process, the distribution of the physical fields such as stress, strain, and temperature of the tool and the workpiece in the cutting process, as well as the changing laws of cutting parameters such as cutting force and cutting temperature, can be obtained, which provides a data basis for the subsequent tool wear state analysis, cutting force model correction, and digital twin model update. By analyzing the key cutting data of the simulation results, key information related to tool wear, cutting force, etc. can be extracted, such as the components of cutting force in three directions, the morphological characteristics of the chips, the temperature distribution in the tool contact area, etc., providing data support for subsequent analysis and model updates.

[0020] Preferably, step S2 comprises the following steps:

[0021] Step S21: extracting tool wear features from cutting process data to obtain tool wear features;

[0022] Step S22: Perform machine learning training using tool wear characteristics and a pre-selected wear model to obtain a wear prediction model;

[0023] Step S23: using the wear prediction model to predict tool wear according to the cutting process data to obtain tool wear status data.

[0024] The present invention can convert the original cutting force, cutting temperature, vibration signal and other data into characteristic indicators that can better reflect the tool wear state, such as the main frequency band energy of the cutting force, the time-frequency domain energy distribution of the cutting temperature, the energy entropy of the inherent modal function of the vibration signal, etc., by extracting the tool wear characteristics of the cutting process data, so as to provide more direct and effective input information for the subsequent tool wear state prediction. By using the extracted tool wear characteristics and the pre-prepared tool wear state label data for machine learning training, a mapping relationship between the tool wear characteristics and the tool wear state can be established, and a prediction model that can accurately predict the tool wear state according to the tool wear characteristics can be obtained. By using the trained tool wear prediction model to predict the new cutting process data, the wear state of the current tool can be evaluated in real time, such as judging whether the tool is in a normal state, a slightly worn state or a severely worn state, providing a basis for the subsequent tool model update, cutting force model correction and digital twin model parameter update.

[0025] Preferably, step S3 comprises the following steps:

[0026] Step S31: extracting a tool model and a workpiece model from the digital twin model to obtain a tool model and a workpiece model;

[0027] Step S32: updating the tool model according to the tool wear state data to obtain an updated tool model;

[0028] Step S33: importing the updated tool model and workpiece model into the contact mechanics analysis software to construct the contact model, and applying the cutting force according to the cutting process data to obtain the contact model;

[0029] Step S34: solving the dynamic contact state of the contact model to obtain a dynamic contact solution result;

[0030] Step S35: analyzing the key contact data of the dynamic contact solution results to obtain tool-workpiece dynamic contact information;

[0031] Step S36: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a corrected cutting force coefficient and a cutting force model after the cutting force coefficient is corrected;

[0032] Step S37: Calculate the cutting force components according to the cutting process data and the cutting force model corrected by the cutting force coefficient to obtain the corrected cutting force components;

[0033] Step S38: synthesize and output the corrected cutting force components to obtain corrected cutting force data.

[0034] The present invention provides a model basis for subsequent tool model update and contact analysis by extracting tool model and workpiece model from digital twin model. Updating tool model according to tool wear state data can make tool model closer to the wear state in actual processing process, and improve the accuracy of subsequent contact analysis and cutting force model correction. Import updated tool model and workpiece model into contact mechanics analysis software, and apply cutting force according to cutting process data to build a contact model that is more in line with actual processing conditions, so as to provide a basis for subsequent dynamic contact analysis. By solving the dynamic contact state of contact model, the contact state change of tool and workpiece in the whole cutting process, such as contact area, contact pressure, friction force, etc., can be obtained, so as to provide more detailed data support for subsequent contact analysis and cutting force model correction. Analyze the dynamic contact solution results, extract tool-workpiece dynamic contact information, such as contact length, contact width, average contact pressure, average friction coefficient, etc., to provide more direct input parameters for subsequent cutting force model correction. Correcting the pre-selected initial cutting force model based on the tool-workpiece dynamic contact information can make the cutting force model more accurately reflect the impact of tool wear and dynamic contact effect on cutting force, and improve the accuracy of cutting force prediction. According to the corrected cutting force model and cutting process data, the corrected cutting force components are calculated, which can more accurately predict the magnitude of cutting force in all directions, providing a more reliable basis for subsequent tool wear monitoring, processing parameter optimization, etc. The corrected cutting force components are synthesized to obtain the final cutting force data, which can more comprehensively reflect the changing law of cutting force during the cutting process, and provide more accurate input information for the subsequent digital twin model parameter update.

[0035] Preferably, step S32 includes the following steps:

[0036] Step S321: extracting the tool edge wear and flank wear from the tool wear state data to obtain tool edge and flank wear parameters;

[0037] Step S322: mapping the wear parameters of the cutting edge and the flank surface to corresponding positions of the cutting tool model according to the cutting tool geometry of the cutting tool model, and obtaining the mapped wear amount;

[0038] Step S323: performing tool geometry deformation processing on the tool model according to the mapped wear amount to obtain a deformed tool model;

[0039] Step S324: performing model smoothing processing on the deformed tool model to obtain an updated tool model.

[0040] The present invention can convert the tool wear state data into specific and quantifiable wear parameters by extracting the tool blade wear and the back face wear from the tool wear state data, thereby providing an accurate modification basis for the subsequent tool model update. According to the tool geometry of the tool model, the blade and back face wear parameters are mapped to the corresponding positions of the tool model, which can ensure that the tool wear is accurately applied to the corresponding parts of the tool model, avoiding model distortion caused by inaccurate wear application positions. According to the mapped wear, the tool model is subjected to tool geometry deformation processing, which can intuitively reflect the influence of tool wear on the tool geometry, such as blunting of the blade, etc., so that the updated tool model is closer to the actual tool shape after wear. Model smoothing processing is performed on the deformed tool model, which can eliminate defects such as sharp corners and depressions generated during the deformation process, make the surface of the tool model smoother and more continuous, improve the model quality, and facilitate subsequent contact analysis and simulation calculations.

[0041] Preferably, step S34 includes the following steps:

[0042] Step S341: defining the tool cutting motion according to the actual machining process data, and defining the workpiece clamping rotation, to obtain the tool cutting motion data and the workpiece clamping rotation data;

[0043] Step S342: setting the time step according to the tool cutting motion data and the workpiece clamping rotation data to obtain the time step;

[0044] Step S343: performing model dynamic processing on the contact model according to the time step, and setting initial conditions to obtain an initial state model;

[0045] Step S344: using contact mechanics analysis software to perform iterative solution of dynamic contact between the tool and the workpiece according to the initial state model to obtain results at each time step;

[0046] Step S345: Integrate and analyze the results of each time step to obtain dynamic contact solution results.

[0047] The present invention can introduce the real processing process into the simulation analysis by defining the tool cutting motion and the workpiece clamping rotation according to the actual processing technology data, so that the simulation result is closer to the actual situation and the accuracy and reliability of the simulation are improved. According to the tool cutting motion data and the workpiece clamping rotation data, the appropriate time step is set to ensure the numerical stability of the simulation process, avoid the distortion of the simulation result caused by the time step being too large, and also avoid the excessive amount of calculation caused by the time step being too small. According to the time step, the contact model is dynamically processed, and the initial position, speed and other information of the tool and the workpiece are set as the initial conditions, which can provide a complete model and initial state for the subsequent dynamic contact solution, and ensure the smooth progress of the simulation calculation. The initial state model is dynamically contacted and solved using the contact mechanics analysis software, and the contact state of the tool and the workpiece at each time step, such as the contact area, contact pressure, friction, etc., as well as the displacement, speed, acceleration and other information of the tool and the workpiece, can be obtained, providing a detailed data basis for the subsequent contact analysis. By integrating and analyzing the results of each time step, we can obtain the changing pattern of the contact state between the tool and the workpiece during the entire cutting process, such as the changing trend of the contact area, the peak value and distribution of the contact pressure, as well as the motion trajectory and speed change of the tool and workpiece, providing data support for subsequent cutting force model correction, tool wear monitoring, etc.

[0048] Preferably, step S36 includes the following steps:

[0049] Step S361: extracting cutting dynamic contact information from tool-workpiece contact information according to a pre-selected initial cutting force model to obtain cutting dynamic contact parameters; performing dynamic cutting process modeling according to the cutting dynamic contact parameters to obtain a dynamic cutting process model;

[0050] Step S362: extracting dynamic cutting parameters from the dynamic cutting process model to obtain dynamic cutting parameters;

[0051] Step S363: quantifying the contact state of the tool-workpiece contact information to obtain quantified contact state data;

[0052] Step S364: analyzing the influence mechanism of tool wear on cutting force coefficient according to the dynamic cutting process model to obtain wear influence parameters;

[0053] Step S365: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the dynamic cutting parameters, the quantified contact state data and the wear influence parameters to obtain a corrected cutting force coefficient;

[0054] Step S366: Generate a cutting force model according to the corrected cutting force coefficient to obtain a cutting force model after the cutting force coefficient is corrected.

[0055] The present invention can convert tool-workpiece contact information into calculable model parameters by extracting cutting dynamic contact parameters and performing dynamic cutting process modeling, thereby providing a basis for subsequent correction of cutting force coefficient. Extracting dynamic cutting parameters, such as dynamic cutting force, dynamic chip thickness, etc., from the dynamic cutting process model can more accurately describe the dynamic changes in the cutting process and improve the accuracy of cutting force coefficient correction. Quantifying the contact state of tool-workpiece contact information can convert complex contact states into quantifiable data, such as light contact, moderate contact, heavy contact, etc., which is convenient for subsequent analysis and processing. By analyzing the influence mechanism of tool wear on the cutting force coefficient, wear influence parameters, such as the relationship between wear amount and cutting force coefficient, can be obtained, providing a theoretical basis for cutting force coefficient correction. According to the dynamic cutting parameters, the quantified contact state data and the wear influence parameters, the initial cutting force model is corrected, and a cutting force coefficient that is more in line with the actual processing situation can be obtained, thereby improving the accuracy of cutting force prediction. According to the corrected cutting force coefficient, a new cutting force model is generated, which can more accurately predict the cutting force under different processing conditions and provide a more reliable basis for tool wear monitoring, processing parameter optimization, etc.

[0056] Preferably, step S37 includes the following steps:

[0057] Step S371: extracting cutting geometric parameters from the cutting process data to obtain cutting geometric parameters, wherein the cutting geometric parameters include cutting thickness parameters, cutting width parameters, tool rake angle parameters, and tool clearance angle parameters;

[0058] Step S372: performing cutting force coefficient interpolation calculation on the corrected cutting force coefficient according to the cutting thickness parameter and the cutting width parameter to obtain the interpolated cutting force coefficient;

[0059] Step S373: performing cutting force model correction of cutting force tool geometric parameters on the cutting force model after cutting force coefficient correction according to the tool rake angle parameter and the tool clearance angle parameter, to obtain the cutting force model after cutting force tool geometric parameter correction;

[0060] Step S374: Calculate the cutting force components according to the interpolated cutting force coefficient, the cutting geometric parameters, and the cutting force model corrected by the tool geometric parameters to obtain corrected cutting force components.

[0061] The present invention can obtain important geometric factors that affect the cutting force, such as cutting thickness, width and tool angle, by extracting cutting geometry parameters, and provide necessary parameter input for subsequent cutting force calculation. By interpolating the corrected cutting force coefficient according to the cutting thickness and width, a cutting force coefficient that matches the current cutting conditions can be obtained, thereby improving the accuracy of the cutting force calculation. By correcting the corrected cutting force model according to the front angle and back angle of the tool, the accuracy of the cutting force model can be further improved, so that it can more accurately reflect the influence of the tool geometry on the cutting force. By calculating the cutting force components according to the interpolated cutting force coefficient, cutting geometry parameters and the corrected cutting force model, a more accurate cutting force prediction result can be obtained, providing a more reliable basis for subsequent tool wear monitoring, processing parameter optimization and the like.

[0062] Preferably, step S4 comprises the following steps:

[0063] Step S41: extracting cutting force characteristics from tool wear state data, tool-workpiece contact information and corrected cutting force data to obtain cutting force characteristics;

[0064] Step S42: using the cutting force characteristics to identify model parameters of the digital twin model to obtain model parameters; using the model parameters to update the digital twin model to obtain an updated digital twin model;

[0065] Step S43: Use the updated digital twin model to simulate the new processing process to obtain simulation results; obtain new actual processing process data; compare the simulation results and the new actual processing process data, and perform model verification and evaluation to obtain model evaluation results;

[0066] Step S44: Deploy the updated digital twin model to the actual processing system according to the model evaluation results to implement the CNC machine tool digital twin model update operation.

[0067] The present invention can fuse multi-source data into characteristic indicators that can better reflect the parameter changes of the digital twin model by extracting the tool wear state data, tool-workpiece contact information and the cutting force characteristics in the corrected cutting force data, and provide more comprehensive and accurate input information for subsequent model parameter identification. The parameters of the digital twin model are identified by using the extracted cutting force characteristics, and the digital twin model is updated by using the identified parameters, so that the parameters of the digital twin model can be more consistent with the actual processing process, and the simulation accuracy and prediction ability of the model can be improved. The updated digital twin model is used to simulate the new processing process, and the simulation results are compared and analyzed with the actual processing process data to verify whether the updated model can accurately predict the actual processing process, and the model can be further optimized and improved according to the evaluation results. The updated digital twin model is deployed to the actual processing system according to the model evaluation results, and the model can be used to monitor, optimize and predict the processing process in real time, such as predicting the tool wear state, optimizing cutting parameters, predicting processing quality, etc., thereby improving processing efficiency and product quality.

[0068] Preferably, the present invention also provides an updating system for a digital twin model of a CNC machine tool, which is used to execute the updating method for the digital twin model of a CNC machine tool as described above, and the updating system for the digital twin model of a CNC machine tool comprises:

[0069] The virtual cutting simulation module is used to obtain the actual processing data; the cutting process is simulated based on the actual processing data and the digital twin model of the CNC machine tool to obtain the cutting process data;

[0070] Tool wear prediction module, used to analyze tool wear state characteristics of cutting process data to obtain tool wear state data;

[0071] A dynamic cutting parameter correction module is used to update the tool model according to the tool wear state data to obtain an updated tool model; solve the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyze the contact key data of the dynamic contact solution result to obtain tool-workpiece dynamic contact information; perform cutting force coefficient correction on the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient correction; calculate the cutting force component according to the cutting process data and the cutting force model after the cutting force coefficient correction, and output the cutting force to obtain corrected cutting force data;

[0072] The digital twin model update module is used to update the model parameters of the digital twin model using tool wear status data, tool-workpiece contact information and corrected cutting force data to obtain an updated digital twin model to realize the digital twin model update operation of the CNC machine tool.

[0073] The present invention can integrate the real processing environment and conditions into the digital twin model by acquiring the actual processing technology data in real time and simulating the cutting process, providing a data basis closer to reality for subsequent analysis, and improving the accuracy and reliability of the analysis results. By analyzing the cutting process data, extracting the tool wear characteristics and predicting the tool wear state, the health of the tool can be monitored in real time, providing a decision basis for tool maintenance and replacement, and avoiding the decline in processing quality and production accidents caused by excessive tool wear. According to the predicted tool wear state, the tool model is updated, and dynamic contact analysis and cutting force model correction are performed, which can more accurately simulate the tool-workpiece interaction in the actual cutting process, improve the accuracy of cutting force prediction, and provide a more reliable basis for the subsequent digital twin model parameter update. Using the tool wear state, contact information and corrected cutting force data to update the parameters of the digital twin model, the model can be made to reflect the actual processing process more realistically, improve the prediction accuracy and reliability of the model, and provide more effective support for processing process monitoring, optimization and prediction. Therefore, the present invention provides a method for updating a digital twin model of a numerically controlled machine tool, which is used to solve the disadvantages caused by the static processing method of the model. By acquiring processing data in real time and updating the model, the digital twin model is synchronized with the actual processing status, greatly improving the modeling accuracy and simulation reliability. The impact of tool wear development and tool-workpiece dynamic contact on the cutting process is taken into account, so that the digital twin model can more realistically reflect complex processing conditions and provide a reliable decision-making basis for process optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic diagram of the steps of a method for updating a digital twin model of a CNC machine tool;

[0075] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0076] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0077] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0078] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0079] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0080] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0081] To achieve this, please refer to Figures 1 to 3 , a method for updating a digital twin model of a CNC machine tool comprises the following steps:

[0082] Step S1: Acquire actual processing data; simulate the cutting process according to the actual processing data and the digital twin model of the CNC machine tool to obtain cutting process data;

[0083] Step S2: Analyze the tool wear state characteristics of the cutting process data to obtain tool wear state data;

[0084] Step S3: updating the tool model according to the tool wear state data to obtain an updated tool model; solving the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyzing the contact key data on the dynamic contact solution result to obtain tool-workpiece dynamic contact information; correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient is corrected; calculating the cutting force components according to the cutting process data and the cutting force model after the cutting force coefficient is corrected, and outputting the cutting force to obtain corrected cutting force data;

[0085] Step S4: Use the tool wear status data, tool-workpiece contact information and corrected cutting force data to update the model parameters of the digital twin model to obtain an updated digital twin model to implement the CNC machine tool digital twin model update operation.

[0086] In the embodiment of the present invention, reference Figure 1The above is a schematic flow chart of the steps of a method for updating a digital twin model of a CNC machine tool according to the present invention. In this example, the method for updating a digital twin model of a CNC machine tool includes the following steps:

[0087] Step S1: Acquire actual processing data; simulate the cutting process according to the actual processing data and the digital twin model of the CNC machine tool to obtain cutting process data;

[0088] In an embodiment of the present invention, the sensor system of the CNC machine tool collects the process data such as the spindle speed, feed speed, cutting depth, etc. in the actual processing process in real time, and determines the specific process parameters such as the cutting speed and feed per revolution according to the collected data and the properties of the processed material. The target workpiece model and the tool model are called from the digital twin model database, and the model is preprocessed and converted into a format that can be used for finite element analysis. The preprocessed model data, process parameters, and material parameters are input into the finite element analysis software, the boundary conditions of the cutting process are defined, the meshing and solution parameter setting are performed, and finally the calculation task is submitted to simulate the cutting process. After the simulation, key data such as cutting force and cutting temperature are extracted and output as cutting process data.

[0089] Step S2: Analyze the tool wear state characteristics of the cutting process data to obtain tool wear state data;

[0090] In the embodiment of the present invention, features related to tool wear are extracted from cutting process data, and the features are fused to form a tool wear feature vector. A support vector machine is selected as a pre-selected wear model, and the extracted tool wear feature vector and the pre-prepared tool wear state label data are divided into a training set and a test set respectively. The kernel function type, penalty coefficient and kernel function parameter of the support vector machine model are optimized using a grid search method, and finally the performance of the trained tool wear prediction model is evaluated using the test set. The new cutting process data is input into the trained tool wear prediction model to obtain the prediction result of the tool wear state, which is output as the tool wear state data.

[0091] Step S3: updating the tool model according to the tool wear state data to obtain an updated tool model; solving the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyzing the contact key data on the dynamic contact solution result to obtain tool-workpiece dynamic contact information; correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient is corrected; calculating the cutting force components according to the cutting process data and the cutting force model after the cutting force coefficient is corrected, and outputting the cutting force to obtain corrected cutting force data;

[0092] In the embodiment of the present invention, the tool model and workpiece model corresponding to the current processing task are extracted respectively. According to the tool wear state data, the tool model is modified by using CATIA software, and the corresponding volume of material is removed along the wear direction at the tool tip and the back face area, respectively, to generate a new tool shape, and smoothing is performed to obtain an updated tool model. The updated tool model and workpiece model are imported into the static analysis module of ANSYS Workbench, the material properties and contact relationship are defined, and the cutting force load is applied to complete the establishment of the contact model. According to the tool trajectory and cutting parameters in the actual processing process, the contact model is converted into a transient dynamic analysis model, the tool motion trajectory, speed, acceleration and other parameters are defined, the time step and solution time are set, and the initial conditions are defined, and the solver is started to solve the dynamic contact state. The key data of the tool-workpiece contact area is extracted from the solution result, and the tool-workpiece dynamic contact information is obtained by analysis and calculation. The classic Merchant cutting force model is selected as the initial cutting force model, and the cutting force coefficient in the Merchant cutting force model is corrected according to the extracted tool-workpiece dynamic contact information to obtain the corrected cutting force coefficient and cutting force model. According to the cutting process data and the modified cutting force model, the modified cutting force components are calculated, and the modified cutting force data are calculated by using the vector synthesis method.

[0093] Step S4: using the tool wear state data, tool-workpiece contact information, and corrected cutting force data to update the model parameters of the digital twin model, and obtain an updated digital twin model to implement the digital twin model update operation of the CNC machine tool;

[0094] In an embodiment of the present invention, tool wear characteristics are extracted from tool wear state data, contact characteristics are extracted from tool-workpiece contact information, and cutting force characteristics are extracted from corrected cutting force data, and the three are fused to form a cutting force characteristic vector. The parameters to be identified in the digital twin model are set as parameters to be optimized, and parameter identification is performed. The cutting force characteristic vector is used as the input of the objective function, and the error between the simulation output of the digital twin model and the actual cutting force characteristic vector is minimized by iteratively optimizing the parameters to be optimized. The digital twin model is updated using the optimized parameters to obtain an updated digital twin model. The new machining process is simulated using the updated digital twin model, and compared and analyzed with the actual machining process data to verify and evaluate the model. If the model evaluation results meet the requirements, the updated digital twin model is deployed to the actual machining system to realize machining process monitoring, optimization, and prediction operations.

[0095] Preferably, step S1 comprises the following steps:

[0096] Step S11: extracting a workpiece model and a tool model from the digital twin model of the CNC machine tool, and performing model preprocessing to obtain model data;

[0097] Step S12: Acquire actual processing data; set process parameters according to the actual processing data to obtain process parameters;

[0098] Step S13: using the model data as input data and the process parameters as boundary conditions, finite element analysis technology is used to simulate the cutting process to obtain simulation results;

[0099] Step S14: Analyze key cutting data on the simulation results to obtain cutting process data.

[0100] In an embodiment of the present invention, the target workpiece model and the tool model are first called from the digital twin model database of the CNC machine tool. For the acquired workpiece model, the STL file format is used for storage, and the model is repaired using the FreeCAD software, including operations such as removing model defects, gap repairing, and model simplification, and finally the model is converted into a STEP file format that can be used for finite element analysis. For the acquired tool model, the standard tool library is used for calling, and the tool model is three-dimensionally modeled using the CATIA software to define the geometric parameters of the tool, such as tool radius, blade length, tool angle, etc., and finally the model is saved in the STEP file format. The process data in the actual processing process is collected in real time by the sensor system of the CNC machine tool, including spindle speed, feed speed, cutting depth, and back cutting amount. According to the actual processing technology data collected, combined with the properties of the processed material and the processing requirements, the specific process parameters are determined. For example, the cutting speed is calculated according to the spindle speed and the tool diameter, and the feed per revolution is calculated according to the feed speed and the spindle speed, and these parameters are used as input conditions for the simulation of the cutting process. The workpiece model and tool model processed in step S11 are imported into the ABAQUS finite element analysis software, and the boundary conditions of the cutting process, such as the motion trajectory of the tool, cutting speed, cutting depth, etc., are defined according to the process parameters set in step S12. At the same time, the constitutive model of the material, such as the Johnson-Cook model, is defined, and material parameters, such as elastic modulus, Poisson's ratio, yield strength, etc., are set. After the model and boundary conditions are set, meshing is performed, the model is discretized into a finite number of units, and solution parameters, such as time step, solution accuracy, etc., are set. Finally, the calculation task is submitted to simulate the cutting process to obtain simulation results, such as cutting force, cutting temperature, tool wear, etc. The key data of the cutting process, such as node displacement, stress, strain, and temperature data in the contact area between the tool and the workpiece, are extracted from the simulation results of step S13. The extracted data is processed and analyzed using Python scripts, such as calculating the components of the cutting force in three directions, calculating the morphological characteristics of the chips, and calculating the temperature distribution in the contact area of ​​the tool. These analysis results will be used as cutting process data for subsequent tool wear state analysis and updating of the digital twin model.

[0101] Preferably, step S2 comprises the following steps:

[0102] Step S21: extracting tool wear features from cutting process data to obtain tool wear features;

[0103] Step S22: Perform machine learning training using tool wear characteristics and a pre-selected wear model to obtain a wear prediction model;

[0104] Step S23: using the wear prediction model to predict tool wear according to the cutting process data to obtain tool wear status data.

[0105] In an embodiment of the present invention, features related to tool wear, such as cutting force, cutting temperature and vibration signals, are extracted from the cutting process data obtained in step S14. The cutting force signal is analyzed in the frequency domain by using fast Fourier transform (FFT) to extract the energy and spectrum characteristic parameters of its main frequency band. The cutting temperature signal is analyzed in the time-frequency domain by using wavelet packet transform (WPT) to extract the energy characteristics of different frequency bands. The vibration signal is decomposed by using empirical mode decomposition (EMD) to extract characteristic parameters such as energy entropy and permutation entropy of its main intrinsic mode function (IMF). The extracted multi-source signal features are fused to form a tool wear feature vector for subsequent machine learning model training. Support vector machine (SVM) is selected as the pre-selected wear model, and the tool wear features extracted in step S21 and the corresponding tool wear state label data are used for model training. First, the tool wear feature vector and label data are divided into a training set and a test set. Then, the kernel function type, penalty coefficient and kernel function parameters of the SVM model are optimized by using the grid search method to minimize the cross-validation error of the model on the training set. Finally, the optimized SVM model is used as the tool wear prediction model, and its prediction performance is evaluated using the test set, such as calculating the prediction accuracy, recall rate, and F1 value of the model. The new cutting process data is input into the tool wear prediction model trained in step S22 for prediction. First, the new cutting process data is subjected to the same feature extraction operation as step S21 to obtain a new tool wear feature vector. Then, the feature vector is input into the SVM model for prediction to obtain the prediction result of the tool wear state, such as the tool being in a normal state, a slightly worn state, a severely worn state, etc. The prediction result is used as the tool wear state data for subsequent tool model updates and digital twin model parameter updates.

[0106] Preferably, step S3 comprises the following steps:

[0107] Step S31: extracting a tool model and a workpiece model from the digital twin model to obtain a tool model and a workpiece model;

[0108] Step S32: updating the tool model according to the tool wear state data to obtain an updated tool model;

[0109] Step S33: importing the updated tool model and workpiece model into the contact mechanics analysis software to construct the contact model, and applying the cutting force according to the cutting process data to obtain the contact model;

[0110] Step S34: solving the dynamic contact state of the contact model to obtain a dynamic contact solution result;

[0111] Step S35: analyzing the key contact data of the dynamic contact solution results to obtain tool-workpiece dynamic contact information;

[0112] Step S36: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a corrected cutting force coefficient and a cutting force model after the cutting force coefficient is corrected;

[0113] Step S37: Calculate the cutting force components according to the cutting process data and the cutting force model corrected by the cutting force coefficient to obtain the corrected cutting force components;

[0114] Step S38: synthesize and output the corrected cutting force components to obtain corrected cutting force data.

[0115] As an example of the present invention, refer to Figure 2 As shown, in this example, step S3 includes:

[0116] Step S31: extracting a tool model and a workpiece model from the digital twin model to obtain a tool model and a workpiece model;

[0117] In the embodiment of the present invention, the tool model and workpiece model corresponding to the current processing task are respectively extracted from the digital twin model database of the CNC machine tool. The tool model is stored in the STEP file format, including information such as the geometric shape and size parameters of the tool. The workpiece model is also stored in the STEP file format, including information such as the geometric shape, size parameters and processing area of ​​the workpiece.

[0118] Step S32: updating the tool model according to the tool wear state data to obtain an updated tool model;

[0119] In the embodiment of the present invention, the tool model extracted in step S31 is updated according to the tool wear state data obtained in step S23. For example, if the tool wear state data shows that the tool tip wear is 0.1 mm, the tool model is modified using the 3D modeling software CATIA, and 0.1 mm of material is removed from the tool tip along the tool wear direction to generate a new tool tip shape. Then, the modified tool model is smoothed to eliminate sharp corners and discontinuous geometric features to obtain an updated tool model.

[0120] Step S33: importing the updated tool model and workpiece model into the contact mechanics analysis software to construct the contact model, and applying the cutting force according to the cutting process data to obtain the contact model;

[0121] In the embodiment of the present invention, the tool model updated in step S32 and the workpiece model extracted in step S31 are imported into the static analysis module of ANSYS Workbench. The material properties of the tool and the workpiece are defined in the software, such as elastic modulus, poise ratio, yield strength, etc. According to the actual processing parameters, the contact relationship between the tool and the workpiece is set, such as contact type, contact stiffness, friction coefficient, etc. According to the cutting force data obtained in step S14, the corresponding cutting force load is applied to the cutting edge of the tool, and the direction and magnitude of the load are set.

[0122] Step S34: solving the dynamic contact state of the contact model to obtain a dynamic contact solution result;

[0123] In the embodiment of the present invention, in ANSYS Workbench, the contact model established in step S33 is converted into a transient dynamic analysis model. According to the tool trajectory and cutting parameters in the actual processing process, the tool motion trajectory, speed, acceleration and other parameters are defined. The time step and solution time are set, and the initial conditions, such as initial speed, initial displacement, etc., are defined. The solver is started to solve the dynamic contact state, and the displacement, stress, strain and other data of the tool and the workpiece at each time step are obtained.

[0124] Step S35: analyzing the key contact data of the dynamic contact solution results to obtain tool-workpiece dynamic contact information;

[0125] In the embodiment of the present invention, key data of the tool-workpiece contact area are extracted from the dynamic contact solution result of step S34, such as contact area, contact pressure, contact time, relative slip velocity, etc. The extracted data are processed and analyzed using a Python script, such as calculating the average contact pressure, maximum contact pressure, contact area change rate, and other indicators of the tool-workpiece contact area to obtain tool-workpiece dynamic contact information.

[0126] Step S36: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a corrected cutting force coefficient and a cutting force model after the cutting force coefficient is corrected;

[0127] In the embodiment of the present invention, the classic Merchant cutting force model is selected as the initial cutting force model. According to the tool-workpiece dynamic contact information obtained in step S35, such as contact area, contact pressure, relative sliding speed, etc., the cutting force coefficient in the Merchant cutting force model is corrected. For example, the relationship curve between the contact area and the cutting force coefficient is fitted using the least squares method to obtain a corrected cutting force coefficient. The corrected cutting force coefficient is substituted into the Merchant cutting force model to obtain a cutting force model after the cutting force coefficient is corrected.

[0128] Step S37: Calculate the cutting force components according to the cutting process data and the cutting force model corrected by the cutting force coefficient to obtain the corrected cutting force components;

[0129] In the embodiment of the present invention, the corrected cutting force components are calculated based on the cutting process data acquired in step S14, such as cutting thickness, cutting width, tool rake angle, tool clearance angle, etc., and the corrected cutting force model in step S36. For example, the cutting thickness, cutting width, tool rake angle, tool clearance angle, and the corrected cutting force coefficient are substituted into the corrected Merchant cutting force model to calculate the cutting force components in the cutting direction, feed direction, and back feed direction.

[0130] Step S38: synthesizing and outputting the corrected cutting force components to obtain corrected cutting force data;

[0131] In the embodiment of the present invention, according to the cutting force component calculated in step S37, the corrected cutting force is calculated by using a vector synthesis method. The corrected cutting force data is output to the digital twin model for updating the parameters of the digital twin model.

[0132] Preferably, step S32 includes the following steps:

[0133] Step S321: extracting the tool edge wear and flank wear from the tool wear state data to obtain tool edge and flank wear parameters;

[0134] Step S322: mapping the wear parameters of the cutting edge and the flank surface to corresponding positions of the cutting tool model according to the cutting tool geometry of the cutting tool model, and obtaining the mapped wear amount;

[0135] Step S323: performing tool geometry deformation processing on the tool model according to the mapped wear amount to obtain a deformed tool model;

[0136] Step S324: performing model smoothing processing on the deformed tool model to obtain an updated tool model.

[0137] In the embodiment of the present invention, the predicted values ​​of the tool blade wear and the flank wear are read from the tool wear state data obtained in step S23. For example, the tool wear state data includes information such as the tool wear state type, the blade wear and the flank wear. By parsing the data, the blade wear is obtained to be 0.2 mm and the flank wear is 0.15 mm. According to the geometric information of the tool model, the blade wear and the flank wear obtained in step S321 are mapped to the corresponding positions of the tool model. The tool model is opened using the three-dimensional modeling software CATIA, and the area range of the blade and the flank is determined according to the tool geometry and size information. Then, according to the blade wear and the flank wear, the corresponding offset distances are defined on the blade area and the flank area, for example, the blade area is offset outward by 0.2 mm along the tool wear direction, and the flank area is offset outward by 0.15 mm in the direction perpendicular to the flank. According to the wear amount mapped in step S322, the tool model is subjected to geometric deformation processing. By using the free deformation function of the CATIA software, the nodes of the blade area and the back face area are moved according to the offset distance defined in step S322, so that the blade and back face of the tool model produce corresponding wear deformation. For example, the nodes of the blade area are moved outward by 0.2mm in the direction of tool wear, and the nodes of the back face area are moved outward by 0.15mm in the direction perpendicular to the back face. The deformed tool model in step S323 is smoothed. By using the surface smoothing function of the CATIA software, the deformed blade area and back face area are smoothed to eliminate defects such as sharp corners and depressions generated during the deformation process, so as to make the model surface smoother and continuous. For example, the deformed blade area and back face area are fitted with a cubic spline curve to generate a smooth surface, and a smooth transition is made with other parts of the tool model to finally obtain an updated tool model.

[0138] Preferably, step S34 includes the following steps:

[0139] Step S341: defining the tool cutting motion according to the actual machining process data, and defining the workpiece clamping rotation, to obtain the tool cutting motion data and the workpiece clamping rotation data;

[0140] Step S342: setting the time step according to the tool cutting motion data and the workpiece clamping rotation data to obtain the time step;

[0141] Step S343: performing model dynamic processing on the contact model according to the time step, and setting initial conditions to obtain an initial state model;

[0142] Step S344: using contact mechanics analysis software to perform iterative solution of dynamic contact between the tool and the workpiece according to the initial state model to obtain results at each time step;

[0143] Step S345: Integrate and analyze the results of each time step to obtain dynamic contact solution results.

[0144] In an embodiment of the present invention, the motion module of ANSYS Workbench is used to define the cutting motion trajectory of the tool according to the actual processing data obtained in step S12, such as spindle speed, feed speed, cutting depth, tool path, etc. For example, according to the tool path information, the motion trajectory curve of the tool center is generated using an interpolation algorithm, and the rotational motion of the tool is defined in combination with the spindle speed. At the same time, the clamping and rotation of the workpiece are defined according to the clamping method and processing requirements of the workpiece. For example, if the workpiece is clamped by a three-jaw chuck, a three-jaw chuck model is created in ANSYS Workbench, and it is assembled with the workpiece model. Constrain; if the workpiece needs to be indexed and rotated during the processing, the rotational motion of the workpiece is defined according to the indexing angle and time node. According to the tool cutting motion data and the workpiece clamping and rotation data defined in step S341, the maximum motion speed and the minimum geometric feature size in the simulation process are analyzed, such as the cutting edge radius of the tool, the chamfer size on the workpiece, etc. According to the Courant-Friedrichs-Lewy (CFL) condition, the simulation time step is calculated to ensure the numerical stability of the simulation process. For example, assuming that the radius of the tool cutting edge is 0.8 mm and the maximum cutting speed during the simulation is 200 mm / min, the time step can be set to 0.02 ms. According to the time step determined in step S342, the contact model established in step S33 is converted into a transient dynamic analysis model in ANSYS Workbench. Set the simulation time range, such as 0-100 ms, and discretize the simulation time into a series of time steps according to the time step. Define initial conditions, such as the initial position, velocity, acceleration, etc. of the tool and the workpiece. For example, set the tool to be at the starting point of cutting at the initial moment, the initial velocity to 0, and the workpiece to be stationary at the initial moment. Use the transient dynamic analysis module of ANSYS Workbench to perform iterative solution of the dynamic contact between the tool and the workpiece on the initial state model established in step S343. At each time step, the solver calculates the displacement, velocity, acceleration, stress, strain and other physical quantities of the tool and the workpiece according to the motion state, contact relationship and material properties of the tool and the workpiece. Integrate and analyze the results of each time step obtained in step S344. For example, the contact area, contact pressure, friction and other data of the tool and workpiece at each time step can be extracted, and the corresponding curve graphs can be drawn to analyze the change of the tool-workpiece contact state over time. At the same time, the stress and strain distribution cloud graphs of the tool and workpiece at a specific time point can be extracted to analyze the mechanical behavior and material removal mechanism during the cutting process. These analysis results are integrated into tool-workpiece dynamic contact information for subsequent cutting force model correction and digital twin model update.

[0145] Preferably, step S36 includes the following steps:

[0146] Step S361: extracting cutting dynamic contact information from tool-workpiece contact information according to a pre-selected initial cutting force model to obtain cutting dynamic contact parameters; performing dynamic cutting process modeling according to the cutting dynamic contact parameters to obtain a dynamic cutting process model;

[0147] Step S362: extracting dynamic cutting parameters from the dynamic cutting process model to obtain dynamic cutting parameters;

[0148] Step S363: quantifying the contact state of the tool-workpiece contact information to obtain quantified contact state data;

[0149] Step S364: analyzing the influence mechanism of tool wear on cutting force coefficient according to the dynamic cutting process model to obtain wear influence parameters;

[0150] Step S365: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the dynamic cutting parameters, the quantified contact state data and the wear influence parameters to obtain a corrected cutting force coefficient;

[0151] Step S366: Generate a cutting force model according to the corrected cutting force coefficient to obtain a cutting force model after the cutting force coefficient is corrected.

[0152] In the embodiment of the present invention, the classic Merchant cutting force model is selected as the initial cutting force model. According to the model, the cutting force is mainly composed of shear force, friction force and plowing force, and is related to factors such as cutting area, cutting speed, and tool geometry angle. Cutting dynamic contact parameters, such as tool-workpiece contact length, contact width, average contact pressure, average friction coefficient, etc., are extracted from the tool-workpiece dynamic contact information obtained in step S35. A dynamic cutting process model is established using MATLAB software, and the extracted cutting dynamic contact parameters are used as input parameters of the model. According to the theoretical formula of the Merchant cutting force model, the components of the cutting force in three directions, as well as the flow direction of the chips, the chip thickness and other parameters are calculated. The tool-workpiece contact information obtained in step S35 is quantified to obtain quantized contact state data. For example, according to the size of the tool-workpiece contact area, the contact state is divided into three levels of light contact, moderate contact and heavy contact, and different values ​​are assigned, such as 1, 2, and 3, respectively, to characterize the strength of the contact state. According to the dynamic cutting process model established in step S361, the influence mechanism of tool wear on the cutting force coefficient is analyzed. For example, by changing the parameters of tool wear in the model, the change of cutting force under different wear states is simulated and calculated, and its change law and influencing factors are analyzed. For example, the influence of tool wear on the three components of cutting force, the influence on the chip flow direction, the influence on the chip morphology, etc. can be analyzed. According to the dynamic cutting parameters extracted in step S362, the contact state data quantified in step S363, and the wear influence parameters analyzed in step S364, the cutting force coefficient in the initial cutting force model is corrected. For example, a mathematical model between the cutting force coefficient and the dynamic cutting parameters, contact state data, and wear influence parameters can be established by using a multivariate linear regression method, and the model parameters can be fitted according to the actual cutting force measurement data to obtain a corrected cutting force coefficient. The corrected cutting force coefficient in step S365 is substituted into the initial cutting force model to replace the original cutting force coefficient to obtain a cutting force model after the cutting force coefficient is corrected. For example, the corrected cutting force coefficient is substituted into the theoretical formula of the Merchant cutting force model to obtain an updated cutting force model.

[0153] Preferably, step S37 includes the following steps:

[0154] Step S371: extracting cutting geometric parameters from the cutting process data to obtain cutting geometric parameters, wherein the cutting geometric parameters include cutting thickness parameters, cutting width parameters, tool rake angle parameters, and tool clearance angle parameters;

[0155] Step S372: performing cutting force coefficient interpolation calculation on the corrected cutting force coefficient according to the cutting thickness parameter and the cutting width parameter to obtain the interpolated cutting force coefficient;

[0156] Step S373: performing cutting force model correction of cutting force tool geometric parameters on the cutting force model after cutting force coefficient correction according to the tool rake angle parameter and the tool clearance angle parameter, to obtain the cutting force model after cutting force tool geometric parameter correction;

[0157] Step S374: Calculate the cutting force components according to the interpolated cutting force coefficient, the cutting geometric parameters, and the cutting force model corrected by the tool geometric parameters to obtain corrected cutting force components.

[0158] In an embodiment of the present invention, cutting geometric parameters are extracted from the cutting process data obtained in step S14, including cutting thickness parameters, cutting width parameters, tool rake angle parameters, and tool back angle parameters. For example, the cutting process data includes information such as tool position, workpiece size, and cutting trajectory. The cutting thickness and cutting width of the tool at each cutting moment can be obtained by using a geometric calculation method, and the tool rake angle and back angle parameters can be directly obtained from the tool model. According to the cutting thickness parameters and cutting width parameters extracted in step S371, the cutting force coefficient corrected in step S365 is interpolated to obtain the interpolated cutting force coefficient. For example, assuming that the corrected cutting force coefficient is a two-dimensional table data about the cutting thickness and cutting width, a bilinear interpolation method can be used to calculate the corresponding cutting force coefficient according to the cutting thickness and cutting width at the current cutting moment. According to the tool rake angle parameters and tool back angle parameters extracted in step S371, the cutting force model after the cutting force coefficient is corrected obtained in step S366 is corrected. For example, the cutting force model obtained in step S366 only considers the influence of parameters such as cutting thickness and cutting width, but does not consider the influence of tool rake angle and back angle. Therefore, it is necessary to correct the correlation coefficients in the cutting force model according to the tool rake angle and clearance angle parameters, for example, by introducing the trigonometric function terms of the tool rake angle and clearance angle to reflect the influence of the tool geometric angle on the cutting force. The interpolated cutting force coefficient obtained in step S372, the cutting geometric parameters extracted in step S371, and the cutting force model corrected by the tool geometric parameters obtained in step S373 are substituted into the cutting force model to calculate the corrected cutting force components. For example, the interpolated cutting force coefficient, cutting thickness, cutting width, tool rake angle, tool clearance angle and other parameters are substituted into the corrected cutting force model to calculate the components of the cutting force in the cutting direction, feed direction and back-cutting direction.

[0159] Preferably, step S4 comprises the following steps:

[0160] Step S41: extracting cutting force characteristics from tool wear state data, tool-workpiece contact information and corrected cutting force data to obtain cutting force characteristics;

[0161] Step S42: using the cutting force characteristics to identify model parameters of the digital twin model to obtain model parameters; using the model parameters to update the digital twin model to obtain an updated digital twin model;

[0162] Step S43: Use the updated digital twin model to simulate the new processing process to obtain simulation results; obtain new actual processing process data; compare the simulation results and the new actual processing process data, and perform model verification and evaluation to obtain model evaluation results;

[0163] Step S44: Deploy the updated digital twin model to the actual processing system according to the model evaluation results to implement the CNC machine tool digital twin model update operation.

[0164] As an example of the present invention, refer to Figure 3 As shown, in this example, step S4 includes:

[0165] Step S41: extracting cutting force characteristics from tool wear state data, tool-workpiece contact information and corrected cutting force data to obtain cutting force characteristics;

[0166] In the embodiment of the present invention, tool wear characteristics, such as tool tip wear, flank wear width, etc., are extracted from the tool wear state data obtained in step S23. Contact characteristics, such as average contact pressure, maximum contact pressure, contact area, etc., are extracted from the tool-workpiece contact information obtained in step S35. Cutting force characteristics, such as cutting force mean, cutting force peak, cutting force frequency, etc., are extracted from the corrected cutting force data obtained in step S38. The extracted tool wear characteristics, contact characteristics, and cutting force characteristics are fused to form a cutting force feature vector for subsequent model parameter identification.

[0167] Step S42: using the cutting force characteristics to identify model parameters of the digital twin model to obtain model parameters; using the model parameters to update the digital twin model to obtain an updated digital twin model;

[0168] In an embodiment of the present invention, the parameters that need to be identified in the digital twin model, such as tool-workpiece contact stiffness, friction coefficient, material parameters, etc., are set as parameters to be optimized. The genetic algorithm is used to identify the parameters of the digital twin model, and the cutting force characteristic vector extracted in step S41 is used as the input of the objective function. The error between the simulation output of the digital twin model and the actual cutting force characteristic vector is minimized by continuously iteratively optimizing the parameters to be optimized. The optimized parameters are used as updated model parameters to update the digital twin model, such as updating the tool-workpiece contact model, updating the material constitutive model, etc., to obtain an updated digital twin model.

[0169] Step S43: Use the updated digital twin model to simulate the new processing process to obtain simulation results; obtain new actual processing process data; compare the simulation results and the new actual processing process data, and perform model verification and evaluation to obtain model evaluation results;

[0170] In an embodiment of the present invention, the updated digital twin model is used to simulate the new processing process. For example, new processing parameters, such as cutting speed, feed speed, cutting depth, etc., as well as new tool and workpiece models are input, and the digital twin model is run for simulation calculation to obtain simulation results, such as cutting force, cutting temperature, tool wear, etc. At the same time, the sensor system is used to collect new actual processing data in real time, such as cutting force, cutting temperature, vibration signal, etc. The simulation results and the new actual processing data are compared and analyzed, such as calculating the error and correlation coefficient between the two, and the model is verified and evaluated according to the pre-set model evaluation indicators, such as accuracy, stability, reliability, etc., to obtain the model evaluation results.

[0171] Step S44: deploying the updated digital twin model to the actual processing system according to the model evaluation result to implement the digital twin model update operation of the CNC machine tool;

[0172] In an embodiment of the present invention, according to the model evaluation result obtained in step S43, it is determined whether the updated digital twin model meets the accuracy and reliability requirements of the actual processing process. If the model evaluation result meets the requirements, the updated digital twin model is deployed to the actual processing system, such as the control system of the CNC machine tool, the tool management system, etc., to realize the processing process monitoring, optimization and prediction operations. For example, the digital twin model can be used to predict the cutting force in real time, and the processing parameters can be adjusted according to the prediction results to ensure the processing quality and tool life; the digital twin model can also be used to monitor the tool wear status in real time, and send an alarm signal when the tool reaches the preset wear limit to remind the operator to replace the tool.

[0173] Preferably, the present invention also provides an updating system for a digital twin model of a CNC machine tool, which is used to execute the updating method for the digital twin model of a CNC machine tool as described above, and the updating system for the digital twin model of a CNC machine tool comprises:

[0174] The virtual cutting simulation module is used to obtain the actual processing data; the cutting process is simulated based on the actual processing data and the digital twin model of the CNC machine tool to obtain the cutting process data;

[0175] Tool wear prediction module, used to analyze tool wear state characteristics of cutting process data to obtain tool wear state data;

[0176] A dynamic cutting parameter correction module is used to update the tool model according to the tool wear state data to obtain an updated tool model; solve the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyze the contact key data of the dynamic contact solution result to obtain tool-workpiece dynamic contact information; perform cutting force coefficient correction on the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient correction; calculate the cutting force component according to the cutting process data and the cutting force model after the cutting force coefficient correction, and output the cutting force to obtain corrected cutting force data;

[0177] The digital twin model update module is used to update the model parameters of the digital twin model using tool wear status data, tool-workpiece contact information and corrected cutting force data to obtain an updated digital twin model to realize the digital twin model update operation of the CNC machine tool.

[0178] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0179] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for updating a digital twin model of a CNC machine tool, characterized in that: The following steps are involved: Step S1: obtaining actual processing data; The cutting process is simulated based on the actual processing data and the digital twin model of the CNC machine tool to obtain the cutting process data; Step S2: Analyze the tool wear state characteristics of the cutting process data to obtain tool wear state data; Step S3: updating the tool model according to the tool wear state data to obtain an updated tool model; solving the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyzing the contact key data of the dynamic contact solution result to obtain tool-workpiece dynamic contact information; correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient is corrected; The cutting force components are calculated according to the cutting process data and the cutting force model corrected by the cutting force coefficient, and the cutting force is output to obtain the corrected cutting force data; Step S4: Use the tool wear status data, tool-workpiece contact information and corrected cutting force data to update the model parameters of the digital twin model to obtain an updated digital twin model to implement the CNC machine tool digital twin model update operation.

2. The method for updating the digital twin model of a CNC machine tool according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: extracting a workpiece model and a tool model from the digital twin model of the CNC machine tool, and performing model preprocessing to obtain model data; Step S12: Acquire actual processing data; set process parameters according to the actual processing data to obtain process parameters; Step S13: using the model data as input data and the process parameters as boundary conditions, finite element analysis technology is used to simulate the cutting process to obtain simulation results; Step S14: Analyze key cutting data on the simulation results to obtain cutting process data.

3. The method for updating the digital twin model of a CNC machine tool according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: extracting tool wear features from cutting process data to obtain tool wear features; Step S22: Perform machine learning training using tool wear characteristics and a pre-selected wear model to obtain a wear prediction model; Step S23: using the wear prediction model to predict tool wear according to the cutting process data to obtain tool wear status data.

4. The method for updating the digital twin model of a CNC machine tool according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: extracting a tool model and a workpiece model from the digital twin model to obtain a tool model and a workpiece model; Step S32: updating the tool model according to the tool wear state data to obtain an updated tool model; Step S33: importing the updated tool model and workpiece model into the contact mechanics analysis software to construct the contact model, and applying the cutting force according to the cutting process data to obtain the contact model; Step S34: solving the dynamic contact state of the contact model to obtain a dynamic contact solution result; Step S35: analyzing the key contact data of the dynamic contact solution results to obtain tool-workpiece dynamic contact information; Step S36: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a corrected cutting force coefficient and a cutting force model after the cutting force coefficient is corrected; Step S37: Calculate the cutting force components according to the cutting process data and the cutting force model corrected by the cutting force coefficient to obtain the corrected cutting force components; Step S38: synthesize and output the corrected cutting force components to obtain corrected cutting force data.

5. The method for updating the digital twin model of a CNC machine tool according to claim 4, characterized in that: Step S32 includes the following steps: Step S321: extracting the tool edge wear and flank wear from the tool wear state data to obtain the tool edge and flank wear parameters; Step S322: mapping the wear parameters of the cutting edge and the flank surface to corresponding positions of the cutting tool model according to the cutting tool geometry of the cutting tool model, and obtaining the mapped wear amount; Step S323: performing tool geometry deformation processing on the tool model according to the mapped wear amount to obtain a deformed tool model; Step S324: performing model smoothing processing on the deformed tool model to obtain an updated tool model.

6. The method for updating the digital twin model of a CNC machine tool according to claim 4, characterized in that: Step S34 includes the following steps: Step S341: defining the tool cutting motion according to the actual machining process data, and defining the workpiece clamping rotation, to obtain the tool cutting motion data and the workpiece clamping rotation data; Step S342: setting the time step according to the tool cutting motion data and the workpiece clamping rotation data to obtain the time step; Step S343: performing model dynamic processing on the contact model according to the time step, and setting initial conditions to obtain an initial state model; Step S344: using contact mechanics analysis software to perform iterative solution of dynamic contact between the tool and the workpiece according to the initial state model to obtain results at each time step; Step S345: Integrate and analyze the results of each time step to obtain dynamic contact solution results.

7. The method for updating the digital twin model of a CNC machine tool according to claim 4, characterized in that: Step S36 includes the following steps: Step S361: extracting cutting dynamic contact information from tool-workpiece contact information according to a pre-selected initial cutting force model to obtain cutting dynamic contact parameters; performing dynamic cutting process modeling according to the cutting dynamic contact parameters to obtain a dynamic cutting process model; Step S362: extracting dynamic cutting parameters from the dynamic cutting process model to obtain dynamic cutting parameters; Step S363: quantifying the contact state of the tool-workpiece contact information to obtain quantified contact state data; Step S364: analyzing the influence mechanism of tool wear on cutting force coefficient according to the dynamic cutting process model to obtain wear influence parameters; Step S365: Correcting the cutting force coefficient of the pre-selected initial cutting force model according to the dynamic cutting parameters, the quantified contact state data and the wear influence parameters to obtain a corrected cutting force coefficient; Step S366: Generate a cutting force model according to the corrected cutting force coefficient to obtain a cutting force model after the cutting force coefficient is corrected.

8. The method for updating the digital twin model of a CNC machine tool according to claim 4, characterized in that: Step S37 includes the following steps: Step S371: extracting cutting geometric parameters from the cutting process data to obtain cutting geometric parameters, wherein the cutting geometric parameters include cutting thickness parameters, cutting width parameters, tool rake angle parameters, and tool clearance angle parameters; Step S372: performing cutting force coefficient interpolation calculation on the corrected cutting force coefficient according to the cutting thickness parameter and the cutting width parameter to obtain the interpolated cutting force coefficient; Step S373: performing cutting force model correction of cutting force tool geometric parameters on the cutting force model after cutting force coefficient correction according to the tool rake angle parameter and the tool clearance angle parameter, to obtain the cutting force model after cutting force tool geometric parameter correction; Step S374: Calculate the cutting force components according to the interpolated cutting force coefficient, the cutting geometric parameters, and the cutting force model corrected by the tool geometric parameters to obtain corrected cutting force components.

9. The method for updating the digital twin model of a CNC machine tool according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: extracting cutting force characteristics from tool wear state data, tool-workpiece contact information and corrected cutting force data to obtain cutting force characteristics; Step S42: using the cutting force characteristics to identify model parameters of the digital twin model to obtain model parameters; using the model parameters to update the digital twin model to obtain an updated digital twin model; Step S43: Use the updated digital twin model to simulate the new processing process to obtain simulation results; obtain new actual processing process data; compare the simulation results and the new actual processing process data, and perform model verification and evaluation to obtain model evaluation results; Step S44: Deploy the updated digital twin model to the actual processing system according to the model evaluation results to implement the CNC machine tool digital twin model update operation.

10. A system for updating a digital twin model of a CNC machine tool, characterized in that: The updating method for the digital twin model of a CNC machine tool according to claim 1 is used to execute the updating system for the digital twin model of a CNC machine tool, and the updating system for the digital twin model of a CNC machine tool comprises: The virtual cutting simulation module is used to obtain the actual processing data; the cutting process is simulated based on the actual processing data and the digital twin model of the CNC machine tool to obtain the cutting process data; Tool wear prediction module, used to analyze tool wear state characteristics of cutting process data to obtain tool wear state data; A dynamic cutting parameter correction module is used to update the tool model according to the tool wear state data to obtain an updated tool model; solve the dynamic contact state according to the updated tool model to obtain a dynamic contact solution result; analyze the contact key data of the dynamic contact solution result to obtain tool-workpiece dynamic contact information; perform cutting force coefficient correction on the pre-selected initial cutting force model according to the tool-workpiece contact information to obtain a cutting force model after the cutting force coefficient correction; calculate the cutting force component according to the cutting process data and the cutting force model after the cutting force coefficient correction, and output the cutting force to obtain corrected cutting force data; The digital twin model update module is used to update the model parameters of the digital twin model using tool wear status data, tool-workpiece contact information and corrected cutting force data to obtain an updated digital twin model to realize the digital twin model update operation of the CNC machine tool.

Images