Machine tool machining control method and system

Abstract

The invention relates to the technical field of machine tool control, and provides a machine tool machining control method and system.The method comprises the steps that in the machine tool machining process, microenvironment parameters of a cutting area are obtained and monitored; obtaining a cutter type, a workpiece material attribute and a preset cutting parameter corresponding to the machine tool; based on the tool type, the workpiece material attributes and the preset cutting parameters, an initial judgment basis of the tool wear speed is established; according to the deviation between the microenvironment parameters and a preset standard value, the initial judgment basis is corrected, and the corrected tool wear speed is obtained; according to the actual tool abrasion loss, the rule for subsequently correcting the initial judgment basis is optimized; and based on the corrected tool wear speed, the tool remaining life is updated, and when tool wear reaches a preset threshold value, a compensation strategy of the cutting parameters or the tool path is decided and executed in combination with the value of the machined workpiece and the remaining machining task. The method has the effect of improving the prediction accuracy of the tool wear speed.

Description

Technical Field

[0001] This invention relates to the technical field of machine tool control, and specifically to a machine tool machining control method and system. Background Technology

[0002] In modern high-end manufacturing, such as the production of aerospace engine components, precision medical devices, or complex molds, CNC machine tools are the core for achieving high-precision and high-efficiency machining. However, in actual machining processes, especially when dealing with difficult-to-machine materials such as high-strength alloys and heat-resistant alloys, cutting tools are subjected to enormous mechanical stress, high temperatures, and abrasive wear, leading to gradual tool wear. Tool wear alters the effective cutting angle and edge sharpness, affecting the stability of the cutting process and ultimately resulting in a decline in surface quality or even dimensional deviations. If tool wear is not detected and addressed in a timely manner, in extreme cases, it may lead to sudden tool chipping or breakage. This not only damages the high-value workpiece being machined but may also cause impact damage to critical components such as the machine tool spindle and tool magazine, triggering downtime, maintenance, and resulting in significant economic losses and production delays.

[0003] As the manufacturing industry continues to demand higher processing efficiency and product quality, traditional tool management and wear monitoring methods are no longer sufficient. This is especially true when machining high-value, highly complex critical components, where even minor machining errors can render the entire workpiece unusable. The machining cycle for these workpieces often lasts tens or even hundreds of hours, involving a wide variety of tools, each with a very high cost. In such cases, relying solely on experience or rudimentary monitoring methods to determine tool condition is prohibitively risky and costly. This raises a crucial technical challenge: how to achieve real-time, accurate perception of tool wear, dynamically adjust wear predictions based on this information, and intelligently select and execute compensation strategies for cutting parameters or toolpaths to ensure that while extending tool life, stringent machining accuracy and surface quality requirements are consistently met.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] This application discloses a machine tool processing control method and system, which aims to solve the problems of inaccurate tool wear monitoring, untimely prediction, and unintelligent compensation strategies in the prior art, so as to improve processing accuracy, extend tool life, and reduce production costs.

[0006] The technical solution of this application is as follows: In a first aspect, this application discloses a machine tool machining control method, comprising: acquiring and monitoring micro-environmental parameters of the cutting area during machine tool machining; acquiring the tool type, workpiece material properties, and preset cutting parameters corresponding to the machine tool; establishing an initial judgment basis for tool wear rate based on the tool type, workpiece material properties, and preset cutting parameters; correcting the initial judgment basis according to the deviation between the micro-environmental parameters and the preset standard value to obtain the corrected tool wear rate; optimizing the rules for subsequent correction of the initial judgment basis according to the actual tool wear amount; updating the remaining tool life based on the corrected tool wear rate; and, when the tool wear reaches a preset threshold, deciding and executing a compensation strategy for cutting parameters or tool path based on the value of the machined workpiece and the remaining machining tasks.

[0007] This technical solution enables real-time and accurate perception of tool wear status, and on this basis, dynamically adjusts wear prediction. At the same time, it intelligently selects and executes compensation strategies for cutting parameters or tool paths, effectively solving the problems of tool resource waste, strong subjectivity of manual judgment, and insufficient sensitivity of indirect monitoring methods in traditional tool management methods, and significantly improving machining accuracy and production efficiency.

[0008] Furthermore, according to the machine tool processing control method, the steps for optimizing the rules for subsequent correction of the initial judgment criteria based on the actual tool wear include: capturing the original analog signal of the cutting area; triggering the storage of the instantaneous data block when the instantaneous rate of change of the original analog signal exceeds a preset multiple standard deviation of the historical average rate of change within a preset time period; performing multi-dimensional feature extraction and correlation analysis on the instantaneous data block, calculating the maximum instantaneous rise slope and peak temperature of the infrared temperature signal, and performing Fourier transform analysis on the high-frequency acceleration signal to extract the energy spectral density and total energy peak of the ultrasonic band; and based on the maximum instantaneous rise slope... Transient extreme wear acceleration events are identified by the peak temperature, energy spectral density, total energy peak, and peak time synchronization of all characteristics. Transient failures of the cutting fluid micro-interface are inferred based on the peak temperature and energy spectral density of the ultrasonic band of the transient extreme wear acceleration events. Event wear equivalents are generated based on the transient failures of the cutting fluid micro-interface and the intensity characteristics of the transient extreme wear acceleration events. The total tool wear is obtained by accumulating the conventional wear amount and the event wear equivalent. The coefficients in the nonlinear mapping function are adjusted based on the deviation between the actual tool wear amount and the total tool wear amount to optimize and correct the rules used for the initial judgment.

[0009] This technical solution enables a deep understanding and accurate modeling of tool wear mechanisms through refined analysis of the original analog signals, particularly the identification and quantification of transient extreme wear acceleration events. This significantly improves the accuracy of tool wear prediction and overcomes the limitations of traditional methods in distinguishing different wear types.

[0010] In some preferred embodiments, according to the machine tool machining control method, the step of generating event wear equivalent based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of transient extreme wear acceleration events includes: identifying transient extreme wear acceleration events; recording the peak temperature, peak vibration energy, and duration of the transient extreme wear acceleration events; determining whether at least two transient extreme wear acceleration events are continuously identified within a preset time window; when at least two transient extreme wear acceleration events are continuously identified, extracting the total duration of all transient extreme wear acceleration events, the highest temperature peak value within the sequence, the highest vibration energy peak value, and the average time interval between events; based on the total duration, the highest temperature peak value within the sequence, the highest vibration energy peak value, and the average time interval, consulting a nonlinear damage superposition mapping table to generate a sequence wear equivalent; and using the sequence wear equivalent as the event wear equivalent.

[0011] This technical solution enables more accurate assessment of cumulative wear effects by performing sequential analysis and damage superposition on continuously occurring transient extreme wear acceleration events. It effectively captures the nonlinear characteristics of complex wear processes and further improves the accuracy of wear prediction.

[0012] As an optional solution, based on the machine tool machining control method, the steps for optimizing the rules for subsequent correction of the initial judgment criteria based on the actual tool wear include: obtaining the actual tool wear during the machining clearance; comparing the actual tool wear with the total tool wear to obtain the deviation; when the deviation meets the preset conditions, identifying the transient extreme wear acceleration event related to the deviation; and iteratively adjusting the coefficients in the nonlinear mapping function based on the deviation and the characteristic parameters of the transient extreme wear acceleration event.

[0013] This technical solution enables adaptive optimization of the wear prediction model by calibrating the actual wear amount during machining clearance and combining the correlation analysis between deviation and transient extreme wear events, thus ensuring the accuracy and robustness of the prediction model in long-term operation.

[0014] Furthermore, according to the machine tool processing control method, when at least two transient extreme wear acceleration events are continuously identified, the steps of extracting the total duration of all transient extreme wear acceleration events, the highest temperature peak value within the sequence, the highest vibration energy peak value, and the average time interval between events include: identifying transient extreme wear acceleration events; recording the start time, end time, peak temperature, and peak vibration energy of the transient extreme wear acceleration events; maintaining a sliding time window and calculating the occurrence density of transient extreme wear acceleration events within the sliding time window in real time; when the occurrence density is continuously higher than a preset threshold within a consecutive preset time period, marking the first transient extreme wear acceleration event in the corresponding region as the starting point of the continuous sequence. The system continuously tracks the occurrence of transient extreme wear acceleration events until the occurrence density remains below a preset threshold or the total duration of the continuous sequence reaches a preset maximum sequence duration. The last transient extreme wear acceleration event is then marked as the end point of the continuous sequence. Based on the start and end times of each transient extreme wear acceleration event, the total duration of the continuous sequence is extracted. Based on the peak temperature and peak vibration energy of each transient extreme wear acceleration event, the system identifies the highest temperature peak and the highest vibration energy peak of each transient extreme wear acceleration event within the continuous sequence. The system calculates the weighted average of the start time differences of all adjacent transient extreme wear acceleration events within the continuous sequence as the average time interval between events.

[0015] This technical solution enables the intelligent identification and precise definition of continuous sequences of transient extreme wear acceleration events by introducing a sliding time window and occurrence density analysis. This provides more accurate input parameters for subsequent calculation of the wear equivalent of the sequence, further improving the precision of wear assessment.

[0016] Based on this, according to the machine tool processing control method, and based on the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval, the steps to generate the sequence wear equivalent by consulting the nonlinear damage superposition mapping table include: calculating the local geometric feature parameters of the current cutting point based on the three-dimensional coordinates of the current cutting point and the CAD model of the workpiece; obtaining the geometric constraint influence factor based on the local geometric feature parameters; weighting and fusing the geometric constraint influence factor with the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval to obtain the corrected sequence feature parameters; and generating the sequence wear equivalent by consulting the nonlinear damage superposition mapping table based on the corrected sequence feature parameters.

[0017] This technical solution allows the local geometric feature parameters of the workpiece to be incorporated into the wear equivalent calculation. By correcting the sequence feature parameters through geometric constraint influence factors, the wear prediction becomes more consistent with the actual processing conditions. In particular, it significantly improves the accuracy of the prediction in the processing of complex geometric areas.

[0018] To enhance functionality, based on machine tool machining control methods and local geometric feature parameters, the steps for obtaining geometric constraint influence factors include: real-time monitoring of acoustic emission signals and spindle torque signals in the cutting area; frequency domain analysis of the acoustic emission signals to extract the energy distribution of characteristic frequency bands; wavelet decomposition of the spindle torque signals to extract high-frequency components; and obtaining and adjusting geometric constraint influence factors based on energy distribution, high-frequency components, and local geometric feature parameters.

[0019] This technical solution enables a more comprehensive capture of dynamic mechanical and energy changes during the cutting process by combining the analysis of acoustic emission signals and spindle torque signals. This allows for more accurate acquisition and adjustment of geometric constraint influencing factors, further enhancing the adaptability of the wear prediction model to complex cutting environments.

[0020] For specific situations, according to machine tool machining control methods, when in complex geometric regions, the steps for obtaining and adjusting the geometric constraint influence factor based on energy distribution, high-frequency components, and local geometric feature parameters include: performing real-time time-frequency analysis on the acoustic emission signal to extract the energy distribution and duration of the transient shock wave; performing high-order spectrum analysis on the spindle torque signal to extract non-Gaussian characteristics and bispectral correlation; taking the energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric feature parameters of the transient shock wave as input, calculating and outputting the correction amount of the geometric constraint influence factor in real time through a nonlinear mapping function, and superimposing the correction amount onto the initially obtained geometric constraint influence factor.

[0021] This technical solution enables a deeper understanding of the cutting process in complex geometric regions by introducing more advanced signal processing techniques (such as time-frequency analysis and higher-order spectral analysis). This allows for more accurate calculation and correction of geometric constraint influence factors, significantly improving the accuracy of wear prediction under extremely complex working conditions.

[0022] To further address the problem, based on machine tool machining control methods, the following steps are taken: Using the energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric feature parameters of the transient shock wave as input, a nonlinear mapping function is used to calculate and output the correction amount of the geometric constraint influence factor in real time. This correction amount is then superimposed onto the initially acquired geometric constraint influence factor. The steps include: acquiring the actual tool wear at the end of the machining gap or a specific machining stage; comparing the actual tool wear with the total tool wear to obtain the deviation; when the deviation exceeds a preset threshold, backtracking historical data to identify the transient extreme wear acceleration event sequence causing the deviation, and extracting the feature parameters, local geometric feature parameters, energy distribution of characteristic frequency bands, and high-frequency components of the spindle torque signal from the transient extreme wear acceleration event sequence, integrating them into the corresponding historical data. Based on historical data and deviations, adjust the weight parameters or activation function parameters in the nonlinear mapping function; during processing, monitor material batch information and machine tool status parameters; based on material batch information and machine tool status parameters, when a change in material batch or a drift in machine tool status parameters is detected, trigger the online adaptive adjustment mechanism of the nonlinear mapping function; based on the online adaptive adjustment mechanism, fine-tune the parameters of the nonlinear mapping function according to the new material batch characteristics or machine tool status drift trend; using the energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components of spindle torque signal, and local geometric feature parameters of transient shock waves as inputs, calculate and output the correction amount of geometric constraint influence factor in real time through the adjusted nonlinear mapping function, and superimpose the correction amount onto the initially acquired geometric constraint influence factor.

[0023] This technical solution enables dynamic optimization of the nonlinear mapping function by introducing an online adaptive adjustment mechanism, combined with the actual wear deviation, historical data backtracking, and monitoring of material batches and machine tool conditions. This ensures that the wear prediction model can maintain high accuracy and robustness when facing uncertainties such as material changes and machine tool aging.

[0024] Secondly, this application also discloses a machine tool processing control system for performing machine tool processing control, including: a micro-environment parameter acquisition module for acquiring and monitoring micro-environment parameters of the cutting area during machine tool processing; a work information acquisition module for acquiring the tool type, workpiece material properties, and preset cutting parameters corresponding to the machine tool; a judgment basis establishment module for establishing an initial judgment basis for tool wear rate based on tool type, workpiece material properties, and preset cutting parameters; a wear rate correction module for correcting the initial judgment basis according to the deviation between the micro-environment parameters and preset standard values ​​to obtain the corrected tool wear rate; a judgment basis optimization module for optimizing the rules for subsequent correction of the initial judgment basis based on the actual tool wear amount; and a compensation strategy execution module for updating the remaining tool life based on the corrected tool wear rate, and when the tool wear reaches a preset threshold, deciding and executing a compensation strategy for cutting parameters or tool path based on the value of the processed workpiece and the remaining processing tasks.

[0025] This technical solution provides a system that integrates multi-sensor data acquisition, intelligent wear prediction, and adaptive compensation strategies, enabling comprehensive and precise control of the machine tool processing process. It effectively addresses the shortcomings of traditional systems in tool management and wear monitoring, significantly improving production efficiency and product quality.

[0026] Beneficial Effects: The machine tool machining control method disclosed in this application acquires and monitors the micro-environment parameters of the cutting area in real time during machine tool machining, and establishes an initial judgment basis for tool wear rate by combining tool type, workpiece material properties, and preset cutting parameters. This method further corrects the initial judgment basis based on the deviation between the micro-environment parameters and preset standard values, thereby obtaining a more accurate tool wear rate. Based on this, this application can continuously optimize the rules for subsequently correcting the initial judgment basis according to the actual tool wear amount, enabling the wear prediction model to have adaptive learning capabilities. Finally, based on the corrected tool wear rate, the system can update the remaining tool life, and when the tool wear reaches a preset threshold, intelligently decides and executes compensation strategies for cutting parameters or tool paths, combining the value of the machined workpiece and the remaining machining tasks.

[0027] Through the above technical solution, this application effectively solves the technical problems of inaccurate tool wear monitoring, untimely prediction, and unintelligent compensation strategies in the prior art. Specifically, traditional tool life management strategies are often too conservative or rely too heavily on manual experience, leading to wasted tool resources or misjudgments. This application, by introducing real-time monitoring and adaptive correction mechanisms for micro-environment parameters, significantly improves the accuracy of tool wear rate prediction, avoiding workpiece scrap and machine tool damage caused by excessive wear, and also preventing premature tool replacement, thereby saving production costs. Furthermore, the intelligent compensation strategy can dynamically adjust cutting parameters or tool paths according to actual machining needs and workpiece value, ensuring that while extending tool life, stringent machining accuracy and surface quality requirements are always met, making it particularly suitable for precision machining of high-value, complex parts. Therefore, this application provides a more efficient, economical, and reliable machine tool machining control solution. Attached Figure Description

[0028] Figure 1 This is a flowchart of a machine tool processing control method according to one embodiment of the present invention; Figure 2 This is a flowchart of a machine tool processing control method according to another embodiment of the present invention; Figure 3 This is a system block diagram of a machine tool processing control system according to another embodiment of the present invention; Explanation of reference numerals in the attached figures: 1. Machine tool processing control system; 11. Micro-environment parameter acquisition module; 12. Working information acquisition module; 13. Judgment basis establishment module; 14. Wear rate correction module; 15. Judgment basis optimization module; 16. Compensation strategy execution module. Detailed Implementation

[0029] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0030] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] This application proposes a machine tool machining control method, combined with Figure 1 As shown, it includes: S1, during machine tool processing, acquires and monitors the microenvironmental parameters of the cutting area; S2, obtain the tool type, workpiece material properties and preset cutting parameters corresponding to the machine tool; S3 establishes the initial judgment criteria for tool wear rate based on tool type, workpiece material properties, and preset cutting parameters; S4. Based on the deviation between the microenvironment parameters and the preset standard values, the initial judgment criteria are corrected to obtain the corrected tool wear rate. S5, based on the actual tool wear, optimize the rules for subsequent correction of the initial judgment basis; S6 updates the remaining tool life based on the corrected tool wear rate, and when the tool wear reaches a preset threshold, it decides and executes a compensation strategy for cutting parameters or toolpaths based on the value of the workpiece being machined and the remaining machining tasks.

[0032] Among them, "microenvironmental parameters" refer to the local physical and chemical conditions within the machine tool cutting area that directly affect tool wear, such as the temperature, pressure, flow rate, pH value, and concentration of the cutting fluid, as well as vibration, acoustic emission, cutting force, torque, acoustic characteristics, and infrared radiation in the cutting area. These parameters can reflect the energy dissipation, friction state, and material removal mechanism during the cutting process in real time, and are an important basis for assessing the tool wear condition.

[0033] "Tool type" refers to the type of cutting tool used in machining, such as milling cutters, turning tools, and drills, as well as their specific geometry, coating materials, base materials, and other characteristics. Different tool types exhibit different wear mechanisms and wear rates during the cutting process.

[0034] "Workpiece material properties" refer to the physical and chemical characteristics of the workpiece being machined, such as hardness, toughness, thermal conductivity, and chemical activity. These properties directly affect the cutting force, cutting heat, and tool wear during the cutting process.

[0035] "Preset cutting parameters" refer to parameters such as cutting speed, feed rate, and depth of cut that are set before machining based on process requirements and experience. These parameters are key factors affecting the tool wear rate.

[0036] "Initial judgment basis for tool wear rate" refers to the preliminary estimate of tool wear rate established before machining begins, based on known tool type, workpiece material properties, and preset cutting parameters, through empirical formulas, database queries, or preliminary models.

[0037] "Preset standard values" refer to the normal or ideal values ​​set for microenvironment parameters. When microenvironment parameters deviate from these standard values, it usually means that the cutting conditions have changed, which may lead to accelerated tool wear.

[0038] "Actual tool wear" refers to the true degree of tool wear over a period of time, obtained through direct or indirect measurement methods, such as measuring the width of the wear band on the cutting edge using an optical microscope, or inferring it indirectly through force sensors, acoustic emission sensors, etc.

[0039] "Preset threshold" refers to the critical value at which the tool wear reaches the point where compensation measures or tool replacement are required.

[0040] "Workpiece value" refers to the economic value of a workpiece currently being processed. High-value workpieces typically require more conservative compensation strategies to avoid scrapping.

[0041] "Remaining machining tasks" refers to the amount of machining that has not yet been completed for the current workpiece or batch of workpieces. This will affect the choice of compensation strategy to ensure that the tool life is maximized while completing the tasks.

[0042] This method is typically implemented in a smart manufacturing environment that integrates sensors, a data acquisition system, a data processing unit (such as an industrial PC or embedded controller), and a machine tool numerical control (CNC) system. The sensors are responsible for acquiring micro-environment parameters in real time, the data acquisition system converts analog signals into digital signals, the data processing unit executes algorithmic logic, and communicates with the CNC system to adjust cutting parameters or toolpaths.

[0043] The machine tool processing control method proposed in this application achieves accurate prediction and intelligent compensation for tool wear through a series of closely related steps.

[0044] First, during machine tool processing, it is necessary to acquire and monitor the micro-environmental parameters of the cutting area. This can be achieved in several ways. For example, infrared temperature sensors can be deployed to measure the temperature of the cutting area in real time; acoustic emission sensors can be installed to capture high-frequency elastic wave signals during the cutting process; vibration sensors can be used to monitor the vibration of the machine tool spindle or cutting tool; and even vision systems can be used to capture changes in the shape of cutting sparks or chips using high-speed cameras. These sensors can work independently or collaboratively, transmitting the collected data to a data processing unit.

[0045] Secondly, it is necessary to obtain the corresponding tool type, workpiece material properties, and preset cutting parameters for the machine tool. Tool type information can be obtained from the tool management system or the machine tool's tool magazine, such as tool model, diameter, and coating material. Workpiece material properties can be obtained from the production planning system or material database, such as material hardness, strength, and thermal conductivity. Preset cutting parameters are typically set by process engineers during programming, including cutting speed, feed rate, and depth of cut. This information forms the basis for establishing the initial judgment of tool wear.

[0046] Next, based on the tool type, workpiece material properties, and preset cutting parameters, an initial basis for judging the tool wear rate is established. This can be achieved by consulting preset empirical formulas or databases. For example, for a specific combination of tool and workpiece materials, under given cutting parameters, a preliminary wear rate model can be established based on historical data or Taylor formulas. This model provides a benchmark for subsequent dynamic adjustments.

[0047] Then, based on the deviation of the microenvironment parameters from the preset standard values, the initial judgment criteria are revised to obtain the corrected tool wear rate. For example, if the temperature in the cutting area is significantly higher than the preset standard value, this may indicate accelerated tool wear. In this case, the wear rate in the initial judgment criteria can be adjusted upwards based on the magnitude of the temperature increase. Similarly, if the energy spectrum of the acoustic emission signal undergoes abnormal changes, it may also indicate a change in the wear state, thereby correcting the wear rate. This correction can be linear or nonlinear, depending on the relationship model between the microenvironment parameters and the wear rate.

[0048] Furthermore, based on the actual tool wear, the rules for subsequent corrections are optimized to establish the initial judgment criteria. For example, during machining gaps or after completing a workpiece, the actual tool wear can be obtained through offline measurement (such as microscopic observation) or online measurement (such as wear measurement based on image recognition). The actual wear is then compared with the predicted wear obtained through the above correction method. If there is a significant deviation, the weighting coefficients or model parameters in the correction rules need to be adjusted to make subsequent predictions more accurate. This optimization is a continuous learning and iterative process, enabling the system to adapt to different machining conditions and wear patterns.

[0049] Finally, based on the corrected tool wear rate, the remaining tool life is updated. When tool wear reaches a preset threshold, the system considers the value of the workpiece and the remaining machining tasks to determine and execute compensation strategies for cutting parameters or toolpaths. For example, if the remaining tool life is predicted to be nearing its end, and the current workpiece being machined is of high value, the system may choose to reduce the cutting speed, decrease the feed rate, or even adjust the toolpath to avoid high-load areas, thereby extending tool life and protecting the workpiece. If there are few remaining machining tasks, the system may also choose to replace the tool after the current workpiece is completed. This decision-making process is intelligent and comprehensively considers both economic benefits and production efficiency.

[0050] Optional, combined Figure 2 As shown, the steps in S5 to optimize the rules for subsequent correction of the initial judgment based on the actual tool wear include: S51 captures the original analog signal of the cutting area; S52, when the instantaneous rate of change of the original analog signal exceeds the preset multiple standard deviation of the historical average rate of change within a preset time period, the storage of the instantaneous data block is triggered; S53 performs multi-dimensional feature extraction and correlation analysis on instantaneous data blocks, calculates the maximum instantaneous rise slope and peak temperature of infrared temperature signals, and performs Fourier transform analysis on high-frequency acceleration signals to extract the energy spectral density and total energy peak of the ultrasonic band. S54 identifies transient extreme wear acceleration events based on the maximum instantaneous rise slope, peak temperature, energy spectral density, total energy peak value, and the peak time synchronization of all features. S55, based on the peak temperature of the transient extreme wear acceleration event and the energy spectral density of the ultrasonic band, the transient failure of the cutting fluid micro-interface is inferred; S56, Based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of the transient extreme wear acceleration event, generate the event wear equivalent; S57, accumulate the normal wear amount and the event wear equivalent to obtain the total tool wear amount; S58, based on the deviation between the actual tool wear and the total tool wear, adjusts the coefficients in the nonlinear mapping function to optimize and correct the rules used for the initial judgment.

[0051] Specifically, capturing the raw analog signals of the cutting zone refers to acquiring physical signals during the cutting process in real time using sensors installed near the cutting zone of the machine tool, such as infrared temperature sensors, high-frequency acceleration sensors, and acoustic emission sensors. These signals are typically continuous analog signals that reflect the dynamic changes in the cutting zone.

[0052] Specifically, when the instantaneous rate of change of the original analog signal exceeds a preset multiple of the standard deviation of the historical average rate of change within a preset time period, the storage of instantaneous data blocks is triggered. This can be understood as an anomaly event detection mechanism. For example, the instantaneous rate of change is obtained by calculating the first derivative or difference of the signal and comparing it with the statistical characteristics of historical data (such as the mean and standard deviation). When a significant instantaneous change is detected, it indicates that an abnormal cutting condition may have occurred. At this time, the system will immediately store the original data for that time period for subsequent analysis.

[0053] In practical applications, multi-dimensional feature extraction and correlation analysis are performed on instantaneous data blocks to calculate the maximum instantaneous rise slope and peak temperature of the infrared temperature signal, and Fourier transform analysis is performed on the high-frequency acceleration signal to extract the energy spectral density and total energy peak of the ultrasonic band. The aim is to extract key features closely related to tool wear from the original signals. The rise slope and peak temperature of the infrared temperature signal can reflect the instantaneous changes in cutting heat and local high temperatures, while the high-frequency acceleration signal, through Fourier transform analysis, can reveal the vibration energy of the ultrasonic band, which is closely related to phenomena such as tool micro-damage and cutting fluid failure. The peak time synchronization of all features is used to confirm whether these features are caused by the same transient event.

[0054] Identifying transient extreme wear acceleration events involves using the extracted features, such as the maximum instantaneous rise rate, peak temperature, energy spectral density, total energy peak value, and their temporal synchronicity, to determine whether a transient event causing a sharp increase in tool wear has occurred, through preset rules or machine learning models. These events may include micro-chipping, coating peeling, and instantaneous failure of cutting fluid.

[0055] Furthermore, based on the peak temperature and energy spectral density of the ultrasonic band in transient extreme wear acceleration events, the transient failure of the cutting fluid micro-interface is inferred, with the aim of deeply analyzing the root causes of extreme wear events. High instantaneous temperature and high ultrasonic energy are often direct manifestations of local vaporization, rupture, or lubrication film failure of the cutting fluid at the tool-workpiece interface, and such failures can significantly exacerbate tool wear.

[0056] Specifically, generating event wear equivalents, based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of transient extreme wear acceleration events, refers to quantifying the contribution of identified transient extreme wear acceleration events to tool wear. This can be achieved by establishing a mapping model that transforms the severity of transient failures (such as peak temperature and energy spectral density) and the intensity characteristics of events (such as duration and amplitude) into an equivalent wear quantity.

[0057] Therefore, by accumulating both conventional wear and event wear equivalents, the total tool wear is obtained, aiming to provide a more comprehensive and accurate assessment of tool wear. Conventional wear can be obtained through traditional wear models or continuous monitoring data, while event wear equivalents supplement the contribution of transient extreme events to the total wear.

[0058] Finally, based on the deviation between the actual tool wear and the total tool wear, the coefficients in the nonlinear mapping function are adjusted to optimize and correct the rules used for the initial judgment. The purpose is to calibrate and improve the wear prediction model through actual feedback. When there is a deviation between the predicted total wear and the actual measured wear, the system will adaptively adjust the parameters in the nonlinear mapping function used to correct the initial judgment based on the magnitude and direction of the deviation, thereby making subsequent wear predictions more accurate.

[0059] In some preferred embodiments, this application is implemented as follows: Assuming that during high-speed milling of a titanium alloy workpiece, the machine tool machining control system captures the original analog signals of the cutting area in real time using an infrared temperature sensor and a high-frequency acceleration sensor. At a certain moment, the system detects that the instantaneous rise rate of the infrared temperature signal exceeds three times the standard deviation of the historical average rise rate within 10 milliseconds, and simultaneously, a significant peak appears in the energy spectral density of the high-frequency acceleration signal in the ultrasonic band, with both peak times highly synchronized. The system immediately triggers the storage of the instantaneous data block.

[0060] Subsequently, the stored data blocks were analyzed, and the maximum instantaneous rise slope and peak temperature of the infrared temperature signal were calculated to be 800°C. The ultrasonic energy spectral density of the high-frequency acceleration signal reached 1.5 J / Hz, and the total peak energy was 0.05 J. Based on these characteristics and their time synchronization, the system identified this as a transient extreme wear acceleration event and inferred that transient failure of the cutting fluid at the tool-workpiece micro-interface may have occurred.

[0061] Based on the transient failure scenario and event intensity characteristics, the system generates an event wear equivalent, for example, 0.5 micrometers, using a pre-defined nonlinear model. This event wear equivalent is accumulated into the regular wear amount to obtain the total tool wear. During the machining gap, the actual tool wear measured using an optical microscope is 10.2 micrometers, while the system predicts a total tool wear of 9.8 micrometers, resulting in a deviation of 0.4 micrometers. Based on this deviation, the system adjusts the coefficients in the nonlinear mapping function used to correct the initial judgment criteria, such as fine-tuning the weighting parameters, so that subsequent wear predictions can more accurately reflect the actual situation, thereby optimizing the rules for the initial judgment criteria of tool wear rate.

[0062] Optionally, the step of generating the event wear equivalent, based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of the aforementioned transient extreme wear acceleration events, includes: Identify transient, extreme wear acceleration events; Record the peak temperature, peak vibration energy, and duration of transient extreme wear acceleration events; Determine whether at least two transient extreme wear acceleration events are continuously identified within a preset time window; When at least two transient extreme wear acceleration events are identified consecutively, the total duration of all transient extreme wear acceleration events, the highest temperature peak, the highest vibration energy peak, and the average time interval between events are extracted. Based on the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval, consult the nonlinear damage superposition mapping table to generate the sequence wear equivalent. Use the sequence wear equivalent as the event wear equivalent.

[0063] Identifying transient extreme wear acceleration events involves processing and extracting features from the raw simulated signals of the cutting area using the methods described above to determine whether transient, abnormal wear acceleration phenomena exist. Once such an event is identified, its key characteristic parameters, such as peak temperature, peak vibration energy, and the duration of the event, are accurately recorded. These parameters are crucial for assessing the intensity of a single transient extreme wear acceleration event.

[0064] Furthermore, to capture consecutive transient extreme wear acceleration events, the system determines whether at least two of these events are consecutively identified within a preset time window. The preset time window can be understood as a relatively short time interval used to define the "continuity" between events. For example, if the interval between two transient extreme wear acceleration events is less than the preset time window, they are considered to have occurred consecutively.

[0065] When the system consecutively identifies at least two of the aforementioned transient extreme wear acceleration events, it is necessary to extract comprehensive features of the entire event sequence to assess the cumulative effect of these consecutive events. Specifically, this involves extracting the total duration of all transient extreme wear acceleration events, i.e., the total time elapsed from the start of the first event to the end of the last event in the sequence; the highest temperature peak and the highest vibration energy peak within the sequence, i.e., the maximum values ​​among the peak temperatures and peak vibration energies of all events in the entire sequence; and the average time interval between events, used to measure the intensity of event occurrence. These comprehensive features can more fully reflect the overall intensity and sustained impact of the consecutive event sequence.

[0066] Subsequently, based on the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval within the sequence, a nonlinear damage superposition mapping table is consulted to generate the sequence wear equivalent. The nonlinear damage superposition mapping table is a pre-established database or model that stores the nonlinear relationships between different combinations of sequence characteristic parameters and their corresponding wear equivalents. Since consecutive transient extreme wear acceleration events may exhibit complex superposition and synergistic effects, their impact on tool wear is not a simple linear superposition. Therefore, using a nonlinear mapping table can more accurately reflect this complex relationship. By consulting this mapping table, a sequence wear equivalent representing the contribution of the entire event sequence to tool wear can be obtained based on the extracted sequence characteristic parameters.

[0067] Ultimately, the sequence wear equivalent is used as the event wear equivalent. This means that when consecutive transient extreme wear acceleration events occur, the wear equivalents of individual events are no longer simply accumulated. Instead, a more representative event wear equivalent is generated by comprehensively evaluating the entire sequence, thereby improving the accuracy of tool wear assessment.

[0068] Optionally, the steps to optimize the rules for subsequent correction of the initial judgment based on the actual tool wear include: Obtain the actual tool wear during machining gaps; By comparing the actual tool wear with the total tool wear, the deviation can be obtained. When the deviation meets the preset conditions, identify transient extreme wear acceleration events related to the deviation; The coefficients in the nonlinear mapping function are iteratively adjusted based on the deviation and the characteristic parameters of the transient extreme wear acceleration event.

[0069] Specifically, obtaining actual tool wear during machining intervals refers to measuring the actual tool wear using appropriate measuring equipment during machine tool processing, when the machine tool is in a non-cutting state, such as during tool changes, workpiece loading and unloading, measurement, or short machine pauses. This machining interval provides favorable conditions for obtaining accurate tool wear data and avoids interference that may occur during the cutting process. Actual tool wear can refer to parameters such as the wear width, wear depth, or wear volume of the tool's rake face, flank face, or tool tip.

[0070] Specifically, comparing the actual tool wear with the total tool wear to obtain the deviation refers to comparing the total tool wear accumulated through the above technical solution (i.e., the predicted wear) with the actual tool wear obtained during the machining clearance, and calculating the difference between the two. This deviation can be an absolute value, a relative percentage, or a value calculated through a specific error function, used to quantify the inconsistency between the prediction model and the actual situation.

[0071] In practical applications, identifying transient extreme wear acceleration events related to the deviation when it meets preset conditions means that when the calculated deviation reaches or exceeds a preset threshold, or when the deviation's trend conforms to a specific pattern, the system will trigger an event identification mechanism. This mechanism will backtrack historical processing data and analyze transient extreme wear acceleration events that occurred during the processing stage that led to the deviation. These events are identified according to the method described in the above technical solution, and their characteristic parameters include, but are not limited to, peak temperature, peak vibration energy, and duration. Identifying deviation-related events aims to determine which specific transient wear events contribute most to the final predicted deviation.

[0072] Furthermore, based on the deviation and the characteristic parameters of the transient extreme wear acceleration event, iteratively adjusting the coefficients in the nonlinear mapping function means taking the overall deviation information and the specific characteristic parameters of the identified transient extreme wear acceleration event as input, and correcting the coefficients in the nonlinear mapping function through an iterative optimization process. The nonlinear mapping function is a model used to correlate micro-environment parameters with tool wear rate, and its coefficients determine the model's response characteristics. Iterative adjustment can employ various optimization algorithms, such as gradient descent, genetic algorithms, or particle swarm optimization, to gradually reduce prediction deviation. By combining adjustments with specific transient event characteristics, the nonlinear mapping function can more accurately capture the micro-mechanisms leading to wear acceleration, thereby improving the model's prediction accuracy and adaptability.

[0073] In some preferred embodiments, a specific example is given below. Assume that in a milling task, the machine tool enters the machining gap after completing the roughing stage of a workpiece. At this time, the wear width of the tool's flank face is measured by a laser displacement sensor integrated into the machine tool, obtaining an actual tool wear of 0.25 mm. Simultaneously, the system predicts a total tool wear of 0.20 mm based on the aforementioned method. Comparing the two, a deviation of 0.05 mm is obtained. A preset condition is set to trigger optimization when the deviation exceeds 0.03 mm. Since 0.05 mm is greater than 0.03 mm, the system is triggered and begins to backtrack the monitoring data from the roughing stage. By analyzing historical infrared temperature signals and high-frequency acceleration signals, the system identifies two major transient extreme wear acceleration events that occurred during the roughing process, one with a high peak temperature and the other with a large peak vibration energy. The system takes the characteristic parameters of these two events (e.g., peak temperature, peak vibration energy, duration) and a deviation of 0.05 mm as input, and uses a genetic algorithm-based optimizer to iteratively adjust the weight coefficients in the nonlinear mapping function. After several iterations, the coefficients of the nonlinear mapping function are corrected, making the predicted wear more closely approximate the actual wear of 0.25 mm when simulating wear during this roughing stage. In this way, subsequent tool wear predictions will be more accurate, providing a more reliable basis for updating tool life and deciding on compensation strategies.

[0074] Optionally, when at least two transient extreme wear acceleration events are identified consecutively, the steps of extracting the total duration of all transient extreme wear acceleration events, the highest temperature peak within the sequence, the highest vibration energy peak, and the average time interval between events include: Identify transient, extreme wear acceleration events; Record the start time, end time, peak temperature, and peak vibration energy of transient extreme wear acceleration events; Maintain the sliding time window and calculate the occurrence density of transient extreme wear acceleration events within the sliding time window in real time; When the density continuously exceeds a preset threshold within a preset time period, the first transient extreme wear acceleration event in the corresponding region is marked as the starting point of the continuous sequence. The occurrence of transient extreme wear acceleration events is continuously tracked until the occurrence density is consistently below a preset threshold, or the total duration of the continuous sequence reaches the preset maximum sequence duration. The last transient extreme wear acceleration event is then marked as the end point of the continuous sequence. Based on the start and end times of each transient extreme wear acceleration event, the total duration of the continuous sequence is extracted; Based on the peak temperature and peak vibration energy of each transient extreme wear acceleration event, the highest temperature peak and the highest vibration energy peak of each transient extreme wear acceleration event in the continuous sequence are identified. The weighted average of the start time differences of all adjacent transient extreme wear acceleration events within a continuous sequence is calculated as the average time interval between events.

[0075] Specifically, after identifying transient extreme wear acceleration events, each identified event needs to be recorded in detail. The record includes the event's start and end times, as well as the peak temperature and peak vibration energy reached during the event. This detailed time and energy information forms the basis for subsequent analysis and sequence feature extraction. To more accurately determine the continuity of events, a sliding time window mechanism is introduced. The sliding time window is a fixed-length time period that moves along the time axis. The system calculates the occurrence density of transient extreme wear acceleration events within this window in real time. Occurrence density can be understood as the ratio of the number of transient extreme wear acceleration events detected within the current sliding time window to the window length, used to quantify the intensity of events within a specific time period.

[0076] In practical applications, when the occurrence density remains consistently above a preset threshold within a preset time period, it indicates that the frequency of transient extreme wear acceleration events is high and stable within that period. In this case, the first transient extreme wear acceleration event in the corresponding region is marked as the starting point of the continuous sequence. This ensures that only after events occur densely and persist for a period of time is it considered a true continuous sequence. Once the sequence starting point is marked, the system will continuously track the occurrence of subsequent transient extreme wear acceleration events. The tracking process will continue until the occurrence density consistently falls below the preset threshold, indicating that the dense occurrence of events has ended; or until the total duration of the continuous sequence reaches the preset maximum sequence duration, which is used to prevent a single sequence from becoming too long and losing analytical significance. When either condition is met, the last tracked transient extreme wear acceleration event is marked as the end point of the continuous sequence.

[0077] After determining the start and end points of the continuous sequence, the total duration of the sequence can be calculated and extracted based on the start and end times of all transient extreme wear acceleration events within the sequence. For example, the total duration can be the difference between the start time of the first event and the end time of the last event in the sequence, or the durations of all events in the sequence can be summed. Simultaneously, by iterating through the peak temperature and peak vibration energy of all transient extreme wear acceleration events within the continuous sequence, the highest temperature peak and highest vibration energy peak are identified, representing the moments of most intense wear acceleration in the sequence. Furthermore, to quantify the closeness between events, the weighted average of the start time differences of all adjacent transient extreme wear acceleration events within the continuous sequence is calculated as the average time interval between events. The weighted average can be weighted according to the intensity or duration of the events to more accurately reflect the dynamic characteristics of the event sequence.

[0078] Optionally, the step of generating the sequence wear equivalent by consulting a nonlinear damage superposition mapping table based on the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval includes: Calculate the local geometric feature parameters of the current cutting point based on the three-dimensional coordinates of the current cutting point and the CAD model of the workpiece; Based on local geometric feature parameters, obtain the geometric constraint influence factor; The geometric constraint influence factor is weighted and fused with the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval to obtain the corrected sequence characteristic parameters. Based on the corrected sequence characteristic parameters, consult the nonlinear damage superposition mapping table to generate the sequence wear equivalent.

[0079] Specifically, during machine tool processing, it is necessary to acquire the three-dimensional coordinates of the current contact point between the tool and the workpiece in real time or near real time. This coordinate information, combined with a pre-loaded workpiece CAD model, can be used to calculate the local geometric feature parameters of the current cutting point through geometric analysis algorithms. These local geometric feature parameters may include, but are not limited to, curvature, slope, rate of change of depth of cut, material removal rate, and the contact angle between the cutting edge and the workpiece surface. For example, for machining a curved surface, its radius of curvature and normal vector are important local geometric feature parameters.

[0080] Among these, the geometric constraint influence factor can be obtained based on local geometric feature parameters. The geometric constraint influence factor is a parameter that quantifies the degree to which local geometric features affect tool wear. For example, when cutting acute angles or small-radius arcs, the tool experiences concentrated stress, making heat dissipation difficult and potentially exacerbating wear; in this case, the geometric constraint influence factor will increase accordingly. The geometric constraint influence factor can be obtained through a pre-established mapping table between geometric features and wear effects, empirical formulas, or models trained based on simulation data such as finite element analysis.

[0081] In practical applications, the geometric constraint influence factor is weighted and fused with the total duration, the highest temperature peak within the sequence, the highest vibration energy peak, and the average time interval to obtain the corrected sequence characteristic parameters. This weighted fusion aims to comprehensively consider the influence of the local geometric environment on wear and the characteristics of the transient extreme wear acceleration event itself. For example, linear weighting, nonlinear weighting, or fusion methods based on machine learning models can be used. The weighting coefficients can be trained and optimized based on actual processing data to ensure the effectiveness of the fusion.

[0082] Therefore, based on the corrected sequence characteristic parameters, the nonlinear damage superposition mapping table is consulted to generate the sequence wear equivalent. The nonlinear damage superposition mapping table is a multidimensional lookup table or function whose input parameters are expanded from the original event characteristic parameters to corrected sequence characteristic parameters that include geometric constraint influence factors. In this way, the mapping table can more comprehensively consider the combined influence of various factors on tool wear, thus outputting a more accurate sequence wear equivalent.

[0083] In some preferred embodiments, a specific example is given below. Assume a machine tool is machining an aero-engine blade with a complex curved surface and multiple acute angles. During machining, the system acquires the three-dimensional coordinates of the cutting point in real time and, combined with the blade's CAD model, calculates the local geometric feature parameters of the current cutting point. For example, when the tool enters a groove with a small radius of curvature from a flat area, the system identifies parameters such as the radius of curvature and the rate of change of cutting depth in that area. Based on these parameters, the corresponding geometric constraint influence factor is obtained through preset mapping rules or models. For example, in a small-radius groove, due to difficulties in chip removal and stress concentration, the geometric constraint influence factor might be set to 1.2, indicating that wear may be increased by 20%.

[0084] Simultaneously, the system continuously monitors the microenvironmental parameters of the cutting zone and identifies transient extreme wear acceleration events. Assuming that a series of transient extreme wear acceleration events are detected shortly after entering the groove, their total duration, the highest temperature peak within the sequence, the highest vibration energy peak, and the average time interval between events are recorded. Subsequently, the geometric constraint influence factor 1.2 obtained above is weighted and fused with these event characteristic parameters. For example, the geometric constraint influence factor can be simply multiplied by the original wear equivalent calculation result, or a more complex nonlinear function can be used for fusion to obtain the corrected sequence characteristic parameters. Finally, the system uses these corrected sequence characteristic parameters to consult a nonlinear damage superposition mapping table, thereby generating a more accurate sequence wear equivalent. This corrected sequence wear equivalent will more accurately reflect the actual wear experienced by the tool in this complex geometric region, avoiding underestimation or overestimation of wear due to neglecting geometric factors, thus providing a more reliable basis for subsequent tool life management and machining parameter adjustment.

[0085] Optionally, the steps for obtaining the geometric constraint influence factor based on local geometric feature parameters include: Real-time monitoring of acoustic emission signals and spindle torque signals in the cutting zone; Frequency domain analysis is performed on the acoustic emission signal to extract the energy distribution of characteristic frequency bands; Wavelet decomposition is performed on the spindle torque signal to extract high-frequency components; Based on energy distribution, high-frequency components, and local geometric characteristic parameters, the geometric constraint influence factor is obtained and adjusted.

[0086] Specifically, real-time monitoring of acoustic emission signals in the cutting zone refers to continuously acquiring elastic wave signals generated during the cutting process using acoustic emission sensors installed on the machine tool spindle or workpiece fixture. These signals can reflect dynamic processes such as microscopic friction, wear, and crack propagation in the contact area between the tool and the workpiece. Real-time monitoring of the spindle torque signal is achieved through a torque sensor on the spindle motor. This signal directly reflects the magnitude and changes in cutting force and is closely related to cutting load, tool condition, and material removal rate.

[0087] Frequency domain analysis of acoustic emission signals, extracting the energy distribution of characteristic frequency bands, can be understood as converting the acquired time-domain acoustic emission signals to the frequency domain using methods such as Fourier transform, and analyzing their energy distribution across different frequency ranges. For example, energy increases in certain specific frequency bands may be related to phenomena such as tool wear, chip formation, or microcrack propagation. By identifying energy changes in these characteristic frequency bands, microscopic events during the cutting process can be captured more precisely.

[0088] In practical applications, wavelet decomposition of the spindle torque signal to extract high-frequency components refers to using wavelet transform to decompose the spindle torque signal into components of different frequency scales. High-frequency components typically contain information about transient impacts, vibrations, and rapidly changing interactions between the tool and workpiece during the cutting process. This information is crucial for identifying phenomena such as tool chipping and chatter. By extracting these high-frequency components, subtle changes in the cutting state can be reflected more sensitively.

[0089] Therefore, obtaining and adjusting the geometric constraint influence factor based on energy distribution, high-frequency components, and local geometric characteristic parameters involves comprehensively analyzing the energy distribution of characteristic frequency bands extracted from the acoustic emission signal, the high-frequency components extracted from the spindle torque signal, and pre-calculated local geometric characteristic parameters (such as the curvature, slope, and contact angle of the cutting point). These dynamic signals provide real-time information about the actual physical interaction of the cutting area, which can be used to correct or adjust the initial geometric constraint influence factor obtained solely based on geometric parameters, making it more accurately reflect the impact of geometric constraints on wear under the current cutting state.

[0090] Optionally, when dealing with complex geometric regions, the steps for obtaining and adjusting the geometric constraint influence factor based on energy distribution, high-frequency components, and local geometric characteristic parameters include: Real-time time-frequency analysis of acoustic emission signals is performed to extract the energy distribution and duration of transient shock waves; High-order spectral analysis was performed on the spindle torque signal to extract non-Gaussian characteristics and bispectral correlation. The transient shock wave's energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric characteristic parameters are used as inputs. Through a nonlinear mapping function, the correction amount of the geometric constraint influence factor is calculated and output in real time, and then superimposed on the initially obtained geometric constraint influence factor.

[0091] Specifically, when machining in complex geometric regions, real-time time-frequency analysis of acoustic emission signals is performed to more accurately capture transient characteristics during the cutting process. Real-time time-frequency analysis refers to using methods such as short-time Fourier transform, wavelet transform, or Hilbert-Huang transform to analyze the frequency components of the acoustic emission signal and their changes over time. This allows for the precise identification and extraction of the energy distribution and duration of transient shock waves caused by instantaneous contact between the tool and workpiece, chip breaking, vibration, etc. These transient shock waves are important indicators of accelerated tool wear under complex cutting conditions.

[0092] Simultaneously, high-order spectral analysis is performed on the spindle torque signal. High-order spectral analysis, such as bispectral or trispectral analysis, can reveal the non-Gaussian and nonlinear coupling characteristics of the signal, which is crucial for understanding the sources of nonlinear vibration and impact loads in complex cutting processes. Through high-order spectral analysis, non-Gaussian features and bispectral correlations can be extracted from the spindle torque signal. These features reflect the nonlinear relationships between cutting force, vibration, and wear, providing deeper information for a more accurate assessment of geometric constraint influence factors.

[0093] Furthermore, the energy distribution and duration of the transient shock wave extracted through real-time time-frequency analysis, the non-Gaussian characteristics and bispectral correlation extracted through high-order spectral analysis, and the energy distribution, high-frequency components, and local geometric feature parameters of the aforementioned characteristic frequency bands are used as inputs. These multi-dimensional, high-precision features are input into a nonlinear mapping function. This nonlinear mapping function can be a machine learning model such as a multilayer perceptron, radial basis function network, or support vector machine. It learns the nonlinear relationship between the cutting characteristics of complex geometric regions and the geometric constraint influence factor in historical data, and calculates and outputs the correction amount of the geometric constraint influence factor in real time. This correction amount is then superimposed on the initially obtained geometric constraint influence factor to obtain a more accurate geometric constraint influence factor that is adapted to complex cutting conditions.

[0094] Optionally, the steps of taking the energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric characteristic parameters of the transient shock wave as input, calculating and outputting the correction amount of the geometric constraint influence factor in real time through a nonlinear mapping function, and superimposing the correction amount onto the initially obtained geometric constraint influence factor include: Obtain the actual tool wear at the end of the machining interval or at the end of a specific machining stage; By comparing the actual tool wear with the total tool wear, the deviation can be obtained. When the deviation exceeds the preset threshold, historical data is reviewed to identify the transient extreme wear acceleration event sequence that caused the deviation, and the characteristic parameters, local geometric characteristic parameters, energy distribution of characteristic frequency bands and high-frequency components of spindle torque signal of the transient extreme wear acceleration event sequence are extracted and integrated into the corresponding historical data. Adjust the weight parameters or activation function parameters in the nonlinear mapping function based on historical data and deviations; During the processing, monitor material batch information and machine tool status parameters; Based on material batch information and machine tool status parameters, when a change in material batch or a drift in machine tool status parameters is detected, an online adaptive adjustment mechanism for the nonlinear mapping function is triggered. Based on the online adaptive adjustment mechanism, the parameters of the nonlinear mapping function are fine-tuned according to the new material batch characteristics or machine tool state drift trend; The transient shock wave's energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution in the characteristic frequency band, high-frequency components of the spindle torque signal, and local geometric feature parameters are used as inputs. Through an adjusted nonlinear mapping function, the correction amount of the geometric constraint influence factor is calculated and output in real time, and then superimposed on the initially obtained geometric constraint influence factor.

[0095] Specifically, obtaining the actual tool wear at the end of a machining interval or a specific machining stage refers to accurately measuring the actual tool wear through offline measurement methods, such as microscopic observation or laser scanning. This actual tool wear is a crucial basis for calibrating and optimizing the model. The deviation is obtained by comparing the actual tool wear with the total tool wear, where the total tool wear is a predicted value obtained by accumulating the conventional wear and event wear equivalent using the method described above. By comparing the actual measured value with the predicted value, the prediction accuracy of the current model can be quantified. When the deviation exceeds a preset threshold, historical data is reviewed to identify the transient extreme wear acceleration event sequence that caused the deviation. The characteristic parameters, local geometric characteristic parameters, energy distribution of characteristic frequency bands, and high-frequency components of the spindle torque signal of the transient extreme wear acceleration event sequence are extracted and integrated into the corresponding historical data. The preset threshold can be set based on experience or statistical analysis to determine whether the model prediction has significant errors. Reviewing historical data aims to identify specific machining events and conditions related to large deviations. This historical data includes the original signal characteristics that led to the wear acceleration event, the geometric characteristics of the machining area, and the frequency domain and high-frequency components of the acoustic emission and torque signals. This information is crucial for understanding the causes of the deviation.

[0096] Further, based on historical data and deviations, the weight parameters or activation function parameters in the nonlinear mapping function are adjusted. This step is an offline optimization process. By analyzing patterns in historical data that lead to prediction deviations, the internal parameters of the nonlinear mapping function (e.g., neural networks, support vector machines) are adjusted to improve its prediction accuracy under similar conditions. Weight parameters and activation function parameters are key elements affecting the behavior of the nonlinear mapping function. During processing, material batch information and machine tool status parameters are monitored. Material batch information may include the material's hardness, toughness, chemical composition, etc., while machine tool status parameters may include spindle speed, feed rate, coolant flow rate, vibration level, temperature, etc. Changes in these parameters may directly affect tool wear behavior. Based on material batch information and machine tool status parameters, when a change in the material batch or a drift in the machine tool status parameters is detected, the online adaptive adjustment mechanism of the nonlinear mapping function is triggered. The online adaptive adjustment mechanism refers to the system's ability to automatically adjust model parameters in real time or near real time according to changes in the external environment or internal state during processing. For example, when a change in the hardness of a new batch of workpiece material is detected, the system will immediately initiate adjustment. Based on an online adaptive adjustment mechanism, the parameters of the nonlinear mapping function are fine-tuned according to new material batch characteristics or machine tool state drift trends. Fine-tuning refers to making small, gradual adjustments to the parameters to adapt to the latest machining conditions and avoid system instability caused by parameter mutations. Finally, the energy distribution, duration, non-Gaussianity characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components of the spindle torque signal, and local geometric feature parameters of the transient shock wave are used as inputs. The adjusted nonlinear mapping function calculates and outputs the correction amount of the geometric constraint influence factor in real time, and then superimposes the correction amount onto the initially acquired geometric constraint influence factor. This means that the nonlinear mapping function, after offline optimization and online adaptive adjustment, can more accurately calculate the correction amount of the geometric constraint influence factor, thereby improving the overall accuracy of tool wear prediction.

[0097] This application also discloses a machine tool processing control system, combined with... Figure 3 As shown, the machine tool processing control system 1 includes: The microenvironment parameter acquisition module 11 is used to acquire and monitor the microenvironment parameters of the cutting area during machine tool processing. The work information acquisition module 12 is used to acquire the tool type, workpiece material properties and preset cutting parameters corresponding to the machine tool; The judgment criteria establishment module 13 is used to establish the initial judgment criteria for tool wear rate based on tool type, workpiece material properties and preset cutting parameters; The wear rate correction module 14 is used to correct the initial judgment criteria based on the deviation between the micro-environment parameters and the preset standard values, and obtain the corrected tool wear rate. The judgment basis optimization module 15 is used to optimize the rules for subsequent correction of the initial judgment basis based on the actual tool wear amount. The compensation strategy execution module 16 is used to update the remaining tool life based on the corrected tool wear rate, and when the tool wear reaches a preset threshold, it decides and executes a compensation strategy for cutting parameters or tool path based on the value of the workpiece being processed and the remaining processing tasks.

[0098] Specifically, the microenvironment parameter acquisition module is used to acquire and monitor the microenvironment parameters of the cutting area during machine tool processing. The specific methods and examples for acquiring and monitoring the microenvironment parameters of the cutting area have been described in detail in the above embodiments and will not be repeated here. It is important to emphasize that this microenvironment parameter acquisition module can consist of one or more sensor units, a data acquisition unit, and a preprocessing unit. Sensor units may include, for example, infrared temperature sensors, acoustic emission sensors, vibration sensors, force sensors, etc., which are configured near the cutting area to acquire raw physical signals in real time. The data acquisition unit is responsible for converting these analog signals into digital signals and performing preliminary data cleaning and formatting. The preprocessing unit performs filtering, noise reduction, and other operations on the acquired data to ensure data quality and usability. These units can be integrated into a single hardware module or deployed in a distributed manner and connected via a communication interface.

[0099] The work information acquisition module is used to acquire the tool type, workpiece material properties, and preset cutting parameters corresponding to the machine tool. The specific methods and examples for acquiring these information have been described in detail in the above embodiments and will not be repeated here. It is important to emphasize that this work information acquisition module can be a data interface or communication unit used to automatically read relevant information from the machine tool CNC system, tool management database, production management system, or CAD / CAM software. For example, the tool type can be obtained from the tool magazine management system, the workpiece material properties can be queried from the material database, and the preset cutting parameters can be directly parsed from the machining program. This module may also include a user interface, allowing operators to manually input or confirm this information. The purpose is to ensure that the system can acquire all necessary background information to establish accurate wear judgment criteria.

[0100] The judgment criterion establishment module is used to establish initial judgment criteria for tool wear rate based on tool type, workpiece material properties, and preset cutting parameters. The specific methods and examples for establishing the initial judgment criteria for tool wear rate have been described in detail in the above embodiments and will not be repeated here. It is important to emphasize that this judgment criterion establishment module can be a software algorithm unit, internally storing a large number of empirical formulas, lookup tables, or pre-trained machine learning models. Upon receiving the tool type, workpiece material properties, and preset cutting parameters, this module calculates a preliminary estimate of the tool wear rate based on these inputs. For example, it can be calculated based on Taylor's tool life formula or a more complex physical model. The implementation of this module can be based on a software program on an embedded processor or industrial PC.

[0101] The wear rate correction module is used to correct the initial judgment criteria based on the deviation of micro-environment parameters from preset standard values, thereby obtaining the corrected tool wear rate. The specific methods and examples for correcting the initial judgment criteria based on the deviation of micro-environment parameters from preset standard values ​​have been described in detail in the above embodiments, and will not be repeated here. It is important to emphasize that this wear rate correction module receives real-time data from the micro-environment parameter acquisition module and initial judgment criteria from the judgment criteria establishment module. Internally, it includes a correction algorithm that dynamically adjusts the initial judgment criteria based on the degree to which micro-environment parameters (such as temperature, vibration, etc.) deviate from their preset standard values. For example, when the temperature in the cutting area is too high, the correction algorithm increases the estimated wear rate. The implementation of this module can be based on software logic for real-time data processing, ensuring that the wear rate prediction can adapt to changes in the machining environment in real time.

[0102] The judgment criterion optimization module is used to optimize the rules for subsequent correction of the initial judgment criterion based on the actual tool wear. The specific methods and examples for optimizing the rules for subsequent correction of the initial judgment criterion based on the actual tool wear have been described in detail in the above embodiments, and will not be repeated here. It is important to emphasize that this judgment criterion optimization module receives actual tool wear data (e.g., obtained through offline measurement or online visual inspection) after the machining gap or a specific machining stage. It compares the actual wear with the wear predicted by the system and adjusts the correction rules or model parameters within the wear rate correction module based on the deviation between the two. For example, if the system continuously underestimates wear, this module will adjust the correction coefficient to make it more inclined to overestimate wear in the future. This module can be implemented using machine learning or adaptive control algorithms to achieve continuous learning and optimization of the wear prediction model.

[0103] The compensation strategy execution module updates the remaining tool life based on the corrected tool wear rate and, when the tool wear reaches a preset threshold, determines and executes a compensation strategy for cutting parameters or toolpaths, taking into account the value of the workpiece and the remaining machining tasks. The specific methods and examples for updating the remaining tool life based on the corrected tool wear rate and determining and executing a compensation strategy for cutting parameters or toolpaths, when the tool wear reaches a preset threshold, have been described in detail in the above embodiments and will not be repeated here. It is important to emphasize that this compensation strategy execution module receives the corrected tool wear rate and calculates the remaining tool life accordingly. When the remaining tool life reaches the preset threshold, the module comprehensively considers the current workpiece value (e.g., obtained through a production management system) and the remaining machining tasks (e.g., obtained through a machining program or production plan), intelligently selects and sends instructions to the machine tool CNC system to adjust cutting parameters (e.g., reduce feed rate, depth of cut) or modify the toolpath. This module can be implemented as an intelligent decision-making unit, making decisions through a preset strategy library or a rule-based expert system and communicating with the machine tool CNC system in real time.

[0104] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A machine tool processing control method, characterized in that, include: During machine tool processing, acquire and monitor the microenvironmental parameters of the cutting area; Obtain the tool type, workpiece material properties, and preset cutting parameters corresponding to the machine tool; Based on tool type, workpiece material properties, and preset cutting parameters, an initial basis for judging tool wear rate is established. Based on the deviation between the microenvironment parameters and the preset standard values, the initial judgment criteria are corrected to obtain the corrected tool wear rate. Based on the actual tool wear, optimize the rules for subsequent correction of the initial judgment criteria; Based on the corrected tool wear rate, the remaining tool life is updated, and when the tool wear reaches a preset threshold, a compensation strategy for cutting parameters or toolpaths is decided and executed, taking into account the value of the workpiece being machined and the remaining machining tasks.

2. The machine tool processing control method according to claim 1, characterized in that, The step of optimizing the rules for subsequent correction of the initial judgment based on the actual tool wear includes: Capture the original analog signal of the cutting area; When the instantaneous rate of change of the original analog signal exceeds a preset multiple standard deviation of the historical average rate of change within a preset time period, the storage of the instantaneous data block is triggered. Multi-dimensional feature extraction and correlation analysis are performed on the instantaneous data block to calculate the maximum instantaneous rise slope and peak temperature of the infrared temperature signal, and Fourier transform analysis is performed on the high-frequency acceleration signal to extract the energy spectral density and total energy peak of the ultrasonic band. Transient extreme wear acceleration events are identified based on the maximum instantaneous rise slope, the peak temperature, the energy spectral density, the total energy peak value, and the peak time synchronization of all features. Based on the peak temperature and energy spectral density of the ultrasonic band of the transient extreme wear acceleration event, the transient failure of the cutting fluid micro-interface is inferred; Based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of the transient extreme wear acceleration event, an event wear equivalent is generated; The total tool wear is obtained by accumulating the regular wear amount and the wear equivalent of the event. Based on the deviation between the actual tool wear and the total tool wear, the coefficients in the nonlinear mapping function are adjusted to optimize and correct the rules used for the initial judgment.

3. The machine tool processing control method according to claim 2, characterized in that, The step of generating the event wear equivalent based on the transient failure of the cutting fluid micro-interface and the intensity characteristics of the transient extreme wear acceleration event includes: Identify transient extreme wear acceleration events; Record the peak temperature, peak vibration energy, and duration of the transient extreme wear acceleration event; Determine whether at least two transient extreme wear acceleration events are continuously identified within a preset time window; When at least two transient extreme wear acceleration events are identified consecutively, the total duration of all transient extreme wear acceleration events, the highest temperature peak value within the sequence, the highest vibration energy peak value, and the average time interval between events are extracted. Based on the total duration, the highest temperature peak within the sequence, the highest vibration energy peak, and the average time interval, a nonlinear damage superposition mapping table is consulted to generate the sequence wear equivalent. The sequence wear equivalent is used as the event wear equivalent.

4. The machine tool processing control method according to claim 2, characterized in that, The step of optimizing the rules for subsequent correction of the initial judgment based on the actual tool wear includes: Obtain the actual tool wear during machining gaps; The deviation is obtained by comparing the actual tool wear with the total tool wear. When the deviation meets the preset conditions, identify the transient extreme wear acceleration event related to the deviation; The coefficients in the nonlinear mapping function are iteratively adjusted based on the deviation and the characteristic parameters of the transient extreme wear acceleration event.

5. A machine tool processing control method according to claim 3, characterized in that, The step of extracting the total duration, the highest temperature peak, the highest vibration energy peak, and the average time interval between all transient extreme wear acceleration events when at least two such events are consecutively identified includes: Identify transient extreme wear acceleration events; Record the start time, end time, peak temperature, and peak vibration energy of the transient extreme wear acceleration event; Maintain a sliding time window and calculate the occurrence density of transient extreme wear acceleration events within the sliding time window in real time; When the occurrence density is continuously higher than a preset threshold within a preset time period, the first transient extreme wear acceleration event in the corresponding region is marked as the starting point of the continuous sequence. The occurrence of transient extreme wear acceleration events is continuously tracked until the occurrence density is consistently below a preset threshold, or the total duration of the continuous sequence reaches a preset maximum sequence duration. The last transient extreme wear acceleration event is then marked as the end point of the continuous sequence. Based on the start and end times of each transient extreme wear acceleration event, the total duration of the continuous sequence is extracted; Based on the peak temperature and peak vibration energy of each transient extreme wear acceleration event, the highest temperature peak and the highest vibration energy peak of each transient extreme wear acceleration event in the continuous sequence are identified. The weighted average of the start time differences of all adjacent transient extreme wear acceleration events within a continuous sequence is calculated as the average time interval between events.

6. The machine tool processing control method according to claim 3, characterized in that, The step of generating the sequence wear equivalent by consulting a nonlinear damage superposition mapping table based on the total duration, the highest temperature peak within the sequence, the highest vibration energy peak, and the average time interval includes: Calculate the local geometric feature parameters of the current cutting point based on the three-dimensional coordinates of the current cutting point and the CAD model of the workpiece; Based on the local geometric feature parameters, obtain the geometric constraint influence factor; The geometric constraint influence factor is weighted and fused with the total duration, the highest temperature peak in the sequence, the highest vibration energy peak, and the average time interval to obtain the corrected sequence feature parameters; Based on the corrected sequence characteristic parameters, consult the nonlinear damage superposition mapping table to generate the sequence wear equivalent.

7. A machine tool processing control method according to claim 6, characterized in that, The step of obtaining the geometric constraint influence factor based on the local geometric feature parameters includes: Real-time monitoring of acoustic emission signals and spindle torque signals in the cutting zone; Frequency domain analysis is performed on the acoustic emission signal to extract the energy distribution of characteristic frequency bands; The spindle torque signal is subjected to wavelet decomposition to extract high-frequency components; Based on the energy distribution, the high-frequency components, and the local geometric feature parameters, the geometric constraint influence factor is obtained and adjusted.

8. A machine tool processing control method according to claim 7, characterized in that, When in a complex geometric region, the step of obtaining and adjusting the geometric constraint influence factor based on the energy distribution, the high-frequency components, and the local geometric feature parameters includes: Real-time time-frequency analysis is performed on the acoustic emission signal to extract the energy distribution and duration of the transient shock wave; High-order spectral analysis was performed on the spindle torque signal to extract non-Gaussian characteristics and bispectral correlation. The energy distribution, duration, non-Gaussianity characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric feature parameters of the transient shock wave are used as inputs. Through a nonlinear mapping function, the correction amount of the geometric constraint influence factor is calculated and output in real time, and the correction amount is superimposed on the initially obtained geometric constraint influence factor.

9. A machine tool processing control method according to claim 8, characterized in that, The steps of taking the energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components, and local geometric characteristic parameters of the transient shock wave as inputs, calculating and outputting the correction amount of the geometric constraint influence factor in real time through a nonlinear mapping function, and superimposing the correction amount onto the initially obtained geometric constraint influence factor include: Obtain the actual tool wear at the end of the machining interval or at the end of a specific machining stage; The deviation is obtained by comparing the actual tool wear with the total tool wear. When the deviation exceeds a preset threshold, historical data is reviewed to identify the transient extreme wear acceleration event sequence that caused the deviation, and the characteristic parameters, local geometric characteristic parameters, energy distribution of characteristic frequency bands and high-frequency components of spindle torque signal of the transient extreme wear acceleration event sequence are extracted and integrated into the corresponding historical data. Based on the historical data and the deviation, adjust the weight parameters or activation function parameters in the nonlinear mapping function; During the processing, monitor material batch information and machine tool status parameters; Based on material batch information and machine tool status parameters, when a change in material batch or a drift in machine tool status parameters is detected, an online adaptive adjustment mechanism for the nonlinear mapping function is triggered. Based on the online adaptive adjustment mechanism, the parameters of the nonlinear mapping function are fine-tuned according to the new material batch characteristics or machine tool state drift trend; The energy distribution, duration, non-Gaussian characteristics, bispectral correlation, energy distribution of characteristic frequency bands, high-frequency components of spindle torque signal, and local geometric feature parameters of transient shock wave are used as inputs. The correction amount of geometric constraint influence factor is calculated and output in real time through the adjusted nonlinear mapping function, and the correction amount is superimposed on the initially obtained geometric constraint influence factor.

10. A machine tool processing control system for performing machine tool processing control, characterized in that, include: The microenvironment parameter acquisition module is used to acquire and monitor the microenvironment parameters of the cutting area during machine tool processing. The work information acquisition module is used to acquire the tool type, workpiece material properties, and preset cutting parameters corresponding to the machine tool; The judgment criteria establishment module is used to establish the initial judgment criteria for tool wear rate based on tool type, workpiece material properties, and preset cutting parameters. The wear rate correction module is used to correct the initial judgment criteria based on the deviation between the microenvironment parameters and the preset standard values, so as to obtain the corrected tool wear rate. The judgment criterion optimization module is used to optimize the rules for subsequent correction of the initial judgment criterion based on the actual tool wear amount. The compensation strategy execution module is used to update the remaining tool life based on the corrected tool wear rate, and when the tool wear reaches a preset threshold, it decides and executes a compensation strategy for cutting parameters or tool path based on the value of the workpiece being machined and the remaining machining tasks.

Images