Cutter cutting force control method and system for special-shaped tooth machining

Abstract

The invention discloses a tool cutting force control method and system for special-shaped tooth machining, and relates to the technical field of cutting force control. The method comprises the steps that machining parameters of a special-shaped tooth workpiece are obtained, cutting analysis is conducted according to tool geometrical characteristic parameters in combination with a machining parameter set, and the initial cutting force threshold range is determined; collecting dynamic cutting force data of the cutter in real time to generate a real-time cutting force frequency spectrum; and the real-time cutting force frequency spectrum is dynamically compared according to the initial cutting force threshold range, cutting force fluctuation characteristics are extracted for tool abrasion analysis, an abrasion compensation coefficient is obtained, cutting force real-time correction is conducted on the special-shaped tooth workpiece, and a tool cutting force control suggestion is generated. The technical problems that in the prior art, cutting force in the special-shaped tooth machining process cannot be sensed and effectively controlled in real time, so that tool abrasion is fast, and machining precision is low are solved, and the technical effects that the cutting force of the tool is dynamically adjusted and controlled, the service life of the tool is prolonged, and the machining stability and precision of the special-shaped tooth are improved are achieved.

Description

Technical Field

[0001] This invention relates to the field of cutting force control technology, and more specifically to a cutting force control method and system for machining irregularly shaped teeth. Background Technology

[0002] In the precision machining of irregularly shaped gear parts, due to the complex tooth structure and variable stress state, the cutting tool is prone to problems such as local overload and stress concentration during cutting, resulting in significant fluctuations in cutting force, which in turn exacerbates tool wear and affects machining accuracy. Traditional machining strategies mostly rely on fixed process parameter settings, lacking dynamic perception and feedback adjustment of the actual cutting state. This makes it difficult to adapt to the constantly changing stress characteristics and machining stages during the machining of irregularly shaped gears, adversely affecting machining stability and tool life. Summary of the Invention

[0003] This application provides a cutting force control method and system for machining irregularly shaped teeth, which addresses the technical problem that existing technologies cannot effectively sense and control the cutting force during the machining process of irregularly shaped teeth, resulting in rapid tool wear and low machining accuracy.

[0004] In view of the above problems, this application provides a cutting force control method and system for machining irregularly shaped teeth.

[0005] The first aspect of this application provides a method for controlling the cutting force of a tool for machining irregularly shaped teeth, the method comprising: The machining parameters of the irregular-shaped tooth workpiece are obtained, and cutting analysis is performed according to the tool geometric feature parameters and the machining parameter set to determine the initial cutting force threshold range. The dynamic cutting force data of the tool is collected in real time to generate a real-time cutting force spectrum. The real-time cutting force spectrum is dynamically compared with the initial cutting force threshold range to extract the cutting force fluctuation characteristics for tool wear analysis and obtain the wear compensation coefficient. The cutting force of the irregular-shaped tooth workpiece is corrected in real time according to the wear compensation coefficient to generate tool cutting force control suggestions.

[0006] A second aspect of this application provides a tool cutting force control system for machining irregularly shaped teeth, the system comprising: The cutting analysis module is used to acquire machining parameters of the irregular-shaped tooth workpiece, perform cutting analysis based on the tool's geometric characteristics and the machining parameter set, and determine the initial cutting force threshold range. The real-time data acquisition module is used to acquire dynamic cutting force data of the tool in real time and generate a real-time cutting force spectrum. The wear analysis module is used to dynamically compare the real-time cutting force spectrum with the initial cutting force threshold range, extract cutting force fluctuation characteristics for tool wear analysis, and obtain a wear compensation coefficient. The real-time correction module is used to correct the cutting force of the irregular-shaped tooth workpiece in real time according to the wear compensation coefficient and generate tool cutting force control suggestions.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application obtains the machining parameters of irregular-shaped tooth workpieces, performs cutting analysis based on tool geometric features and machining parameter sets, and determines the initial cutting force threshold range; it collects dynamic cutting force data of the tool in real time and generates a real-time cutting force spectrum; it dynamically compares the real-time cutting force spectrum with the initial cutting force threshold range, extracts cutting force fluctuation characteristics for tool wear analysis, and obtains a wear compensation coefficient; it then performs real-time correction of the cutting force of the irregular-shaped tooth workpiece according to the wear compensation coefficient, generating tool cutting force control suggestions. This invention solves the technical problem of existing technologies being unable to perceive and effectively control the cutting force during the machining of irregular-shaped teeth in real time, leading to rapid tool wear and low machining accuracy. By combining tool geometric features and machining parameters to model the cutting force, collecting cutting force data in real time and dynamically comparing and analyzing it, calculating the wear compensation coefficient, and implementing closed-loop correction control, it achieves dynamic regulation of tool cutting force, extends tool life, and improves the stability and accuracy of irregular-shaped tooth machining. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic flowchart of a tool cutting force control method for machining irregularly shaped teeth provided in an embodiment of this application; Figure 2 This is a schematic diagram of the cutting force control system for machining irregularly shaped teeth provided in an embodiment of this application.

[0010] Figure labeling: Cutting analysis module 11, real-time data acquisition module 12, wear analysis module 13, real-time correction module 14. Detailed Implementation

[0011] This application provides a cutting force control method and system for machining irregularly shaped teeth. It addresses the technical problem that existing technologies cannot effectively sense and control the cutting force during the machining process of irregularly shaped teeth, resulting in rapid tool wear and low machining accuracy. By combining tool geometry and machining parameters to model the cutting force, collect cutting force data in real time and perform dynamic comparison and analysis, calculate the wear compensation coefficient and implement closed-loop correction control, the technical effect of achieving dynamic regulation of tool cutting force, extending tool life and improving the stability and accuracy of irregularly shaped tooth machining is achieved.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, this application provides a cutting force control method for machining irregularly shaped teeth, the method comprising: Step S100: Obtain the machining parameters of the irregular tooth workpiece, perform cutting analysis according to the tool geometric feature parameters and the machining parameter set, and determine the initial cutting force threshold range.

[0015] In this embodiment of the application, the machining parameters of the irregular tooth workpiece are first obtained. These machining parameters are predetermined basic data, including information such as tooth size, material type, machining path, cutting depth and feed rate.

[0016] Subsequently, by analyzing the preset machining parameters for irregular-shaped teeth, key tooth profile features were extracted, and cutting prediction was performed in conjunction with tool geometry to calculate theoretical cutting force data. Then, actual cutting force data was collected using multiple sensors, and the theoretical values ​​were corrected through simulation to determine an initial cutting force threshold range that conforms to actual working conditions.

[0017] Furthermore, the method provided in the application embodiment, which involves obtaining the machining parameters of the irregularly shaped toothed workpiece, performing cutting analysis based on the tool geometric feature parameters and the machining parameter set, and determining the initial cutting force threshold range, also includes: Based on the machining parameters, the tooth profile of the irregular tooth workpiece is analyzed to obtain tooth profile feature parameters; cutting prediction is performed based on the tooth profile feature parameters and the tool geometric feature parameters to calculate theoretical cutting force data; actual cutting force data is dynamically collected through multiple sensors, and the theoretical cutting force data is simulated and corrected based on the actual cutting force data to determine the initial cutting force threshold range.

[0018] In this embodiment, a three-dimensional surface reconstruction method is first used to analyze the spatial structure of the irregularly shaped tooth workpiece. This method takes the addendum circle diameter, root circle radius, tooth pitch, tooth thickness, and module from the machining parameters as input, and uses NURBS modeling functions in a CAD platform to fit the tooth profile of the workpiece, obtaining a continuous spatial tooth surface representation model. Based on this, tooth profile feature parameters, including principal and secondary curvatures of the tooth surface, root transition curvature radius, and profile change rate, are extracted.

[0019] The extracted tooth profile features are then combined with the tool's geometric features (such as rake angle, clearance angle, cutting edge length, helix angle, and tip radius) to perform theoretical force prediction using the Kienzle empirical cutting force model. This model calculates the force required per unit cutting volume based on the product of the cutting force constant and the cutting thickness and width. By combining the cutting angle and feed path at different tooth surface positions, the spindle force, normal force, and radial force at that position are calculated to generate theoretical cutting force data covering the entire tooth profile.

[0020] In actual machining, triaxial strain gauge sensors installed in the tool holder and piezoelectric vibration sensors located in the workpiece fixture area are used to synchronously acquire real-time cutting force data in multiple directions at a sampling frequency of 2kHz or higher, obtaining actual cutting force data including time stamps and axial force components. The strain gauges record the stress response in the X, Y, and Z directions, while the vibration sensors capture high-frequency disturbances, effectively reflecting the dynamic changes during machining.

[0021] The theoretical cutting force data is then matched one-to-one with the actual collected data, and the difference between the two is calculated point by point. The average of all differences is then calculated for each tooth segment to obtain the average deviation for that segment. Subsequently, using the theoretical cutting force value as the center, this average deviation is extended upwards and downwards to form corresponding upper and lower boundaries. For example, if the theoretical value for a tooth segment is 180N and the average deviation is 15N, the system determines the cutting force threshold range for that segment to be 165N to 195N. In this way, an initial cutting force threshold range that conforms to actual machining conditions is constructed.

[0022] Step S200: Collect dynamic cutting force data of the tool in real time and generate a real-time cutting force spectrum.

[0023] In this embodiment, a triaxial strain sensor array is deployed on the cutting tool to collect dynamic cutting force data generated in real time between the cutting tool and the contact area of ​​the irregularly shaped toothed workpiece, obtaining cutting force components in multiple axes. Simultaneously, a vibration sensor is deployed on the workpiece to collect high-frequency vibration signals generated during machining. These two types of signals are fused in the time and frequency domains to construct a cutting force time-frequency feature matrix containing both time and frequency information. The matrix is ​​then decomposed according to preset frequency bands to extract the energy distribution within the corresponding frequency bands, ultimately generating a real-time cutting force spectrum reflecting changes in the tool's stress state.

[0024] Furthermore, the method provided in the application embodiments, which involves real-time acquisition of dynamic cutting force data of the tool and generation of a real-time cutting force spectrum, further includes: A triaxial strain sensor array is deployed on the cutting tool to sense and collect dynamic cutting force data, which includes multiple axial cutting force components. A vibration sensor is deployed on the irregularly shaped toothed workpiece to sense the machining process and obtain high-frequency vibration signals. The multiple axial cutting force components and the high-frequency vibration signals are fused in the time-frequency domain to construct a cutting force time-frequency feature matrix. Based on the cutting force time-frequency feature matrix, it is decomposed according to a preset frequency band to extract the energy distribution data of the preset frequency band, and the real-time cutting force spectrum is plotted based on the energy distribution data.

[0025] In this embodiment, a three-dimensional strain sensor array is first uniformly arranged around the circumference of the tool holder. Each sensor group has strain detection capabilities in the X, Y, and Z directions. Multiple sensors are arranged in a ring-shaped force measurement structure with an interval of 45°±2° to achieve comprehensive monitoring of the forces acting on the tool in each axis. These sensors are based on the strain resistance principle and obtain dynamic cutting force data by detecting the minute deformation of the tool caused by external forces during machining. This data includes the cutting force components in the X, Y, and Z directions at the contact area between the tool and the irregularly shaped toothed workpiece.

[0026] Simultaneously, a laser Doppler vibration meter is installed on the fixture structure of the irregularly shaped toothed workpiece. The laser beam focusing direction is at a 30° angle to the tool feed direction, enabling non-contact capture of micron-level vibration displacement on the workpiece surface. This vibration meter relies on the laser frequency shift principle to record the dynamic response behavior of the workpiece in real time when subjected to cutting disturbances. It is suitable for identifying high-frequency vibration fluctuations caused by asymmetrical tooth profiles, slight tool wear, or material inhomogeneity, thereby obtaining high-frequency vibration signals during machining.

[0027] Subsequently, multiple axial cutting force components and high-frequency vibration signals are fused in the time-frequency domain, and the original time series signals are processed using short-time Fourier transform. Specifically, each set of original signals is divided into equal-length time windows, and a fast Fourier transform (FFT) is performed within each window, thereby mapping the continuous time signal to the corresponding frequency space. This method obtains frequency variation information while maintaining time resolution, ultimately generating a three-dimensional structure composed of time, frequency, and energy amplitude, called the cutting force time-frequency characteristic matrix.

[0028] After constructing the cutting force time-frequency characteristic matrix, based on process laws and signal response characteristics, the frequency range is divided according to the sensitive intervals of machining mechanics changes. Specifically, the overall frequency range is divided into a low-frequency band (0~500Hz), a mid-frequency band (500~1500Hz), and a high-frequency band (1500~5000Hz). This preset frequency band setting is based on the fact that the low-frequency band is mainly used to capture basic cutting force fluctuations and process load changes; the mid-frequency band reflects local contact disturbances or tooth runout between the tool and workpiece; and the high-frequency band is most sensitive to high-frequency anomalies such as micro-wear, impact vibration, and chipping. Within each frequency band, energy data for the corresponding frequency interval in the cutting force time-frequency characteristic matrix is ​​extracted to generate a corresponding energy distribution sequence. Then, peak detection is performed on the energy sequence within each frequency band to identify the dominant frequency peak of that band, i.e., the frequency point where energy is most concentrated. This value typically reflects the main excitation frequency or structural resonance frequency of the current system. Simultaneously, by analyzing the spectral width near the dominant frequency, the bandwidth at that frequency point is calculated to measure the degree of frequency energy diffusion. A wider bandwidth usually indicates decreased signal stability or increased stress disturbance. Furthermore, the trend of energy density changes within different time segments is tracked to determine whether there is a continuous increase or sudden accumulation of energy in a certain frequency band. For example, if a continuous energy surge and peak shift occur in the high-frequency band during the tooth root machining stage, it may indicate the presence of microcracks or localized chipping in the tool at that stage.

[0029] After extracting the energy distribution, a real-time cutting force spectrum is plotted based on the energy density results for each frequency band. This spectrum uses time as the horizontal axis, frequency as the vertical axis, and color intensity as the unit energy density value. Color variations express the energy strength of each frequency component at different time points. A darker color indicates a more concentrated cutting energy at that frequency during that time period, thus clearly showing the evolution of mechanical behavior over time and frequency during machining.

[0030] Step S300: Dynamically compare the real-time cutting force spectrum according to the initial cutting force threshold range, extract the cutting force fluctuation characteristics for tool wear analysis, and obtain the wear compensation coefficient.

[0031] In this embodiment, the real-time cutting force spectrum is dynamically compared with the initial cutting force threshold range to identify and calibrate abnormal frequency bands exceeding the threshold in the spectrum, and cutting force fluctuation characteristics are extracted accordingly. Subsequently, tool stress analysis is performed based on the cutting force fluctuation characteristics to calculate the tool stress concentration factor reflecting local load concentration. Based on this, combined with the stress history and stress concentration distribution, tool degradation analysis is performed to quantify the cumulative tool wear.

[0032] Then, unstable change signals during the wear process are extracted to form wear fluctuation characteristics that reflect the nonlinear trend of wear. Through regression analysis of these wear fluctuation characteristics and wear accumulation, wear compensation coefficients for adaptive adjustment of processing parameters are generated.

[0033] Furthermore, in the method provided in the application embodiment, the real-time cutting force spectrum is dynamically compared according to the initial cutting force threshold range, cutting force fluctuation characteristics are extracted for tool wear analysis, and a wear compensation coefficient is obtained. The method further includes: The real-time cutting force spectrum is dynamically compared with the initial cutting force threshold range to identify abnormal frequency bands. The cutting force fluctuation characteristics are extracted based on the abnormal frequency bands. Tool stress analysis is performed based on the cutting force fluctuation characteristics to obtain the tool stress concentration factor. Tool degradation analysis is performed according to the stress concentration factor to obtain the tool wear accumulation. Feature extraction is performed based on the tool wear accumulation to obtain wear fluctuation characteristics. Regression analysis is performed between the tool wear accumulation and the wear fluctuation characteristics to generate the wear compensation coefficient.

[0034] In this embodiment, when dynamically comparing the real-time cutting force spectrum with the initial cutting force threshold range, the spectrum is first divided into N sub-bands according to a preset frequency band based on its multi-segment structure characteristics, and the energy density per unit time is calculated for each sub-band. By traversing the energy deviation between the N sub-bands and the upper limit of the initial threshold, a set of sub-band energy deviation indices is obtained, and anomaly identification is performed accordingly to initially identify multiple initial abnormal frequency bands. To improve the completeness of the frequency domain analysis, frequency domain continuity detection is performed on these initial abnormal frequency bands, and adjacent, overlapping, or short-term abrupt frequency bands are merged to finally determine the abnormal frequency bands used for diagnosing tool status.

[0035] Next, in-depth feature extraction is performed on the abnormal frequency band. The abnormal frequency band is decomposed into a low-frequency stable component and a high-frequency disturbance component, and their typical features are extracted separately. For the low-frequency component, the abnormal fundamental frequency amplitude and abnormal harmonic energy ratio are obtained through Fourier transform, and the cutting force stability index is calculated based on this, reflecting the stability of the force on the main frequency. For the high-frequency component, sudden peak signals are identified, and the corresponding instantaneous impact load is calculated by combining the geometric characteristic parameters of the tool. Finally, the stability index and impact load data are integrated to construct a cutting force fluctuation characteristic describing the stability of tool operation.

[0036] Subsequently, tool stress analysis was performed based on the cutting force fluctuation characteristics. By analyzing the abnormal fundamental frequency amplitude, abnormal harmonic energy ratio, and instantaneous impact load extracted from the abnormal frequency band, the stress situation of the tool in various directions was identified. These three types of stress characteristic data were synthesized into a three-dimensional cutting force variation map, and the von Mises equivalent stress criterion was applied to transform the complex triaxial stress state into a comprehensive equivalent stress value. Then, the ratio of this equivalent stress to the yield limit extracted from the tool material database was calculated to obtain the tool stress concentration factor under the current state, which is used to indicate whether the tool has experienced significant load concentration in a specific machining stage.

[0037] Based on this, tool degradation analysis is performed according to the tool stress concentration factor. In this process, the entire cutting process is divided into machining time periods of equal length (such as once every 5 seconds or per tooth segment). The tool stress concentration factor of each segment is collected and the values ​​are accumulated to obtain the cumulative tool wear amount of the tool wear change over the entire cycle.

[0038] Next, feature extraction is performed based on the cumulative tool wear to obtain wear fluctuation characteristics. Through joint analysis of the cumulative data, a multidimensional wear feature vector containing multiple state indicators is constructed. From this vector, trend characteristics reflecting long-term wear patterns and fluctuation intensity characteristics reflecting short-term instability are extracted to obtain the wear fluctuation characteristics. Finally, regression analysis is performed on the cumulative tool wear and wear fluctuation characteristics to generate wear compensation coefficients.

[0039] Furthermore, in the method provided in the application embodiments, dynamically comparing the real-time cutting force spectrum with the initial cutting force threshold range to identify and determine abnormal frequency bands, the method further includes: The real-time cutting force spectrum is divided into N sub-bands according to the preset frequency band, where N is an integer greater than 1; the energy density of each of the N sub-bands is compared with the initial cutting force threshold range, and the deviation of the energy value of each sub-band from the upper limit of the initial cutting force threshold range is calculated to obtain multiple sub-band energy deviations; anomaly identification is performed based on the multiple sub-band energy deviations to obtain multiple initial abnormal frequency bands; frequency domain continuity detection is performed based on the multiple initial abnormal frequency bands, and the multiple initial abnormal frequency bands are merged according to the detection results to determine the abnormal frequency band.

[0040] In this embodiment, the real-time cutting force spectrum is first divided into N sub-bands according to a preset frequency band, where N is an integer greater than 1. The division principle is usually based on the frequency response characteristics to divide the frequency band into typical machining sensitive frequency ranges such as low frequency band (0~500Hz), medium frequency band (500~1500Hz) and high frequency band (1500~5000Hz) to ensure that different types of vibration and force changes during the cutting process can be covered.

[0041] Next, the energy density per unit time is calculated for the signal data within each sub-band. Specifically, the original time-domain signal is first processed using a short-time Fourier transform to obtain a cutting force time-frequency characteristic matrix containing the relationship between time, frequency, and amplitude. This matrix describes the variation trend of the cutting force signal at different time windows and frequency points. Within the frequency interval corresponding to each sub-band, the corresponding frequency column is extracted from this matrix, and the amplitude of this frequency column is squared point-by-point across all time windows to obtain a sequence of energy spectral density values ​​at the corresponding frequency. Then, within a set fixed time window (e.g., 0.1 seconds), numerical integration is performed on the energy spectral density values ​​of the sub-band at that frequency to obtain the total energy value of the sub-band in the current time period. Finally, the total energy value is divided by the length of the time window to calculate the energy density per unit time for that sub-band, reflecting the force intensity of that frequency interval under the current machining condition. After obtaining the unit time energy density values ​​of all sub-bands, energy density comparison is performed one by one. That is, the energy density of each sub-band is compared with the energy upper limit of the corresponding frequency band in the initial cutting force threshold range, and the difference between the two is calculated to obtain the energy deviation of the sub-band. In this way, the energy deviation of multiple sub-bands is obtained as the basic indicator for abnormal frequency identification.

[0042] Next, an anomaly detection is performed. Using a fixed over-limit detection method, the energy deviation of each sub-band is compared with the detection threshold (such as 120% of the upper limit of the threshold). If the deviation exceeds the threshold, the sub-band is marked as the initial abnormal frequency band.

[0043] To improve the continuity of frequency analysis, frequency domain continuity detection was performed on all initial anomalous frequency bands using a frequency adjacency merging method. This method compared the center frequency difference, energy variation direction, and bandwidth overlap ratio of adjacent sub-bands. If the center frequency difference was less than 100Hz, the energy deviation direction was consistent, and the bandwidth overlap exceeded 30%, these initial anomalous frequency bands were determined to belong to the same response region and were merged. Finally, based on the above operations, multiple merged anomalous frequency bands were identified.

[0044] Furthermore, in the method provided in the application embodiments, extracting the cutting force fluctuation characteristics based on the abnormal frequency band further includes: The abnormal frequency band is divided into a low-frequency stable component and a high-frequency disturbance component; a Fourier transform is performed on the low-frequency stable component to extract abnormal transform parameters, which include the abnormal fundamental frequency amplitude and the abnormal harmonic energy ratio; a cutting force stability index is calculated based on the abnormal fundamental frequency amplitude and the abnormal harmonic energy ratio, and the cutting force stability index is added to the cutting force fluctuation feature; sudden peak signals are identified in the high-frequency disturbance component, and instantaneous impact loads are calculated based on the sudden peak signals and the tool geometric feature parameters to generate the cutting force fluctuation feature.

[0045] In this embodiment, the identified abnormal frequency bands are decomposed using a characteristic structure method. First, the abnormal frequency bands are divided into low-frequency stable components and high-frequency disturbance components using a frequency threshold division method. The division is typically based on an empirical critical point of 500Hz. Signals below 500Hz are considered low-frequency stable components, reflecting the periodic changes in the cutting load of the tool; signals above 500Hz are defined as high-frequency disturbance components, used to capture non-stationary responses such as micro-wear, vibration excitation, or instantaneous impact.

[0046] For the low-frequency stable components, a Fast Fourier Transform (FFT) is used to reconstruct their frequency domain. The FFT transforms this signal from the time domain to the frequency domain, obtaining a frequency-amplitude spectrum. Two key indicators are extracted from this spectrum to form anomaly transformation parameters: the anomalous fundamental frequency amplitude, i.e., the amplitude intensity of the first dominant frequency point in the spectrum, representing the force response intensity under the main rhythm of the cutting process; and the anomalous harmonic energy ratio, calculated by comparing the sum of the energies of the first k harmonics (usually k=5~7) with the fundamental frequency energy, forming a harmonic interference coefficient reflecting periodic consistency. The higher this ratio, the stronger the higher-order harmonic interference, and the more unstable the cutting process. Subsequently, the cutting force stability index (CFSI) is calculated based on these two parameters, with the following formula: ,in, The fundamental frequency amplitude, Let be the amplitude of the i-th harmonic (i = 1 to k).

[0047] For high-frequency disturbance components, a sliding window peak detection algorithm is used for sudden signal analysis. The sliding window width is set to 10ms and the step size to 2ms. Local maxima are identified within each window, and a threshold (e.g., 2.5 times the mean) is set to filter valid impact signals. These signals typically correspond to impact events caused by tool entry hard points, material inhomogeneity, or localized chipping. After identifying the peak signal, the instantaneous impact load F is estimated using the following model, combined with the tool's geometric parameters (including tip radius, cutting edge width, etc.). Where F is the instantaneous impact load; E is the elastic modulus of the tool material, obtained from a material database; A is the contact area, estimated based on tool geometry (e.g., constructing a semi-ellipsoidal model using fillet radius and cutting edge width); δ is the peak displacement, obtained by measuring the voltage signal using a laser Doppler vibrometer and converting it to micrometer-level displacement based on sensitivity; and L is the impact path length, which can be estimated using tool angle and sound velocity, typically taken as 1~2 mm.

[0048] Ultimately, the instantaneous impact load F, together with the aforementioned cutting force stability index CFSI, constitutes the cutting force fluctuation characteristics under the current abnormal frequency band.

[0049] Furthermore, in the method provided in the application embodiment, feature extraction is performed based on the cumulative tool wear to obtain wear fluctuation features, and regression analysis is performed on the cumulative tool wear and the wear fluctuation features to generate the wear compensation coefficient, which further includes: A joint analysis is performed based on the cumulative tool wear to construct a multidimensional wear feature vector. Wear trend features and fluctuation intensity features are extracted based on the multidimensional wear feature vector. Wear fluctuation analysis is performed according to the wear trend features and the fluctuation intensity features to obtain wear fluctuation features. Kernel principal component analysis is used to reduce the dimensionality of the multidimensional wear feature vector to obtain key wear mode components. Support vector regression is performed by combining the key wear mode components with the wear fluctuation features to obtain the wear compensation coefficient.

[0050] In this embodiment, a joint analysis is first performed based on the cumulative wear of the tool during continuous machining to construct a multidimensional wear feature vector. This feature vector contains multiple feature parameters from different monitoring sources, including wear increment per unit time, average tool force, peak impact load, vibration response amplitude, and cutting temperature change rate. Each feature can be obtained in real time through a sensor system during the machining process, forming a complete wear evolution description sequence.

[0051] Subsequently, based on this multidimensional wear feature vector, linear regression analysis was used to fit the time series data for each dimension, extracting wear trend features representing long-term changes, such as the wear growth slope. Then, combined with sliding window standard deviation analysis, the variance, standard deviation, and coefficient of variation of each feature sequence were calculated within the sliding window to extract fluctuation intensity features representing short-term fluctuation amplitude. These trend and fluctuation features are used to identify the long-term stability and short-term disturbance degree of the tool's operating state.

[0052] After acquiring the two types of features mentioned above, wear fluctuation analysis is performed. By combining and mapping, a composite index that can simultaneously characterize trend stability and local impact is extracted, namely, the wear fluctuation feature. For example, in the stage where high trend slope and high fluctuation coefficient occur simultaneously, it is determined that the tool is experiencing a rapid and unstable wear process, which is then incorporated into subsequent modeling as a key analytical indicator.

[0053] To reduce feature dimensionality and improve regression efficiency, kernel principal component analysis (KPCA) is used to reduce the dimensionality of the original feature vectors. This method uses radial basis functions as kernel functions to map the original nonlinear data to a high-dimensional feature space, and extracts the first few principal components with a cumulative contribution rate higher than 90% as key wear mode components representing the core structure of tool wear changes.

[0054] Finally, the aforementioned key wear mode components and wear fluctuation features are input into a support vector regression model. This model is trained on multiple processed samples with known wear compensation labels. Each training sample contains the sample's key wear mode components, wear fluctuation features, and their corresponding wear compensation coefficients. The wear compensation coefficients serve as the model's output, while the key wear mode components and wear fluctuation features serve as the model's input. During the training phase, the model learns the mapping relationship between the input features and the wear compensation coefficients to establish a prediction model. After training, newly acquired key wear mode components and wear fluctuation features are input into the model, which then predicts and outputs the corresponding wear compensation coefficients in real time.

[0055] Step S400: Correct the cutting force of the irregular tooth workpiece in real time according to the wear compensation coefficient, and generate tool cutting force control suggestions.

[0056] In this embodiment, when real-time correction of the cutting force of an irregularly shaped toothed workpiece according to the wear compensation coefficient, the corresponding spindle speed compensation and feed rate correction values ​​are dynamically calculated based on the wear compensation coefficient obtained from wear analysis. Combined with the actual machining state of the irregularly shaped toothed workpiece, the correction parameters are weighted in stages by extracting the current real-time cutting stage to obtain control parameter correction values ​​that match the current working condition. Subsequently, the multi-axis linkage control path of the CNC system is adjusted in real-time based on this correction value, generating control optimization commands during the machining process. The tool motion trajectory is dynamically updated in a closed-loop manner, and finally, tool cutting force control suggestions are output, achieving precise control of the cutting force and wear adaptability compensation.

[0057] Furthermore, in the method provided in the application embodiments, the real-time correction of the cutting force of the irregular-shaped tooth workpiece according to the wear compensation coefficient to generate tool cutting force control suggestions also includes: The wear compensation coefficient is analyzed to determine the spindle speed compensation and feed rate correction values. The real-time cutting stage of the machining position of the irregular tooth is extracted, and the spindle speed compensation and feed rate correction values ​​are weighted according to the real-time cutting stage to determine the control parameter correction. Multi-axis linkage control is performed on the irregular tooth workpiece based on the control parameter correction to generate control optimization instructions. The control optimization instructions are executed to perform closed-loop control on the machining trajectory of the irregular tooth to generate the tool cutting force control suggestions.

[0058] In this embodiment, the wear compensation coefficient is first analyzed, and a linear proportional coefficient model is used to map it to the spindle speed compensation and feed rate correction values. The spindle speed compensation is determined by adjusting the formula... The feed rate correction value is obtained through Calculations show that Indicates the wear compensation coefficient. , These are the speed and feed adjustment coefficients, respectively, set based on tool material and process experience.

[0059] Secondly, the machining path for irregular-shaped teeth is calibrated in real time to identify the current cutting stage. By matching the timestamp and spatial coordinates of each trajectory point in the CNC code, and combining this with the set geometric feature positions such as tool contact points or tooth boundaries, the entire machining process is divided into three stages: the entry stage, the stable cutting stage on the tooth surface, and the transition stage at the tooth root. Based on the machining stage, the system calls a preset weight set to weight and allocate the aforementioned compensation amounts, ultimately obtaining the control parameter correction amounts for the current machining stage, namely the weighted spindle speed correction value and feed rate adjustment value.

[0060] Next, the current multi-axis machining trajectory is updated using the control parameter correction as input. Specifically, a cubic spline interpolation algorithm is used to insert intermediate control points between the original tool trajectory points to generate a new trajectory with a smooth transition. This avoids trajectory jumps and cutting instability caused by parameter abrupt changes, and outputs a new tool position sequence and spindle / feed control parameters to form a new control optimization instruction.

[0061] Then, control optimization instructions are executed to perform closed-loop control of the machining trajectory of the irregular-shaped teeth. In this process, machining analysis is first performed based on the control optimization instructions to extract the actual machining trajectory of the current irregular-shaped tooth workpiece, and this trajectory is compared with the original planned trajectory to obtain offset information during trajectory execution. Subsequently, a trajectory inverse solution method is used to deduce the existing path offset based on the actual machining path. Combining the local curvature distribution of the irregular-shaped tooth profile, dynamic optimization analysis is performed on the offset region, identifying multiple dynamic optimization nodes with compensation value. These nodes are embedded into the original trajectory, and an optimized machining trajectory is generated through interpolation reconstruction. Finally, the optimized trajectory is uploaded to the CNC controller to construct the closed-loop control logic between the machining path and the feedback response, while simultaneously outputting corresponding tool cutting force control suggestions.

[0062] Furthermore, in the method provided in the application embodiment, executing the control optimization command to perform closed-loop control on the machining trajectory of the irregular tooth and generating the tool cutting force control suggestion further includes: The machining analysis of the irregular-shaped tooth workpiece is performed according to the control optimization instructions to obtain the machining trajectory of the irregular-shaped tooth; the machining trajectory of the irregular-shaped tooth is solved in reverse according to the control optimization instructions to determine the path offset; the path offset is combined with the tooth curvature characteristics of the irregular-shaped tooth workpiece for optimization analysis to obtain multiple dynamic optimization nodes; the multiple dynamic optimization nodes are inserted into the machining trajectory of the irregular-shaped tooth for reconstruction to generate an optimized machining trajectory; the optimized machining trajectory is synchronously updated to the CNC system for closed-loop control to generate the tool cutting force control suggestion.

[0063] In this embodiment, the machining analysis of the irregular-shaped tooth workpiece is first performed according to the control optimization instructions. This process involves collecting the displacement encoding values ​​fed back by the servo system during CNC machining and combining them with theoretical machining trajectory instructions (such as the tool displacement path defined by G-code). A synchronization timestamp algorithm is used to align the actual and theoretical trajectories, forming the true motion trajectory of the tool in three-dimensional space. The trajectory points are referenced to the tool tip, and the sequence of position changes within each machining time segment is recorded to obtain the machining trajectory of the irregular-shaped tooth.

[0064] Subsequently, the machining trajectory of the irregular teeth is solved in reverse based on the control optimization instructions. By using spatial vector distance calculation, the Euclidean distance between each actual machining trajectory point and its corresponding theoretical path point is calculated to obtain the deviation value in the three axes, i.e., the path offset. By traversing each trajectory segment, the path offset curve of the entire machining path is finally output.

[0065] Next, a joint optimization analysis is performed by combining the path offset and the tooth curvature characteristics of the irregular-shaped tooth workpiece. First, based on the 3D tooth profile modeling data of the irregular-shaped tooth workpiece, the local curvature values ​​of each point on the machining path are calculated using the finite difference method to obtain the tooth curvature change curve. Then, the aforementioned path offset curve and curvature change curve are compared and analyzed. By setting offset gradient thresholds and curvature change rate thresholds, regions where both change abruptly are identified as accuracy risk points. Representative trajectory points are selected within these regions, and their spatial positions, corresponding curvature values, and offsets are extracted as the basic data for dynamic optimization nodes, ultimately obtaining multiple dynamic optimization nodes.

[0066] Next, multiple dynamically optimized nodes are inserted into the machining trajectory of the irregular tooth for reconstruction. The trajectory reconstruction uses cubic B-spline interpolation to maintain the smoothness and derivative continuity of the original trajectory. Smooth curve segments are generated by interpolation between the original trajectory segments and the optimized nodes, and the control point sequence of the path is updated. This method can effectively avoid machining vibration and accuracy degradation caused by abrupt trajectory changes, and finally generate an optimized machining trajectory, ensuring that the tool maintains stable stress and high-quality forming during the machining of irregular tooth surfaces.

[0067] Finally, the optimized machining trajectory is synchronously updated to the CNC system for closed-loop control, generating tool cutting force control suggestions. These suggestions adjust strategies based on the offset trend between the optimized and original trajectories and real-time machining parameters (such as spindle load and current fluctuations), including real-time adjustments to key parameters such as spindle speed, feed rate, and acceleration limits. Simultaneously, the optimized machining trajectory is executed as the main path, and predictive control suggestions are provided for subsequent machining of similar tooth segments, achieving dynamic adjustment and intelligent feedback.

[0068] In summary, the embodiments of this application have at least the following technical effects: This application obtains the machining parameters of irregular-shaped tooth workpieces, performs cutting analysis based on tool geometric features and machining parameter sets, and determines the initial cutting force threshold range; it collects dynamic cutting force data of the tool in real time and generates a real-time cutting force spectrum; it dynamically compares the real-time cutting force spectrum with the initial cutting force threshold range, extracts cutting force fluctuation characteristics for tool wear analysis, and obtains a wear compensation coefficient; it then performs real-time correction of the cutting force of the irregular-shaped tooth workpiece according to the wear compensation coefficient, generating tool cutting force control suggestions. This invention solves the technical problem of existing technologies being unable to perceive and effectively control the cutting force during the machining of irregular-shaped teeth in real time, leading to rapid tool wear and low machining accuracy. By combining tool geometric features and machining parameters to model the cutting force, collecting cutting force data in real time and dynamically comparing and analyzing it, calculating the wear compensation coefficient, and implementing closed-loop correction control, it achieves dynamic regulation of tool cutting force, extends tool life, and improves the stability and accuracy of irregular-shaped tooth machining.

[0069] Example 2, based on the same inventive concept as the cutting force control method for machining irregularly shaped teeth in the foregoing examples, such as... Figure 2 As shown, this application provides a tool cutting force control system for machining irregularly shaped teeth. The system and method embodiments in this application are based on the same inventive concept. The system includes: The cutting analysis module 11 is used to acquire the machining parameters of the irregular-shaped tooth workpiece, perform cutting analysis according to the tool geometric feature parameters and the machining parameter set, and determine the initial cutting force threshold range; the real-time data acquisition module 12 is used to acquire the dynamic cutting force data of the tool in real time and generate a real-time cutting force spectrum; the wear analysis module 13 is used to dynamically compare the real-time cutting force spectrum according to the initial cutting force threshold range, extract the cutting force fluctuation characteristics for tool wear analysis, and obtain the wear compensation coefficient; the real-time correction module 14 is used to correct the cutting force of the irregular-shaped tooth workpiece in real time according to the wear compensation coefficient and generate tool cutting force control suggestions.

[0070] Furthermore, the system is also used to implement the following functions: Based on the machining parameters, the tooth profile of the irregular tooth workpiece is analyzed to obtain tooth profile feature parameters; cutting prediction is performed based on the tooth profile feature parameters and the tool geometric feature parameters to calculate theoretical cutting force data; actual cutting force data is dynamically collected through multiple sensors, and the theoretical cutting force data is simulated and corrected based on the actual cutting force data to determine the initial cutting force threshold range.

[0071] Furthermore, the system is also used to implement the following functions: A triaxial strain sensor array is deployed on the cutting tool to sense and collect dynamic cutting force data, which includes multiple axial cutting force components. A vibration sensor is deployed on the irregularly shaped toothed workpiece to sense the machining process and obtain high-frequency vibration signals. The multiple axial cutting force components and the high-frequency vibration signals are fused in the time-frequency domain to construct a cutting force time-frequency feature matrix. Based on the cutting force time-frequency feature matrix, it is decomposed according to a preset frequency band to extract the energy distribution data of the preset frequency band, and the real-time cutting force spectrum is plotted based on the energy distribution data.

[0072] Furthermore, the system is also used to implement the following functions: The real-time cutting force spectrum is dynamically compared with the initial cutting force threshold range to identify abnormal frequency bands. The cutting force fluctuation characteristics are extracted based on the abnormal frequency bands. Tool stress analysis is performed based on the cutting force fluctuation characteristics to obtain the tool stress concentration factor. Tool degradation analysis is performed according to the stress concentration factor to obtain the tool wear accumulation. Feature extraction is performed based on the tool wear accumulation to obtain wear fluctuation characteristics. Regression analysis is performed between the tool wear accumulation and the wear fluctuation characteristics to generate the wear compensation coefficient.

[0073] Furthermore, the system is also used to implement the following functions: The real-time cutting force spectrum is divided into N sub-bands according to the preset frequency band, where N is an integer greater than 1; the energy density of each of the N sub-bands is compared with the initial cutting force threshold range, and the deviation of the energy value of each sub-band from the upper limit of the initial cutting force threshold range is calculated to obtain multiple sub-band energy deviations; anomaly identification is performed based on the multiple sub-band energy deviations to obtain multiple initial abnormal frequency bands; frequency domain continuity detection is performed based on the multiple initial abnormal frequency bands, and the multiple initial abnormal frequency bands are merged according to the detection results to determine the abnormal frequency band.

[0074] Furthermore, the system is also used to implement the following functions: The abnormal frequency band is divided into a low-frequency stable component and a high-frequency disturbance component; a Fourier transform is performed on the low-frequency stable component to extract abnormal transform parameters, which include the abnormal fundamental frequency amplitude and the abnormal harmonic energy ratio; a cutting force stability index is calculated based on the abnormal fundamental frequency amplitude and the abnormal harmonic energy ratio, and the cutting force stability index is added to the cutting force fluctuation feature; sudden peak signals are identified in the high-frequency disturbance component, and instantaneous impact loads are calculated based on the sudden peak signals and the tool geometric feature parameters to generate the cutting force fluctuation feature.

[0075] Furthermore, the system is also used to implement the following functions: A joint analysis is performed based on the cumulative tool wear to construct a multidimensional wear feature vector. Wear trend features and fluctuation intensity features are extracted based on the multidimensional wear feature vector. Wear fluctuation analysis is performed according to the wear trend features and the fluctuation intensity features to obtain wear fluctuation features. Kernel principal component analysis is used to reduce the dimensionality of the multidimensional wear feature vector to obtain key wear mode components. Support vector regression is performed by combining the key wear mode components with the wear fluctuation features to obtain the wear compensation coefficient.

[0076] Furthermore, the system is also used to implement the following functions: The wear compensation coefficient is analyzed to determine the spindle speed compensation and feed rate correction values. The real-time cutting stage of the machining position of the irregular tooth is extracted, and the spindle speed compensation and feed rate correction values ​​are weighted according to the real-time cutting stage to determine the control parameter correction. Multi-axis linkage control is performed on the irregular tooth workpiece based on the control parameter correction to generate control optimization instructions. The control optimization instructions are executed to perform closed-loop control on the machining trajectory of the irregular tooth to generate the tool cutting force control suggestions.

[0077] Furthermore, the system is also used to implement the following functions: The machining analysis of the irregular-shaped tooth workpiece is performed according to the control optimization instructions to obtain the machining trajectory of the irregular-shaped tooth; the machining trajectory of the irregular-shaped tooth is solved in reverse according to the control optimization instructions to determine the path offset; the path offset is combined with the tooth curvature characteristics of the irregular-shaped tooth workpiece for optimization analysis to obtain multiple dynamic optimization nodes; the multiple dynamic optimization nodes are inserted into the machining trajectory of the irregular-shaped tooth for reconstruction to generate an optimized machining trajectory; the optimized machining trajectory is synchronously updated to the CNC system for closed-loop control to generate the tool cutting force control suggestion.

[0078] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0079] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0080] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for controlling the cutting force of a tool used in machining irregularly shaped teeth, characterized in that, The method includes: Obtain the machining parameters of the irregular toothed workpiece, perform cutting analysis based on the tool geometric feature parameters and the machining parameter set, and determine the initial cutting force threshold range; Real-time acquisition of dynamic cutting force data of the tool, generating a real-time cutting force spectrum; The real-time cutting force spectrum is dynamically compared according to the initial cutting force threshold range, and the cutting force fluctuation characteristics are extracted for tool wear analysis to obtain the wear compensation coefficient. Based on the wear compensation coefficient, the cutting force of the irregular tooth workpiece is corrected in real time, and the tool cutting force control suggestion is generated.

2. The cutting force control method for machining irregularly shaped teeth as described in claim 1, characterized in that, To obtain the machining parameters of the irregularly shaped toothed workpiece, and to perform cutting analysis based on the tool geometry parameters and the machining parameter set, the initial cutting force threshold range is determined. The methods include: Based on the aforementioned processing parameters, tooth profile analysis is performed on the irregular tooth workpiece to obtain tooth profile feature parameters; Based on the tooth profile characteristic parameters and the tool geometric characteristic parameters, cutting prediction is performed, and theoretical cutting force data is calculated. The actual cutting force data is dynamically acquired by multiple sensors, and the theoretical cutting force data is simulated and corrected based on the actual cutting force data to determine the initial cutting force threshold range.

3. The cutting force control method for machining irregularly shaped teeth as described in claim 1, characterized in that, Real-time acquisition of dynamic cutting force data of the tool to generate a real-time cutting force spectrum, including the following methods: A triaxial strain sensor array is deployed on the cutting tool, and the cutting tool is sensed and collected through the triaxial strain sensor array to obtain dynamic cutting force data, which includes multiple axial cutting force components. Vibration sensors are installed on irregularly shaped toothed workpieces to obtain high-frequency vibration signals through processing sensing. The multiple axial cutting force components are fused with the high-frequency vibration signal in the time-frequency domain to construct a cutting force time-frequency feature matrix. Based on the cutting force time-frequency feature matrix, the energy distribution data of the preset frequency band is extracted and the real-time cutting force spectrum is plotted according to the energy distribution data.

4. The cutting force control method for machining irregularly shaped teeth as described in claim 3, characterized in that, The method involves dynamically comparing the real-time cutting force spectrum according to the initial cutting force threshold range, extracting cutting force fluctuation characteristics for tool wear analysis, and obtaining the wear compensation coefficient. The real-time cutting force spectrum is dynamically compared with the initial cutting force threshold range to identify abnormal frequency bands, and the cutting force fluctuation characteristics are extracted based on the abnormal frequency bands. Based on the cutting force fluctuation characteristics, tool stress analysis is performed to obtain the tool stress concentration factor; Tool degradation analysis is performed according to the stress concentration factor to obtain the cumulative tool wear. Feature extraction is performed based on the cumulative tool wear to obtain wear fluctuation characteristics. Regression analysis is then performed between the cumulative tool wear and the wear fluctuation characteristics to generate the wear compensation coefficient.

5. The cutting force control method for machining irregularly shaped teeth as described in claim 4, characterized in that, The method involves dynamically comparing the real-time cutting force spectrum with the initial cutting force threshold range to identify and determine abnormal frequency bands. The real-time cutting force spectrum is divided into N sub-bands according to the preset frequency band, where N is an integer greater than 1; The energy density of each of the N sub-bands is compared with that of the initial cutting force threshold range. The deviation of the energy value of each sub-band from the upper limit of the initial cutting force threshold range is calculated to obtain the energy deviation of multiple sub-bands. Anomalies are identified based on the energy deviation of the multiple sub-bands to obtain multiple initial abnormal frequency bands; Frequency domain continuity detection is performed based on the multiple initial abnormal frequency bands, and the multiple initial abnormal frequency bands are merged according to the detection results to determine the abnormal frequency band.

6. The cutting force control method for machining irregularly shaped teeth as described in claim 5, characterized in that, The method for extracting the cutting force fluctuation characteristics based on the abnormal frequency band includes: The abnormal frequency band is divided into a low-frequency stable component and a high-frequency disturbance component; Perform a Fourier transform on the low-frequency stable component to extract the abnormal transform parameters, which include the abnormal fundamental frequency amplitude and the abnormal harmonic energy ratio. The cutting force stability index is calculated based on the ratio of the abnormal fundamental frequency amplitude to the abnormal harmonic energy, and the cutting force stability index is added to the cutting force fluctuation characteristics. Based on the identification of sudden peak signals in the high-frequency disturbance components, the instantaneous impact load is calculated according to the sudden peak signals and the tool geometric feature parameters to generate the cutting force fluctuation characteristics.

7. The cutting force control method for machining irregularly shaped teeth as described in claim 4, characterized in that, Based on the cumulative tool wear, feature extraction is performed to obtain wear fluctuation characteristics. Regression analysis is then performed between the cumulative tool wear and the wear fluctuation characteristics to generate the wear compensation coefficient. The method includes: Based on the cumulative tool wear, a joint analysis is performed to construct a multidimensional wear feature vector, and wear trend features and fluctuation intensity features are extracted based on the multidimensional wear feature vector. Wear fluctuation analysis is performed based on the wear trend characteristics and the fluctuation intensity characteristics to obtain wear fluctuation characteristics; Kernel principal component analysis was used to reduce the dimensionality of the multidimensional wear feature vector to obtain the key wear mode components. The wear compensation coefficient is obtained by combining the key wear mode components with the wear fluctuation characteristics through support vector regression.

8. The cutting force control method for machining irregularly shaped teeth as described in claim 1, characterized in that, The method includes real-time correction of the cutting force of the irregular-shaped tooth workpiece according to the wear compensation coefficient, and generation of tool cutting force control suggestions. Analyze the wear compensation coefficient to determine the spindle speed compensation and feed rate correction values; Extract the real-time cutting stage of the machining position of the irregular tooth, and allocate the spindle speed compensation and feed correction value according to the real-time cutting stage to determine the control parameter correction amount; Based on the control parameter correction, multi-axis linkage control is performed on the irregular tooth workpiece to generate control optimization instructions; The control optimization instructions are executed to perform closed-loop control on the machining trajectory of the irregular teeth, and the tool cutting force control suggestions are generated.

9. The cutting force control method for machining irregularly shaped teeth as described in claim 8, characterized in that, The method includes executing the control optimization instructions to perform closed-loop control on the machining trajectory of irregular teeth and generating the tool cutting force control suggestions, wherein the control optimization instructions are executed. The machining trajectory of the irregular tooth workpiece is obtained by performing machining analysis on the irregular tooth workpiece according to the control optimization instructions. Based on the control optimization instructions, the machining trajectory of the irregular tooth is solved in reverse to determine the path offset; Based on the path offset and the tooth curvature characteristics of the irregular tooth workpiece, an optimization analysis is performed to obtain multiple dynamic optimization nodes; The multiple dynamic optimization nodes are inserted into the machining trajectory of the irregular tooth to reconstruct the machining trajectory, generating an optimized machining trajectory. The optimized machining trajectory is then synchronously updated to the CNC system for closed-loop control, generating the tool cutting force control suggestion.

10. A tool cutting force control system for machining irregularly shaped teeth, characterized in that, The system includes: The cutting analysis module is used to obtain the machining parameters of irregularly shaped toothed workpieces, perform cutting analysis based on the tool geometric feature parameters and the machining parameter set, and determine the initial cutting force threshold range. The real-time data acquisition module is used to acquire dynamic cutting force data of the tool in real time and generate a real-time cutting force spectrum; The wear analysis module is used to dynamically compare the real-time cutting force spectrum according to the initial cutting force threshold range, extract the cutting force fluctuation characteristics to perform tool wear analysis, and obtain the wear compensation coefficient. The real-time correction module is used to correct the cutting force of the irregular toothed workpiece in real time according to the wear compensation coefficient and generate tool cutting force control suggestions.

Images