Aluminum profile high-precision machining optimization method and system

Abstract

The present application relates to the field of high-end equipment manufacturing technology, in particular to a high-precision machining optimization method and system for aluminum profiles. In the present application, the cutting force signal, the machine tool shell vibration signal and the tool holder component acoustic emission signal of the aluminum profile workpiece are acquired, and the sampling points are time-stamped and interpolated to form a multi-source machining signal perception dataset, so that the machining state can be represented by multiple machining signals. Through fast Fourier transform processing, harmonic frequency band energy extraction, instantaneous impact force identification and instantaneous acoustic emission energy extraction, the frequency domain features, impact features and acoustic emission features are integrated into a machining feature dataset, and the aluminum profile material state is further matched. On this basis, the comprehensive coordinate deviation data is formed according to the tool path elastic deviation coordinates and the cutting contact area temperature expansion compensation coordinates, which plays a role in jointly correcting the tool feed speed and moving path in the current machining cycle.

Description

Technical Field

[0001] This invention relates to the field of high-end equipment manufacturing technology, and in particular to a method and system for optimizing high-precision machining of aluminum profiles. Background Technology

[0002] The high-precision machining optimization method for aluminum profiles refers to the specific planning process for adjusting process parameters and improving machining paths in the cutting and forming processes of aluminum alloy profiles. This method is mainly used to control variables such as thermal deformation and cutting vibration during machining to meet dimensional tolerance requirements.

[0003] Traditional high-precision machining optimization methods for aluminum profiles, while capable of controlling thermal deformation and cutting vibration by adjusting process parameters and improving machining paths, suffer from several challenges. Firstly, when changes in cutting force, machine tool housing vibration, and acoustic emission from the tool clamping components occur simultaneously during aluminum profile machining, a single machining signal may not accurately reflect the current machining state. Secondly, when the tool is subjected to instantaneous impact, the aluminum profile material condition varies, or the cutting contact area experiences temperature rise, it becomes difficult to simultaneously determine the elastic deviation of the tool trajectory and the compensation for temperature expansion in the cutting contact area. Thirdly, if subsequent feed commands continue to execute according to theoretical coordinates and the original feed rate, deviations may arise between the actual and theoretical tool movement trajectories, and there may be a lack of correspondence between feed rate correction and spatial trajectory correction. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an optimized method and system for high-precision machining of aluminum profiles.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a high-precision processing optimization method for aluminum profiles, comprising the following steps: Acquire a multi-source machining signal sensing dataset based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component during the machining process of aluminum profile workpieces; By analyzing the features corresponding to each signal in the multi-source processing signal sensing dataset, a processing feature dataset is obtained; Based on the machining feature dataset, determine the comprehensive coordinate deviation data of the tool within the current machining cycle; The next feed command to be processed is obtained to drive the tool. The tool displacement correction interpolation path and cutting depth correction coordinates are determined by combining the comprehensive coordinate deviation data. The coordinates and paths are mapped and combined in the preset three-dimensional space as the adjusted tool space trajectory data. The current feed rate parameters of the tool are obtained, and the adjusted tool feed rate is calculated by combining the machining feature dataset. Based on the adjusted tool feed rate and the adjusted tool spatial trajectory data, the tool feed rate and movement trajectory in the current machining cycle are corrected to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, and the optimization results of aluminum profile machining are obtained.

[0006] As a further aspect of the present invention, the multi-source machining signal sensing dataset obtained during the machining of aluminum profile workpieces, based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component, includes: During the processing of aluminum profile workpieces, the cutting force signal of the aluminum profile workpiece is obtained by a piezoelectric triaxial force sensor, the vibration signal of the machine tool housing is obtained by an accelerometer, and the acoustic emission signal of the tool clamping component is obtained by an acoustic emission sensor. Initial timestamps are assigned to the sampling points in the sampling point sequences of the aluminum profile workpiece cutting force signal, the machine tool housing vibration signal, and the acoustic emission signal of the tool clamping component. The initial timestamps of the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component are compared to identify the misaligned time nodes between the signals, and the amplitude of the missing sampling points between the misaligned time nodes is calculated. The amplitude of the missing sampling points is added to the sampling point sequence of each signal to generate the interpolated aluminum profile workpiece cutting force signal, the interpolated machine tool housing vibration signal, and the interpolated tool clamping component acoustic emission signal, which are combined into a multi-source machining signal sensing dataset.

[0007] As a further aspect of the present invention, the step of analyzing the features corresponding to each signal in the multi-source processing signal sensing dataset to obtain the processing feature dataset includes: Fast Fourier Transform (FFT) processing is performed on the aluminum profile workpiece cutting force signal and the machine tool housing vibration signal after centralized interpolation and compensation of the multi-source processing signal sensing data to generate cutting force frequency domain representation data and vibration frequency domain representation data, respectively. The cutting force frequency point sequence and the corresponding cutting force spectrum amplitude sequence are determined from the cutting force frequency domain representation data. The vibration frequency point sequence and the corresponding vibration spectrum amplitude sequence are determined from the vibration frequency domain representation data. Corresponding frequency points are selected from the cutting force frequency point sequence and the vibration frequency point sequence according to the preset harmonic frequency band range. The energy value of the specified harmonic frequency band is determined according to the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence corresponding to the corresponding frequency point. The cumulative amplitude result of the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence within the preset total frequency band range is used as the total frequency band energy value. The spectrum amplitude corresponding to the spectrum peak position in the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence is used as the reference spectrum peak amplitude. Based on the initial timestamp of the sampling point and the corresponding cutting force amplitude in the cutting force signal of the aluminum profile workpiece after interpolation compensation, the instantaneous rise rate of the amplitude is calculated. At the same time, the target sampling point in the current processing cycle where the change in cutting force amplitude exceeds the preset impact force judgment threshold is determined. The cutting force amplitude corresponding to the target sampling point is determined as the instantaneous impact force value. Based on the instantaneous impact force value, the envelope demodulation processing of the acoustic emission signal of the tool clamping component after interpolation compensation is performed to extract the instantaneous acoustic emission energy value. The specified harmonic frequency band energy value, the total frequency band energy value, and the reference spectrum peak amplitude are used as common features of the aluminum profile workpiece cutting force signal and the machine tool housing vibration signal. The instantaneous rise rate of the amplitude and the instantaneous impact force value determined based on the aluminum profile workpiece cutting force signal are combined with the corresponding instantaneous acoustic emission energy value as features of the acoustic emission signal of the tool clamping component, and these are combined into a feature value sequence. The cosine similarity between the entire feature value sequence and each reference mode vector in the preset aluminum profile material state mode library is calculated. The aluminum profile material state associated with the reference mode vector with the maximum cosine similarity in the state mode library is selected. The feature value sequence and the corresponding aluminum profile material state are integrated to obtain the processing feature dataset.

[0008] As a further aspect of the present invention, determining the comprehensive coordinate deviation data of the tool within the current machining cycle based on the machining feature dataset includes: The tool overhang length corresponding to the aluminum profile material state in the processing feature dataset is matched from the preset process parameter library, and the preset tool material elastic modulus and tool section moment of inertia value are obtained. Combined with the instantaneous impact force value, the tool trajectory elastic deviation is calculated. At the same time, the three-dimensional cutting force components in the current processing cycle are extracted from the interpolated aluminum profile workpiece cutting force signal and combined into a cutting force direction vector. The cutting force direction vector is normalized. Based on the tool trajectory elastic deviation and the normalized cutting force direction vector, the tool trajectory elastic deviation coordinates are determined. Based on the processing feature dataset, the ratio of the energy value of the specified harmonic frequency band to the energy value of the total frequency band is calculated to obtain the energy distribution concentration value. The product of the energy distribution concentration value and the peak amplitude of the reference spectrum is calculated. Based on the preset temperature rise conversion coefficient, the product is converted into the temperature rise prediction value of the cutting contact area. Combined with the length of the heated cutting section of the aluminum profile workpiece, the temperature expansion compensation amount of the cutting contact area is calculated. The length of the heated cutting section of the aluminum profile workpiece is the distance between the starting coordinates and the ending coordinates of the cutting path of the aluminum profile workpiece collected in the current processing cycle. The temperature expansion compensation amount of the cutting contact area is mapped along the direction of the preset machining coordinate system to obtain the temperature expansion compensation coordinate of the cutting contact area. The temperature expansion compensation coordinate of the cutting contact area is integrated with the elastic deviation coordinate of the tool trajectory to obtain comprehensive coordinate deviation data.

[0009] As a further aspect of the present invention, the step of obtaining the next feed command to be processed for driving the tool, determining the tool displacement correction interpolation path and cutting depth correction coordinates by combining comprehensive coordinate deviation data, and mapping and combining the coordinates and paths in a preset three-dimensional space, as the adjusted tool space trajectory data, includes: Obtain the next feed command to be processed to drive the tool, extract the theoretical coordinate data and cutting depth parameters from the feed command, and unify the theoretical coordinate data, tool trajectory elastic deviation coordinates and cutting contact area temperature expansion compensation coordinates into the preset machining coordinate system. The difference between the elastic deviation coordinates of the tool trajectory in the theoretical coordinate data and the comprehensive coordinate deviation data is calculated to obtain the tool displacement correction interpolation path. The difference between the cutting depth parameter and the component of the temperature expansion compensation coordinate of the cutting contact area in the cutting depth direction in the comprehensive coordinate deviation data is calculated to obtain the cutting depth correction coordinate. The tool displacement correction interpolation path is used as the feed direction movement trajectory corresponding to the theoretical coordinate data in the preset machining coordinate system, and the cutting depth correction coordinate is used as the depth coordinate in the cutting depth direction in the preset machining coordinate system. The tool displacement correction interpolation path and the cutting depth correction coordinate are mapped and combined in the preset three-dimensional space within the preset machining coordinate system to generate adjusted tool space trajectory data.

[0010] As a further aspect of the present invention, the step of obtaining the current feed rate parameter of the tool, calculating the adjusted tool feed rate in conjunction with the machining feature dataset, and correcting the tool feed rate and movement trajectory within the current machining cycle based on the adjusted tool feed rate and the adjusted tool spatial trajectory data, in order to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of the aluminum profile workpiece, and obtaining the aluminum profile machining optimization results include: Obtain the current feed rate parameters of the tool and the instantaneous impact force value in the machining feature dataset, and calculate the adjusted tool feed rate; Based on the preset upper limit and lower limit of feed rate, the adjusted tool feed rate is subjected to speed limiting processing, and based on the preset acceleration constraint, the adjusted tool feed rate after speed limiting processing is subjected to acceleration constraint processing to obtain the constrained adjusted tool feed rate. Based on the constrained adjusted tool feed rate and the adjusted tool spatial trajectory data, the tool feed rate and movement trajectory within the current machining cycle are adjusted to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, thereby obtaining optimized aluminum profile machining results.

[0011] A high-precision machining optimization system for aluminum profiles, the system comprising: The signal synchronization module acquires a multi-source machining signal sensing dataset based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component during the machining process of the aluminum profile workpiece. The feature recognition module analyzes the features corresponding to each signal in the multi-source processing signal sensing dataset to obtain the processing feature dataset. The deviation calculation module determines the comprehensive coordinate deviation data of the tool within the current machining cycle based on the machining feature dataset. The trajectory correction module obtains the next feed command to be processed for driving the tool, determines the tool displacement correction interpolation path and cutting depth correction coordinates by combining comprehensive coordinate deviation data, and maps and combines the coordinates and paths in a preset three-dimensional space as the adjusted tool space trajectory data. The drive control module acquires the current feed rate parameters of the tool, calculates the adjusted tool feed rate based on the machining feature dataset, and corrects the tool feed rate and movement trajectory within the current machining cycle based on the adjusted tool feed rate and adjusted tool spatial trajectory data. This reduces the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, resulting in optimized aluminum profile machining.

[0012] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by acquiring cutting force signals of aluminum profile workpieces, vibration signals of machine tool housings, and acoustic emission signals of tool clamping components, and performing timestamp alignment and interpolation compensation on sampling points, a multi-source machining signal perception dataset is formed, enabling the machining state to be jointly characterized by multiple machining signals. Through fast Fourier transform processing, harmonic frequency band energy extraction, instantaneous impact force identification, and instantaneous acoustic emission energy extraction, frequency domain features, impact features, and acoustic emission features are integrated into a machining feature dataset, which is further matched with the aluminum profile material state. Based on this, comprehensive coordinate deviation data is formed according to the elastic deviation coordinates of the tool trajectory and the temperature expansion compensation coordinates of the cutting contact area. Combined with the next feed command to be processed, adjusted tool spatial trajectory data is generated. At the same time, the adjusted tool feed rate is calculated by combining the instantaneous impact force value and subjected to amplitude limiting and acceleration constraints, which plays a role in jointly correcting the tool feed rate and movement trajectory within the current machining cycle, thereby reducing the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces. Attached Figure Description

[0013] Figure 1 This is a flowchart of the steps of the method of the present invention; Figure 2 This is a block diagram of the system of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0015] Please see Figure 1 This invention provides a technical solution, an optimized method for high-precision processing of aluminum profiles, comprising the following steps: S1: The multi-source machining signal sensing dataset acquired during the machining of aluminum profile workpieces includes cutting force signals from the aluminum profile workpiece, vibration signals from the machine tool housing, and acoustic emission signals from the tool clamping components. During the processing of aluminum profile workpieces, the cutting force signal of the aluminum profile workpiece is obtained by a piezoelectric triaxial force sensor, the vibration signal of the machine tool housing is obtained by an accelerometer, and the acoustic emission signal of the tool clamping component is obtained by an acoustic emission sensor. Initial timestamps are assigned to the sampling points in the sampling point sequences of the aluminum profile workpiece cutting force signal, the machine tool housing vibration signal, and the acoustic emission signal of the tool clamping component. First, on-site sampling preparation is carried out. The aluminum profile workpiece refers to the actual aluminum alloy profile being machined. The cutting force signal of the aluminum profile workpiece refers to the force value record formed in the feed direction, transverse direction, and depth of cut direction when the tool contacts the aluminum profile workpiece. The machine tool housing vibration signal refers to the acceleration amplitude record formed after the machining load is transmitted to the machine tool housing. The acoustic emission signal of the tool clamping component refers to the high-frequency amplitude record formed at the tool clamping position due to force contact, impact, and slight elastic deformation. The piezoelectric triaxial force sensor is installed at the aluminum profile workpiece clamping support position. The three measurement directions of the piezoelectric triaxial force sensor correspond to the feed direction, transverse direction, and depth of cut direction in the preset machining coordinate system, respectively. The acceleration sensor is fixed on the rigid surface of the machine tool housing near the spindle box or machining table. The acoustic emission sensor is fixed on the outer surface of the tool clamping component. The acquisition channels of the piezoelectric triaxial force sensor, accelerometer, and acoustic emission sensor are connected to an industrial control computer (ICSC). The ICSC is equipped with MATLAB-R2023b. The output files of the piezoelectric triaxial force sensor, accelerometer, and acoustic emission sensor are arranged according to sampling point number, sampling time, and corresponding channel amplitude. At the start of acquisition, using the processing cycle start time as the same time reference, the ICSC reads the first sampling time of each acquisition channel of the piezoelectric triaxial force sensor, accelerometer, and acoustic emission sensor, and converts the sampling order and sampling interval of each subsequent sampling point into an initial timestamp. The initial timestamp refers to the time position of the sampling point relative to the processing cycle start time. After importing into MATLAB, blank amplitude fields are marked as positions to be filled. Duplicate sampling points are merged according to the same sampling time, retaining the one with the earliest sampling time and amplitude within the sensor range. When the amplitude exceeds the sensor range, the effective amplitudes adjacent to the sampling point exceeding the sensor range are read, and the sampling point exceeding the sensor range is replaced with the intermediate amplitude between the two effective amplitudes. Finally, the cutting force signal of the aluminum profile workpiece with the initial timestamp, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component were obtained.

[0016] By comparing the initial timestamps of the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component, the misaligned time nodes between the signals are identified, and the amplitude of the missing sampling points between the misaligned time nodes is calculated. A time reference comparison was performed on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component. A misaligned time node refers to a time position where the same initial timestamp exists in one of these signals, but is missing in another. MATLAB first reads the initial timestamp columns of the cutting force signal, machine tool housing vibration signal, and acoustic emission signal of the tool clamping component, and arranges them in ascending order. Then, all initial timestamps that have appeared in these signals are merged into a timestamp lookup table, with each row corresponding to one initial timestamp and each column corresponding to one signal. If the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component all have amplitude in a certain row, then no points need to be added to that row; if the cutting force signal of the aluminum profile workpiece has amplitude but the vibration signal of the machine tool housing is empty, then the machine tool housing vibration signal in the current row is marked as needing to be added; if the cutting force signal of the aluminum profile workpiece has amplitude but the acoustic emission signal of the tool clamping component is empty, then the acoustic emission signal of the tool clamping component in the current row is marked as needing to be added; if the machine tool housing vibration signal has amplitude but the cutting force signal of the aluminum profile workpiece... If the signal is empty, the cutting force signal of the aluminum profile workpiece in the current row is marked as to be supplemented; if the machine tool housing vibration signal has amplitude but the acoustic emission signal of the tool clamping component is empty, the acoustic emission signal of the tool clamping component in the current row is marked as to be supplemented; if the acoustic emission signal of the tool clamping component has amplitude but the cutting force signal of the aluminum profile workpiece is empty, the cutting force signal of the aluminum profile workpiece in the current row is marked as to be supplemented; if the acoustic emission signal of the tool clamping component has amplitude but the vibration signal of the machine tool housing is empty, the vibration signal of the machine tool housing in the current row is marked as to be supplemented. For each position to be supplemented, MATLAB reads the nearest valid sampling point before and after the position in the same signal column, and determines the corresponding amplitude between the previous and subsequent valid amplitudes according to the ratio between the initial timestamp of the position to be supplemented and the previous valid timestamp. Finally, the amplitude of the missing sampling point for each position to be supplemented is obtained.

[0017] The amplitude of the missing sampling points is added to the sampling point sequence of each signal to generate the interpolated aluminum profile workpiece cutting force signal, the interpolated machine tool housing vibration signal, and the interpolated tool clamping component acoustic emission signal, which are combined into a multi-source machining signal sensing dataset. The missing sampling point amplitudes are backfilled into the corresponding sampling point sequences of the aluminum profile workpiece cutting force signal, machine tool housing vibration signal, and tool clamping component acoustic emission signal. The interpolated and compensated aluminum profile workpiece cutting force signal, interpolated and compensated machine tool housing vibration signal, and interpolated and compensated tool clamping component acoustic emission signal refer to the continuous time sequences of the original signals after the corresponding amplitudes are filled into the missing time positions. MATLAB reads the list of positions to be filled and creates new sampling point records in the corresponding signal columns; the positions to be filled for the aluminum profile workpiece cutting force signal are written with the triaxial force values ​​under the same initial timestamp, the positions to be filled for the machine tool housing vibration signal are written with the vibration amplitude under the same initial timestamp, and the positions to be filled for the tool clamping component acoustic emission signal are written with the acoustic emission amplitude under the same initial timestamp. After insertion, the data is reordered in ascending order of the initial timestamp, and the time interval between two adjacent sampling points is read. If the time interval between two adjacent sampling points matches the set sampling interval, the arrangement of the two adjacent sampling points is retained. If the time interval between two adjacent sampling points is greater than the set sampling interval, there are still missing time positions between the two adjacent sampling points, and the amplitude is determined and inserted according to the effective sampling points before and after. If the time interval between two adjacent sampling points is less than the set sampling interval, there are duplicate or abnormally dense samplings between the two adjacent sampling points, and the one with the earlier sampling time and the amplitude within the range is retained. The set sampling interval is obtained by converting the sensor sampling frequency. When the sampling frequencies of the piezoelectric triaxial force sensor, accelerometer, and acoustic emission sensor are inconsistent, the set sampling interval is converted using the unified output sampling frequency set before processing. Finally, a multi-source processed signal sensing dataset is obtained.

[0018] S2: Analyze the features corresponding to each signal in the multi-source processing signal sensing dataset to obtain the processing feature dataset, which includes: Fast Fourier Transform (FFT) processing is performed on the aluminum profile workpiece cutting force signal and the machine tool housing vibration signal after centralized interpolation and compensation of the multi-source machining signal sensing data to generate cutting force frequency domain representation data and vibration frequency domain representation data, respectively. Frequency component processing is performed on the multi-source machining signal sensing dataset. Frequency domain representation converts time-series amplitude values ​​into corresponding records of frequency points and spectral amplitudes. Fast Fourier Transform (FFT) processing decomposes the time-domain amplitudes into numerical values ​​at different frequency points. MATLAB reads the interpolated aluminum profile workpiece cutting force signal amplitude sequence and the interpolated machine tool housing vibration signal amplitude sequence. Blank fields are first removed, and repeated timestamps retain the previously retained one. Sampling points exceeding the sensor's range use the previously completed replacement records. To reduce the influence of static bias, the average amplitude of the aluminum profile workpiece cutting force signal amplitude sequence and the machine tool housing vibration signal amplitude sequence within the same machining cycle is first calculated. Then, the current sampling point amplitude of each sampling point is subtracted from the corresponding average amplitude to obtain the amplitude to be converted. Subsequently, FFT processing is performed in MATLAB, with the input being the amplitude sequence to be converted and the sampling frequency. MATLAB generates discrete frequency points according to the number of sampling points and synthesizes the sine and cosine components corresponding to each frequency point into a spectral amplitude. The frequency interval is determined by both the sampling frequency and the number of sampling points. A higher sampling frequency allows for a wider frequency range to be represented, while a larger number of sampling points results in finer intervals between adjacent frequency points. The spectral amplitude is determined by the sine and cosine components corresponding to the same frequency point. Finally, the cutting force frequency domain representation data and vibration frequency domain representation data are obtained.

[0019] The cutting force frequency point sequence and the corresponding cutting force spectrum amplitude sequence are determined from the cutting force frequency domain representation data. The vibration frequency point sequence and the corresponding vibration spectrum amplitude sequence are determined from the vibration frequency domain representation data. According to the preset harmonic frequency band range, corresponding frequency points are selected in the cutting force frequency point sequence and the vibration frequency point sequence respectively. The energy value of the specified harmonic frequency band is determined according to the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence corresponding to the corresponding frequency point. The cumulative amplitude result of the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence within the preset total frequency band range is used as the total frequency band energy value. The spectrum amplitude corresponding to the spectrum peak position in the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence is used as the reference spectrum peak amplitude. Frequency band and peak value correlations are extracted from the cutting force frequency domain representation data and vibration frequency domain representation data. The cutting force frequency point sequence refers to the discrete frequency records arranged from low to high frequency in the cutting force frequency domain representation data; the cutting force spectrum amplitude sequence refers to the amplitude records corresponding one-to-one with the cutting force frequency point sequence; the vibration frequency point sequence refers to the discrete frequency records arranged from low to high frequency in the vibration frequency domain representation data; the vibration spectrum amplitude sequence refers to the amplitude records corresponding one-to-one with the vibration frequency point sequence; the preset harmonic frequency band range refers to the frequency range set around the periodic cutting frequency of the tool; and the preset total frequency band range refers to the frequency range participating in the accumulation under effective sensor sampling conditions. The preset harmonic frequency band range is jointly set before machining by the tool speed, the number of tool teeth, and the sampling frequency. The higher the tool speed, the higher the center frequency corresponding to the preset harmonic frequency band range; the more tool teeth, the higher the center frequency corresponding to the preset harmonic frequency band range; the sampling frequency limits the highest frequency that can be read. The preset total frequency band range is determined by the effective sampling frequency of the sensor, with an upper limit not exceeding the highest effective frequency that the sampling frequency can represent, and a lower limit starting from the first effective frequency point after the 0 frequency point. MATLAB first arranges the cutting force frequency point sequence and vibration frequency point sequence in ascending order of numerical values, and binds each frequency point to its corresponding spectral amplitude. If the frequency point is less than the lower limit of the preset harmonic frequency band range, the corresponding amplitude is not read; if the frequency point is equal to the lower limit of the preset harmonic frequency band range, the corresponding amplitude is read; if the frequency point is greater than the lower limit of the preset harmonic frequency band range but less than the upper limit of the preset harmonic frequency band range, the corresponding amplitude is read; if the frequency point is equal to the upper limit of the preset harmonic frequency band range, the corresponding amplitude is read; if the frequency point is greater than the upper limit of the preset harmonic frequency band range, the corresponding amplitude is not read. Then, the read cutting force spectral amplitude and vibration spectral amplitude are accumulated to obtain the energy value of the specified harmonic frequency band; finally, the cutting force spectral amplitude and vibration spectral amplitude within the preset total frequency band range are accumulated to obtain the total frequency band energy value. Subsequently, the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence are compared item by item. If the current spectrum amplitude is greater than the recorded maximum spectrum amplitude, the recorded maximum spectrum amplitude is replaced with the current spectrum amplitude; if the current spectrum amplitude is equal to the recorded maximum spectrum amplitude, the item with the earlier frequency point is retained; if the current spectrum amplitude is less than the recorded maximum spectrum amplitude, no replacement is made. Finally, the specified harmonic frequency band energy value, the total frequency band energy value, and the reference spectrum peak amplitude are obtained.

[0020] Based on the initial timestamp of the sampling point and the corresponding cutting force amplitude in the cutting force signal of the interpolated aluminum profile workpiece, the instantaneous rise rate of the amplitude is calculated. At the same time, the target sampling point in the current processing cycle where the change in cutting force amplitude exceeds the preset impact force judgment threshold is determined. The cutting force amplitude corresponding to the target sampling point is determined as the instantaneous impact force value. Based on the instantaneous impact force value, the envelope demodulation processing of the acoustic emission signal of the interpolated tool clamping component is performed to extract the instantaneous acoustic emission energy value. Impact and acoustic emission correlation values ​​are extracted from the interpolated aluminum profile cutting force signal and the interpolated tool clamping component acoustic emission signal. The instantaneous rise rate of amplitude refers to the rate of change of amplitude between adjacent cutting force sampling points relative to the time interval. The target sampling point refers to the sampling point where the change in cutting force amplitude is greater than the preset impact force judgment threshold. The envelope refers to the change line formed by the outer edge of the acoustic emission amplitude. The preset impact force judgment threshold is determined before processing by the allowable fluctuation ratio and allowable impact increment corresponding to the theoretical normal cutting force data. The theoretical normal cutting force data comes from the records in the preset process parameter library corresponding to the aluminum profile material state, tool overhang length, current feed rate parameter, and cutting depth parameter. The allowable fluctuation ratio is determined by the cutting force fluctuation record allowed in the current processing cycle. The allowable impact increment is set according to the tool overhang length and material hardness record. The longer the tool overhang length, the smaller the allowable impact increment. The larger the material hardness record, the smaller the allowable impact increment. The preset impact force judgment threshold is determined by the fluctuation limit corresponding to the theoretical normal cutting force data and the allowable impact increment. MATLAB reads two adjacent cutting force sampling points based on their initial timestamps. It calculates the instantaneous rate of increase of the amplitude based on the change in the cutting force amplitude of the later sampling point relative to the previous sampling point, and the time interval between the initial timestamps of the two sampling points. If the change in the cutting force amplitude within the current machining cycle is greater than a preset impact force threshold, the current sampling point is designated as the current target sampling point. If the change in the cutting force amplitude within the current machining cycle is equal to the preset impact force threshold, the current sampling point is not included in the target sampling point set. If the change in the cutting force amplitude within the current machining cycle is less than the preset impact force threshold, the current sampling point is not included in the target sampling point set. For each target sampling point, the corresponding cutting force amplitude is read, along with the acoustic emission amplitude within a preset time window before and after the same initial timestamp. The preset time window is set by the acoustic emission sensor sampling frequency and the cutting force sampling point interval. A higher acoustic emission sensor sampling frequency allows for more acoustic emission sampling points to be read within the preset time window, and a larger cutting force sampling point interval results in a wider preset time window. MATLAB takes the absolute value of the intercepted acoustic emission amplitude, then reads adjacent local peak points and connects them to form an envelope. The envelope amplitudes within the intercepted time range are accumulated to obtain the instantaneous acoustic emission energy value. Finally, the instantaneous rise rate of amplitude, instantaneous impact force, and instantaneous acoustic emission energy value are obtained.

[0021] The specified harmonic frequency band energy value, total frequency band energy value, and reference spectrum peak amplitude are used as common features of the cutting force signal of the aluminum profile workpiece and the vibration signal of the machine tool housing. The instantaneous rise rate of amplitude and instantaneous impact force value determined based on the cutting force signal of the aluminum profile workpiece are used as features of the acoustic emission signal of the tool clamping component, along with the corresponding instantaneous acoustic emission energy value, and these are combined into a feature value sequence. The cosine similarity between the entire feature value sequence and each reference mode vector in the preset aluminum profile material state mode library is calculated. The aluminum profile material state associated with the reference mode vector with the maximum cosine similarity is selected in the state mode library. The feature value sequence and the corresponding aluminum profile material state are integrated to obtain the processing feature dataset. The reference mode vector includes the specified harmonic frequency band energy value, total frequency band energy value, reference spectrum peak amplitude, instantaneous rise rate of amplitude, instantaneous impact force value, and the corresponding reference value of the instantaneous acoustic emission energy value in the preset aluminum profile material state mode library. Frequency domain features, impact features, and acoustic emission features are organized into material state matching data. Feature value sequences refer to processing feature values ​​arranged in a fixed order. The preset aluminum profile material state pattern library refers to the material state baseline record established before processing. The reference pattern vector refers to a group of values ​​formed by arranging baseline values ​​in the same order for the same aluminum profile material state. Cosine similarity refers to the comparison value of the degree of closeness in the directions of two sets of values. The aluminum profile material state can be set in the preset aluminum profile material state pattern library as, for example, annealed state, extruded state, naturally aged state, artificially aged state, cold-worked state, and heat-affected softening state. The specific names are based on the aluminum profile material grade and the state field in the processing record. The process of setting up the preset aluminum profile material state mode library is as follows: First, select a sample with the same material grade as the current aluminum profile workpiece, and collect sample data under the same sensor arrangement, the same sampling frequency, the same tool overhang length range, and the same cutting depth parameter range; then, according to the aforementioned sampling, timestamp alignment, missing sampling point compensation, frequency domain conversion, impact extraction, and acoustic emission energy extraction process, obtain the specified harmonic frequency band energy value, total frequency band energy value, reference spectrum peak amplitude, instantaneous amplitude rise rate, instantaneous impact force value, and instantaneous acoustic emission energy value corresponding to each aluminum profile material state; then, arrange the same type of values ​​under the same aluminum profile material state from smallest to largest, and take the middle value as the corresponding reference value. If the number of sample records is even, take the average of the two middle items. In the current processing cycle, MATLAB reads the specified harmonic frequency band energy value, total frequency band energy value, reference spectrum peak amplitude, instantaneous amplitude rise rate, instantaneous impact force value, and corresponding instantaneous acoustic emission energy value in a fixed order, and performs scaling conversion using similar reference values. The closer the current feature value is to the similar reference value, the closer the corresponding feature term after conversion is to the reference mode vector; when the similar reference value is 0, the feature value after conversion is recorded as 0. When calculating cosine similarity, the values ​​at the same positions in the feature value sequence and the reference mode vector are read item by item to obtain the cumulative result of the product at the corresponding positions; then, the square root result of the cumulative square of the feature value sequence itself and the square root result of the cumulative square of the reference mode vector itself are obtained respectively; finally, the ratio of the cumulative result of the product at the corresponding positions to the product formed by the two square roots is compared. If the cosine similarity of a certain reference pattern vector is greater than that of other reference pattern vectors, the aluminum profile material state associated with the reference pattern vector is read. If the cosine similarity of two or more reference pattern vectors is the same and all are the maximum values, the absolute deviation between the current instantaneous impact force value and the instantaneous impact force benchmark value in each reference pattern vector is compared, and the aluminum profile material state corresponding to the one with the smallest deviation is taken as the current aluminum profile material state. Finally, the processing feature dataset is obtained.

[0022] S3: Based on the machining feature dataset, determine the comprehensive coordinate deviation data of the tool within the current machining cycle, including: The tool overhang length corresponding to the material state of the aluminum profile in the machining feature dataset is matched with the preset process parameter library, and the preset elastic modulus of the tool material and the moment of inertia of the tool section are obtained. Combined with the instantaneous impact force value, the elastic deviation of the tool trajectory is calculated. At the same time, the three-dimensional cutting force components in the current machining cycle are extracted from the interpolated and compensated cutting force signal of the aluminum profile workpiece and combined into a cutting force direction vector. The cutting force direction vector is normalized, and the elastic deviation coordinates of the tool trajectory are determined according to the elastic deviation of the tool trajectory and the normalized cutting force direction vector. The formula for calculating the elastic deviation of the tool trajectory is: ;in, This indicates the amount of elastic deviation in the tool path. This represents the preset deviation influence coefficient, used to correct the stiffness error of the tool clamping boundary. Its value ranges from 0.6 to 1.4, and is determined based on the tool overhang length, tool clamping preload, and tool clamping contact length. This represents the instantaneous impact force value. Indicates the overhang length of the tool. Indicates the elastic modulus of the tool material. This represents the moment of inertia of the tool section; The coordinate deviation of the tool after being subjected to force is determined based on the machining feature dataset. The tool overhang length refers to the exposed length of the tool from the clamping end to the tool tip. The tool material elastic modulus refers to the material parameter of the tool material resisting elastic deformation. The tool section moment of inertia value refers to the geometric resistance parameter of the tool section to bending deformation. The preset deviation influence coefficient refers to the coefficient that corrects for the difference in stiffness of the clamping boundary. The preset process parameter library is established before machining. The data comes from the current workshop machining parameter record table. The record table records the aluminum profile material state, tool overhang length, tool clamping preload value, tool clamping contact length, theoretical normal cutting force data, tool material elastic modulus, and tool section moment of inertia value. The tool overhang length is read from the tool clamping record, the tool clamping preload value is read from the clamping device preload record, the tool clamping contact length is read from the tool clamping length record, and the tool material elastic modulus and tool section moment of inertia value are read from the tool specification record. After MATLAB reads the current aluminum profile material status, it searches the preset process parameter library. If only one record of the same aluminum profile material status is found, the corresponding record is read. If multiple records of the same aluminum profile material status are found, the filtering continues by tool clamping preload value and tool clamping contact length. If no record of the same aluminum profile material status is found, the tool overhang length already in the current feed command is read. The preset deviation influence coefficient ranges from 0.6 to 1.4, with 1 corresponding to the standard clamping boundary record. The longer the tool overhang, the larger the preset deviation influence coefficient; the larger the tool clamping preload, the smaller the preset deviation influence coefficient; and the longer the tool clamping contact length, the smaller the preset deviation influence coefficient. Before machining, the tool overhang, tool clamping preload, and tool clamping contact length are converted to the same dimension. Then, the preset deviation influence coefficient is determined according to the positive influence of the tool overhang, the negative influence of the tool clamping preload, and the negative influence of the tool clamping contact length. If the set value is less than 0.6, the current preset deviation influence coefficient is 0.6; if the set value is greater than 1.4, the current preset deviation influence coefficient is 1.4; if the set value is between 0.6 and 1.4, the current preset deviation influence coefficient is the set value. Subsequently, the three-dimensional cutting force components of the current machining cycle are read. These components refer to the cutting force components in the feed direction, the transverse direction, and the depth of cut. During normalization, the length of the cutting force direction vector is first determined based on the feed direction, transverse direction, and depth of cut force components. Then, the proportions of each component relative to the length of the cutting force direction vector are read. When the length of the cutting force direction vector is 0, the proportions in the feed direction, transverse direction, and depth of cut are all recorded as 0. Finally, the tool path elastic deviation coordinates are obtained.

[0023] Based on the processing feature dataset, the ratio of the energy value of a specified harmonic frequency band to the total energy value of the frequency band is calculated to obtain the energy distribution concentration value. The product of the energy distribution concentration value and the peak amplitude of the reference spectrum is calculated. Based on a preset temperature rise conversion coefficient, this product is converted into a predicted temperature rise value for the cutting contact area. Combined with the length of the heated cutting section of the aluminum profile workpiece, the temperature expansion compensation amount for the cutting contact area is calculated. The length of the heated cutting section of the aluminum profile workpiece is the distance between the starting and ending coordinates of the cutting path of the aluminum profile workpiece collected in the current processing cycle. The temperature rise conversion coefficient is determined according to the material state of the aluminum profile and the current processing temperature range. The formula for calculating the temperature expansion compensation amount for the cutting contact area is: ;in, This indicates the amount of temperature expansion compensation in the cutting contact area. This represents the preset coefficient of thermal expansion of the material, with a value range of [value missing]. The temperature is determined based on the condition of the aluminum profile, the grade of the aluminum profile, and the current processing temperature range. Indicates the length of the heated cutting section of the aluminum profile workpiece. This represents the predicted temperature rise in the cutting contact area; The thermal expansion-related values ​​are determined based on the processing feature dataset. The energy distribution concentration value refers to the proportion of the energy value of the specified harmonic frequency band in the total frequency band energy value. The predicted temperature rise value of the cutting contact area refers to the temperature rise value calculated based on the energy concentration of the spectrum. The length of the heated cutting section of the aluminum profile workpiece refers to the distance record between the starting and ending coordinates of the cutting path within the current processing cycle. MATLAB reads the energy values ​​of the specified harmonic frequency band and the total frequency band energy value. If the total frequency band energy value is greater than 0, the ratio of the energy value of the specified harmonic frequency band to the total frequency band energy value is calculated to obtain the energy distribution concentration value. If the total frequency band energy value is equal to 0, the energy distribution concentration value is recorded as 0. A preset temperature rise conversion coefficient is established before processing, mapping the aluminum profile material state to the current processing temperature range, which is derived from the processing equipment's temperature record. For example, the current processing temperature range can be set to 20℃ to 40℃, greater than 40℃ but not greater than 60℃, or greater than 60℃ but not greater than 80℃, with the specific range field in the processing equipment's temperature record being the determining factor. For the same aluminum profile material state, the higher the corresponding temperature value within the current processing temperature range, the smaller the preset temperature rise conversion coefficient. For different aluminum profile material states, the larger the preset coefficient of thermal expansion, the larger the preset temperature rise conversion coefficient. The predicted temperature rise value in the cutting contact area is calculated by combining the energy distribution concentration value, the reference spectrum peak amplitude, and the preset temperature rise conversion coefficient. Next, the starting and ending coordinates of the cutting path are read, and the distance between these two points in the processing equipment record is used as the length of the heated cutting section of the aluminum profile workpiece. The preset material thermal expansion coefficient is set according to the linear expansion record of the aluminum profile material grade. The specific aluminum profile material grade is based on the material record of the aluminum profile workpiece. The larger the linear expansion record corresponding to the aluminum profile material grade, the larger the preset material thermal expansion coefficient. The higher the temperature value corresponding to the current processing temperature range, the larger the preset material thermal expansion coefficient. If the set value is less than the lower limit of the above value range, the current preset material thermal expansion coefficient is taken as the lower limit of the above value range. If the set value is greater than the upper limit of the above value range, the current preset material thermal expansion coefficient is taken as the upper limit of the above value range. If the set value is within the above value range, the current preset material thermal expansion coefficient is taken as the set value. Finally, the temperature expansion compensation amount of the cutting contact area is obtained.

[0024] The temperature expansion compensation amount of the cutting contact area is mapped along the direction of the preset machining coordinate system to obtain the temperature expansion compensation coordinate of the cutting contact area. The temperature expansion compensation coordinate of the cutting contact area is integrated with the elastic deviation coordinate of the tool trajectory to obtain comprehensive coordinate deviation data. The temperature expansion compensation amount in the cutting contact area is converted into the machining coordinate direction. The preset machining coordinate system refers to the coordinate reference used by the machining equipment to record the relative position of the tool and the workpiece. The feed direction refers to the direction in which the tool moves along the machining path, the transverse direction refers to the direction in the machining plane that intersects with the feed direction, and the depth of cut direction refers to the direction in which the tool enters the workpiece material. The temperature expansion compensation coordinate of the cutting contact area refers to the components of the temperature expansion compensation amount in the cutting contact area in each coordinate direction. MATLAB reads the direction definition of the preset machining coordinate system and reads the starting and ending coordinates of the cutting path; the path direction vector is determined by the direction record of the ending coordinates relative to the starting coordinates. If the path direction falls only in the feed direction, the feed direction component takes the temperature expansion compensation amount of the cutting contact area, while the transverse and depth-of-cut components are set to 0. If the path direction falls only in the transverse direction, the transverse component takes the temperature expansion compensation amount of the cutting contact area, while the feed and depth-of-cut components are set to 0. If the path direction falls only in the depth-of-cut direction, the depth-of-cut component takes the temperature expansion compensation amount of the cutting contact area, while the feed and transverse components are set to 0. If the path direction falls in two or three coordinate directions simultaneously, the proportion of the path direction in each coordinate direction is read, and the temperature expansion compensation amount of the cutting contact area is allocated according to the proportion. The proportion is determined by the relationship between the component of the path direction vector in the corresponding coordinate direction and the length of the path direction vector. The length of the path direction vector is jointly determined by the path components of the feed direction, the transverse direction, and the depth-of-cut direction. Subsequently, the tool path elastic deviation coordinates are read and recorded in alignment with the same coordinate direction. When there are both temperature expansion compensation coordinates for the cutting contact area and elastic deviation coordinates for the tool path in the same coordinate direction, the values ​​of the temperature expansion compensation coordinates for the cutting contact area and the elastic deviation coordinates for the tool path are recorded separately. Finally, the comprehensive coordinate deviation data is obtained.

[0025] S4: Obtain the next feed command to be processed for driving the tool, determine the tool displacement correction interpolation path and cutting depth correction coordinates by combining the comprehensive coordinate deviation data, and perform coordinate and path mapping combination in the preset three-dimensional space. The adjusted tool space trajectory data includes: Obtain the next feed command to be processed to drive the tool, extract the theoretical coordinate data and cutting depth parameters from the feed command, and unify the theoretical coordinate data, tool trajectory elastic deviation coordinates and cutting contact area temperature expansion compensation coordinates into the preset machining coordinate system. The next feed instruction to be processed is read. The feed instruction refers to the tool movement record that has not yet been executed in the machining equipment. The theoretical coordinate data refers to the target tool position given by the feed instruction, and the depth of cut parameter refers to the depth to which the tool enters the aluminum profile workpiece material along the cutting depth direction. After the control record of the machining equipment is imported into MATLAB, MATLAB sorts the feed instructions by sequence number from smallest to largest. Executed feed instructions are marked as completed, and the feed instruction that has not been executed and has the earliest sequence number is read as the current feed instruction to be processed. Then, the theoretical coordinate data and depth of cut parameter in the current feed instruction to be processed are read, along with the corresponding tool trajectory elastic deviation coordinates and cutting contact area temperature expansion compensation coordinates from the comprehensive coordinate deviation data. When verifying coordinate references, if the theoretical coordinate data, tool path elastic deviation coordinates, and cutting contact area temperature expansion compensation coordinates all use the preset machining coordinate system, the original coordinates are retained. If the theoretical coordinate data uses other coordinate references, the coordinate origin offset record and corresponding coordinate direction record corresponding to the theoretical coordinate data are read and then converted to the preset machining coordinate system. If the tool path elastic deviation coordinates use other coordinate references, the coordinate origin offset record and corresponding coordinate direction record corresponding to the tool path elastic deviation coordinates are read and then converted to the preset machining coordinate system. If the cutting contact area temperature expansion compensation coordinates use other coordinate references, the coordinate origin offset record and corresponding coordinate direction record corresponding to the cutting contact area temperature expansion compensation coordinates are read and then converted to the preset machining coordinate system. During coordinate transformation, the original coordinate reference coordinates are read first, and then the origin offset is superimposed. When the coordinate directions are opposite, the corresponding components are recorded in opposite directions. Finally, the theoretical coordinate data, tool path elastic deviation coordinates, and cutting contact area temperature expansion compensation coordinates under a unified coordinate reference are obtained.

[0026] The difference between the elastic deviation coordinates of the tool trajectory in the theoretical coordinate data and the comprehensive coordinate deviation data is calculated to obtain the tool displacement correction interpolation path. The difference between the cutting depth parameter and the component of the temperature expansion compensation coordinate of the cutting contact area in the cutting depth direction in the comprehensive coordinate deviation data is calculated to obtain the cutting depth correction coordinate. Based on unified coordinate data, a correction path and correction depth are generated. The tool displacement correction interpolation path refers to the sequence of feed direction coordinate points after deducting the elastic deviation coordinates of the tool trajectory. The cutting depth correction coordinates refer to the depth value after deducting the component of the temperature expansion compensation coordinates of the cutting contact area in the cutting depth direction. MATLAB reads the coordinate components of the theoretical coordinate data and the coordinate components of the elastic deviation coordinates of the tool trajectory in the same coordinate direction of the preset machining coordinate system. If the coordinate component of the elastic deviation coordinates of the tool trajectory is positive, the corresponding coordinate component of the theoretical coordinate data is corrected in the opposite direction; if the coordinate component of the elastic deviation coordinates of the tool trajectory is negative, the corresponding coordinate component of the theoretical coordinate data is corrected in the opposite direction; if the coordinate component of the elastic deviation coordinates of the tool trajectory is 0, the corresponding coordinate component of the theoretical coordinate data remains unchanged. Subsequently, the cutting depth parameters and the component of the temperature expansion compensation coordinates of the cutting contact area in the cutting depth direction are read. If the component of the temperature expansion compensation coordinate in the cutting contact area along the cutting depth direction is positively compensated, the cutting depth parameter is corrected in the opposite direction; if the component is negatively compensated, the cutting depth parameter is corrected in the opposite direction; if the component is zero, the cutting depth parameter remains unchanged. Finally, the tool displacement correction interpolation path and the cutting depth correction coordinates are obtained.

[0027] The tool displacement correction interpolation path is used as the feed direction movement trajectory corresponding to the theoretical coordinate data in the preset machining coordinate system, and the cutting depth correction coordinate is used as the depth coordinate in the cutting depth direction in the preset machining coordinate system. The tool displacement correction interpolation path and the cutting depth correction coordinate are mapped and combined in the preset three-dimensional space within the preset machining coordinate system to generate the adjusted tool space trajectory data. The tool displacement correction interpolation path and the depth of cut correction coordinates are combined into a three-dimensional trajectory. The preset three-dimensional space refers to the range of tool position recording defined by the feed direction, transverse direction, and depth of cut direction. The feed direction movement trajectory refers to the correction coordinate points arranged according to the feed command sequence, and the depth coordinates refer to the depth of cut direction coordinates under the same time sequence. MATLAB first arranges the correction coordinate points in the tool displacement correction interpolation path according to the feed command sequence number and the initial timestamp, and then finds the depth of cut correction coordinates under the same time sequence for each correction coordinate point. If a feed direction coordinate point has a cutting depth correction coordinate in the same time sequence, then the feed direction coordinate point and the cutting depth correction coordinate are combined according to the corresponding coordinate direction of the preset machining coordinate system to form the current three-dimensional space coordinates. If a feed direction coordinate point does not have a cutting depth correction coordinate in the same time sequence, and there is a valid cutting depth correction coordinate preceding the feed direction coordinate point, then the feed direction coordinate point is combined with the previous valid cutting depth correction coordinate according to the corresponding coordinate direction to form the current three-dimensional space coordinates. If the first feed direction coordinate point does not have a valid cutting depth correction coordinate, then the first feed direction coordinate point is combined with the cutting depth parameter in the current feed command according to the corresponding coordinate direction to form the current three-dimensional space coordinates. Then, all three-dimensional space coordinates are arranged according to the time sequence of the coordinate points. Finally, the adjusted tool space trajectory data is obtained.

[0028] S5: Obtain the current feed rate parameters of the tool, calculate the adjusted tool feed rate based on the machining feature dataset, and correct the tool feed rate and movement trajectory within the current machining cycle according to the adjusted tool feed rate and adjusted tool spatial trajectory data to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces. The optimized machining results for aluminum profiles include: Obtain the current feed rate parameters of the tool and the instantaneous impact force value from the machining feature dataset, and calculate the adjusted tool feed rate; the formula for calculating the adjusted tool feed rate is: ;in, This indicates the adjusted tool feed rate. This indicates the current feed rate parameter of the tool. This represents the force-velocity conversion coefficient, which is the ratio of velocity to force. This represents the instantaneous impact force value. This represents the theoretical normal cutting force data; The new feed rate is determined based on the machining feature dataset and the current speed record. The current feed rate parameter refers to the speed at which the tool moves along the feed direction within the current machining cycle. The theoretical normal cutting force data refers to the cutting force benchmark value in the preset process parameter library corresponding to the current aluminum profile material state, cutting depth parameter, tool overhang length, and current feed rate parameter. The force-speed conversion coefficient refers to the ratio of force deviation to speed change. The force-speed conversion coefficient is set before machining by the preset upper feed rate limit, preset lower feed rate limit, theoretical normal cutting force data, and preset impact force judgment threshold. The wider the speed range between the preset upper and lower feed rate limits, the larger the force-speed conversion coefficient; the larger the difference between the preset impact force judgment threshold and the theoretical normal cutting force data, the smaller the force-speed conversion coefficient; the longer the tool overhang length, the larger the force-speed conversion coefficient. If there is no usable difference between the preset impact force judgment threshold and the theoretical normal cutting force data, the force-speed conversion coefficient corresponding to the adjacent cutting depth parameter under the same aluminum profile material state is read from the preset process parameter library. MATLAB reads the current feed rate parameters of the tool, the instantaneous impact force value in the machining feature dataset, the theoretical normal cutting force data, and the force-speed conversion coefficient, and takes the difference between the instantaneous impact force value and the theoretical normal cutting force data as the force deviation; then, based on the current feed rate parameters of the tool, the force-speed conversion coefficient, and the force deviation, it calculates the adjusted tool feed rate.

[0029] Based on the preset upper and lower limits of feed rate, the adjusted tool feed rate is subjected to speed limiting processing, and based on the preset acceleration constraint, the adjusted tool feed rate after speed limiting processing is subjected to acceleration constraint processing to obtain the constrained adjusted tool feed rate. The adjusted tool feed rate is subject to speed range and rate of change limits. The preset feed rate upper limit refers to the maximum allowable feed rate under the current machining conditions, the preset feed rate lower limit refers to the minimum allowable feed rate under the current machining conditions, and the preset acceleration constraint refers to the upper limit of the allowable rate of change of speed between adjacent machining cycles. The preset feed rate upper and lower limits are read from the preset process parameter library and correspond to the aluminum profile material state, tool overhang length, and depth of cut parameters. The larger the depth of cut parameter, the smaller the preset feed rate upper limit; the longer the tool overhang length, the smaller the preset feed rate upper limit; the higher the material hardness record corresponding to the aluminum profile material state, the smaller the preset feed rate lower limit. The preset acceleration constraint is set by the current axial drive capability record of the machining equipment and the speed record of the previous machining cycle. The larger the drive capability record, the larger the preset acceleration constraint; the closer the speed of the previous machining cycle is to the preset feed rate upper limit, the smaller the preset acceleration constraint. MATLAB reads the adjusted tool feed rate. If the adjusted tool feed rate is greater than the preset upper limit, the preset upper limit is used as the current speed limiting result; if the adjusted tool feed rate is equal to the preset upper limit, the adjusted tool feed rate is used as the current speed limiting result; if the adjusted tool feed rate is less than the preset lower limit, the preset lower limit is used as the current speed limiting result; if the adjusted tool feed rate is equal to the preset lower limit, the adjusted tool feed rate is used as the current speed limiting result; if the adjusted tool feed rate is between the preset lower limit and the preset upper limit, the adjusted tool feed rate is used as the current speed limiting result. Then, the feed rate of the previous machining cycle and the time interval of the current machining cycle are read. Based on the change in feed rate of the speed limiting result relative to the feed rate of the previous machining cycle, and the time interval of the current machining cycle, the feed rate change rate is obtained. If the feed rate changes at a rate greater than the preset acceleration constraint, the speed limiting result is constrained based on the preset acceleration constraint, the current machining cycle time interval, and the direction of speed change. If the feed rate changes at a rate equal to the preset acceleration constraint, the speed limiting result is used as the current constrained speed. If the feed rate changes at a rate less than the preset acceleration constraint, the speed limiting result is used as the current constrained speed. Finally, the constrained adjusted tool feed rate is obtained.

[0030] Based on the constrained adjusted tool feed rate and adjusted tool spatial trajectory data, the tool feed rate and movement trajectory within the current machining cycle are adjusted to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, thereby obtaining the optimized machining results of aluminum profiles. The adjusted tool space trajectory data and the constrained adjusted tool feed rate are written into the current machining cycle record. The aluminum profile machining optimization result refers to the corresponding records between the feed rate parameters, movement trajectory parameters, machining feature dataset, and comprehensive coordinate deviation data within the current machining cycle. MATLAB reads the constrained adjusted tool feed rate and the three-dimensional coordinate point sequence from the adjusted tool space trajectory data; the three-dimensional coordinate point sequence refers to the tool position record arranged by time or feed command number. If the order of the three-dimensional coordinate point sequence is consistent with the order of the feed commands, it is written into the movement trajectory parameters in the existing order; if the order of the three-dimensional coordinate point sequence is inconsistent with the order of the feed commands, it is rearranged from earliest to latest according to the timestamp, and if the timestamps are the same, they are arranged from smallest to largest according to the feed command number. If a coordinate point lacks a cutting depth direction coordinate but the previous coordinate point has a valid cutting depth direction coordinate, then the previous valid cutting depth direction coordinate is used as the cutting depth direction coordinate of the current point. If a coordinate point lacks a cutting depth direction coordinate and the previous coordinate point does not have a valid cutting depth direction coordinate, then the cutting depth parameter in the current feed command is used as the cutting depth direction coordinate of the current point. If a coordinate point lacks a feed direction coordinate, then the theoretical coordinate data in the feed command is used as the feed direction coordinate of the current point. If both the feed direction coordinate and the cutting depth direction coordinate of a coordinate point are complete, then the coordinate point remains unchanged. Subsequently, the current machining cycle number is written into the corresponding fields of the feed rate parameter, movement trajectory parameter, machining feature dataset, and comprehensive coordinate deviation data. Finally, the optimized machining results for aluminum profiles are obtained.

[0031] A high-precision machining optimization system for aluminum profiles, the system comprising: The signal synchronization module acquires a multi-source machining signal sensing dataset based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component during the machining process of the aluminum profile workpiece. The feature recognition module analyzes the features corresponding to each signal in the multi-source processing signal sensing dataset to obtain the processing feature dataset. The deviation calculation module determines the comprehensive coordinate deviation data of the tool within the current machining cycle based on the machining feature dataset. The trajectory correction module obtains the next feed command to be processed for driving the tool, determines the tool displacement correction interpolation path and cutting depth correction coordinates by combining comprehensive coordinate deviation data, and maps and combines the coordinates and paths in a preset three-dimensional space as the adjusted tool space trajectory data. The drive control module acquires the current feed rate parameters of the tool, calculates the adjusted tool feed rate based on the machining feature dataset, and corrects the tool feed rate and movement trajectory within the current machining cycle based on the adjusted tool feed rate and adjusted tool spatial trajectory data. This reduces the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, resulting in optimized aluminum profile machining.

[0032] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for optimizing high-precision machining of aluminum profiles, characterized in that, Includes the following steps: Acquire a multi-source machining signal sensing dataset based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component during the machining process of aluminum profile workpieces; By analyzing the features corresponding to each signal in the multi-source processing signal sensing dataset, a processing feature dataset is obtained; Based on the machining feature dataset, determine the comprehensive coordinate deviation data of the tool within the current machining cycle; The next feed command to be processed is obtained to drive the tool. The tool displacement correction interpolation path and cutting depth correction coordinates are determined by combining the comprehensive coordinate deviation data. The coordinates and paths are mapped and combined in the preset three-dimensional space as the adjusted tool space trajectory data. The current feed rate parameters of the tool are obtained, and the adjusted tool feed rate is calculated by combining the machining feature dataset. Based on the adjusted tool feed rate and the adjusted tool spatial trajectory data, the tool feed rate and movement trajectory in the current machining cycle are corrected to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, and the optimization results of aluminum profile machining are obtained.

2. The high-precision machining optimization method for aluminum profiles according to claim 1, characterized in that, The multi-source machining signal sensing dataset obtained during the aluminum profile workpiece machining process, based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component, includes: During the processing of aluminum profile workpieces, the cutting force signal of the aluminum profile workpiece is obtained by a piezoelectric triaxial force sensor, the vibration signal of the machine tool housing is obtained by an accelerometer, and the acoustic emission signal of the tool clamping component is obtained by an acoustic emission sensor. Initial timestamps are assigned to the sampling points in the sampling point sequences of the aluminum profile workpiece cutting force signal, the machine tool housing vibration signal, and the acoustic emission signal of the tool clamping component. The initial timestamps of the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component are compared to identify the misaligned time nodes between the signals, and the amplitude of the missing sampling points between the misaligned time nodes is calculated. The amplitude of the missing sampling points is added to the sampling point sequence of each signal to generate the interpolated aluminum profile workpiece cutting force signal, the interpolated machine tool housing vibration signal, and the interpolated tool clamping component acoustic emission signal, which are combined into a multi-source machining signal sensing dataset.

3. The high-precision machining optimization method for aluminum profiles according to claim 2, characterized in that, The analysis of the features corresponding to each signal in the multi-source processing signal sensing dataset yields a processing feature dataset including: Fast Fourier Transform (FFT) processing is performed on the aluminum profile workpiece cutting force signal and the machine tool housing vibration signal after centralized interpolation and compensation of the multi-source processing signal sensing data to generate cutting force frequency domain representation data and vibration frequency domain representation data, respectively. The cutting force frequency point sequence and the corresponding cutting force spectrum amplitude sequence are determined from the cutting force frequency domain representation data. The vibration frequency point sequence and the corresponding vibration spectrum amplitude sequence are determined from the vibration frequency domain representation data. Corresponding frequency points are selected from the cutting force frequency point sequence and the vibration frequency point sequence according to the preset harmonic frequency band range. The energy value of the specified harmonic frequency band is determined according to the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence corresponding to the corresponding frequency point. The cumulative amplitude result of the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence within the preset total frequency band range is used as the total frequency band energy value. The spectrum amplitude corresponding to the spectrum peak position in the cutting force spectrum amplitude sequence and the vibration spectrum amplitude sequence is used as the reference spectrum peak amplitude. Based on the initial timestamp of the sampling point and the corresponding cutting force amplitude in the cutting force signal of the aluminum profile workpiece after interpolation compensation, the instantaneous rise rate of the amplitude is calculated. At the same time, the target sampling point in the current processing cycle where the change in cutting force amplitude exceeds the preset impact force judgment threshold is determined. The cutting force amplitude corresponding to the target sampling point is determined as the instantaneous impact force value. Based on the instantaneous impact force value, the envelope demodulation processing of the acoustic emission signal of the tool clamping component after interpolation compensation is performed to extract the instantaneous acoustic emission energy value. The specified harmonic frequency band energy value, the total frequency band energy value, and the reference spectrum peak amplitude are used as common features of the aluminum profile workpiece cutting force signal and the machine tool housing vibration signal. The instantaneous rise rate of the amplitude and the instantaneous impact force value determined based on the aluminum profile workpiece cutting force signal are combined with the corresponding instantaneous acoustic emission energy value as features of the acoustic emission signal of the tool clamping component, and these are combined into a feature value sequence. The cosine similarity between the entire feature value sequence and each reference mode vector in the preset aluminum profile material state mode library is calculated. The aluminum profile material state associated with the reference mode vector with the maximum cosine similarity in the state mode library is selected. The feature value sequence and the corresponding aluminum profile material state are integrated to obtain the processing feature dataset.

4. The high-precision machining optimization method for aluminum profiles according to claim 3, characterized in that, The determination of the comprehensive coordinate deviation data of the tool within the current machining cycle based on the machining feature dataset includes: The tool overhang length corresponding to the aluminum profile material state in the processing feature dataset is matched from the preset process parameter library, and the preset tool material elastic modulus and tool section moment of inertia value are obtained. Combined with the instantaneous impact force value, the tool trajectory elastic deviation is calculated. At the same time, the three-dimensional cutting force components in the current processing cycle are extracted from the interpolated aluminum profile workpiece cutting force signal and combined into a cutting force direction vector. The cutting force direction vector is normalized. Based on the tool trajectory elastic deviation and the normalized cutting force direction vector, the tool trajectory elastic deviation coordinates are determined. Based on the processing feature dataset, the ratio of the energy value of the specified harmonic frequency band to the energy value of the total frequency band is calculated to obtain the energy distribution concentration value. The product of the energy distribution concentration value and the peak amplitude of the reference spectrum is calculated. Based on the preset temperature rise conversion coefficient, the product is converted into the temperature rise prediction value of the cutting contact area. Combined with the length of the heated cutting section of the aluminum profile workpiece, the temperature expansion compensation amount of the cutting contact area is calculated. The length of the heated cutting section of the aluminum profile workpiece is the distance between the starting coordinates and the ending coordinates of the cutting path of the aluminum profile workpiece collected in the current processing cycle. The temperature expansion compensation amount of the cutting contact area is mapped along the direction of the preset machining coordinate system to obtain the temperature expansion compensation coordinate of the cutting contact area. The temperature expansion compensation coordinate of the cutting contact area is integrated with the elastic deviation coordinate of the tool trajectory to obtain comprehensive coordinate deviation data.

5. The high-precision machining optimization method for aluminum profiles according to claim 3, characterized in that, The reference mode vector includes the specified harmonic frequency band energy value, total frequency band energy value, reference spectrum peak amplitude, instantaneous amplitude rise rate, instantaneous impact force value, and the corresponding reference values ​​of the instantaneous acoustic emission energy value in the preset aluminum profile material state mode library.

6. The high-precision machining optimization method for aluminum profiles according to claim 4, characterized in that, The temperature rise conversion coefficient is determined based on the material condition of the aluminum profile and the current processing temperature range.

7. The high-precision machining optimization method for aluminum profiles according to claim 4, characterized in that, The process of acquiring the next feed command to drive the tool, determining the tool displacement correction interpolation path and cutting depth correction coordinates by combining comprehensive coordinate deviation data, and mapping and combining the coordinates and paths in a preset three-dimensional space to obtain the adjusted tool space trajectory data includes: Obtain the next feed command to be processed to drive the tool, extract the theoretical coordinate data and cutting depth parameters from the feed command, and unify the theoretical coordinate data, tool trajectory elastic deviation coordinates and cutting contact area temperature expansion compensation coordinates into the preset machining coordinate system. The difference between the elastic deviation coordinates of the tool trajectory in the theoretical coordinate data and the comprehensive coordinate deviation data is calculated to obtain the tool displacement correction interpolation path. The difference between the cutting depth parameter and the component of the temperature expansion compensation coordinate of the cutting contact area in the cutting depth direction in the comprehensive coordinate deviation data is calculated to obtain the cutting depth correction coordinate. The tool displacement correction interpolation path is used as the feed direction movement trajectory corresponding to the theoretical coordinate data in the preset machining coordinate system, and the cutting depth correction coordinate is used as the depth coordinate in the cutting depth direction in the preset machining coordinate system. The tool displacement correction interpolation path and the cutting depth correction coordinate are mapped and combined in the preset three-dimensional space within the preset machining coordinate system to generate adjusted tool space trajectory data.

8. The high-precision machining optimization method for aluminum profiles according to claim 7, characterized in that, The process involves acquiring the current feed rate parameters of the cutting tool, calculating the adjusted feed rate using the machining feature dataset, and then correcting the tool's feed rate and movement trajectory within the current machining cycle based on the adjusted feed rate and spatial trajectory data. This aims to reduce the impact of feed rate deviation and trajectory deviation on the machining accuracy of the aluminum profile workpiece, resulting in optimized aluminum profile machining results, including: Obtain the current feed rate parameters of the tool and the instantaneous impact force value in the machining feature dataset, and calculate the adjusted tool feed rate; Based on the preset upper limit and lower limit of feed rate, the adjusted tool feed rate is subjected to speed limiting processing, and based on the preset acceleration constraint, the adjusted tool feed rate after speed limiting processing is subjected to acceleration constraint processing to obtain the constrained adjusted tool feed rate. Based on the constrained adjusted tool feed rate and the adjusted tool spatial trajectory data, the tool feed rate and movement trajectory within the current machining cycle are adjusted to reduce the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, thereby obtaining optimized aluminum profile machining results.

9. The high-precision machining optimization method for aluminum profiles according to claim 8, characterized in that, The formula for calculating the adjusted tool feed rate is as follows: ； in, This indicates the adjusted tool feed rate. This indicates the current feed rate parameter of the tool. This represents the force-velocity conversion coefficient, which is the ratio of velocity to force. This represents the instantaneous impact force value. This represents the theoretical normal cutting force data.

10. A high-precision machining optimization system for aluminum profiles, characterized in that, The system includes: The signal synchronization module acquires a multi-source machining signal sensing dataset based on the cutting force signal of the aluminum profile workpiece, the vibration signal of the machine tool housing, and the acoustic emission signal of the tool clamping component during the machining process of the aluminum profile workpiece. The feature recognition module analyzes the features corresponding to each signal in the multi-source processing signal sensing dataset to obtain the processing feature dataset. The deviation calculation module determines the comprehensive coordinate deviation data of the tool within the current machining cycle based on the machining feature dataset. The trajectory correction module obtains the next feed command to be processed for driving the tool, determines the tool displacement correction interpolation path and cutting depth correction coordinates by combining comprehensive coordinate deviation data, and maps and combines the coordinates and paths in a preset three-dimensional space as the adjusted tool space trajectory data. The drive control module acquires the current feed rate parameters of the tool, calculates the adjusted tool feed rate based on the machining feature dataset, and corrects the tool feed rate and movement trajectory within the current machining cycle based on the adjusted tool feed rate and adjusted tool spatial trajectory data. This reduces the impact of tool feed rate deviation and movement trajectory deviation on the machining accuracy of aluminum profile workpieces, resulting in optimized aluminum profile machining.

Images