Method for boring a stepped deep hole with an adjustable function precision boring tool

Abstract

The present application relates to the technical field of boring processing, in particular to a boring processing method of a step deep hole fine boring cutter with adjustable function. Firstly, based on the vibration data and temperature data collected by the boring cutter in real time, the time domain and frequency domain characteristics of the vibration data are analyzed in real time by using a sliding window, so as to obtain a vibration instability value. Then, according to the vibration data and temperature data of the boring cutter in historical normal processing, a preset vibration deviation value range and a preset normal temperature value range are respectively constructed, so as to identify the abnormal time of the vibration data and temperature data. Finally, according to the time interval of the abnormal time, the abnormal proportion of the vibration data in the time interval and the vibration instability value, a boring cutter instability coefficient is calculated, and the fixed boring parameters are adjusted according to the boring cutter instability coefficient, so as to realize real-time monitoring and closed-loop control of the processing process, and effectively improve the boring stability and processing precision.

Description

Technical Field

[0001] This invention relates to the field of boring technology, specifically to a boring method for precision boring of deep, elongated holes with adjustable functions. Background Technology

[0002] Deep hole boring tools are mainly used for deep hole machining. Their structure includes the tool body, guide block, and cooling channel system. Stepped boring tools can complete multiple machining tasks of different diameters in one operation, effectively reducing tool changes. In finish boring, the boring tool performs precision cutting within the workpiece's inner hole, resulting in a bored hole structure with high dimensional accuracy and surface quality. Therefore, finish boring requires high stability and rationality of cutting parameters, necessitating control of these parameters to ensure machining accuracy. However, boring tool wear can cause vibration, leading to uneven cutting forces and unstable temperature changes during boring.

[0003] In existing control methods, a Programmable Logic Controller (PLC) is typically used to set and adjust basic parameters such as cutting speed and feed rate during the boring process. However, this method does not consider that the wear of the boring tool will continuously increase during the boring process, and the probability of boring tool vibration will also continuously increase. When the boring tool vibration exceeds the stable critical point, uneven stress will occur on the boring tool, leading to an increase in temperature during the boring process and a reduction in boring accuracy. Controlling basic cutting parameters based on fixed parameters lacks the ability to monitor and dynamically adjust these dynamic changes in real time, easily overlooking the actual state changes during the boring process, resulting in boring accuracy failing to meet process requirements. Summary of the Invention

[0004] To address the technical problem of low boring accuracy caused by using fixed parameters to control boring without fully considering the boring tool's condition, the present invention aims to provide a boring method for precision boring of deep, elongated holes with adjustable parameters. The specific technical solution adopted is as follows:

[0005] This invention proposes a method for boring deep, elongated holes with an adjustable precision boring tool, the method comprising:

[0006] Real-time acquisition of vibration and temperature data of the boring tool;

[0007] A sliding window is performed on the vibration data time series. Based on the fluctuation degree and distribution pattern of the vibration data within the sliding window, the time-domain wear confidence of the sliding window is obtained. The high-frequency components of the vibration data within the sliding window are obtained. Combining the proportion of the high-frequency components and the time-domain wear confidence, the vibration instability value is obtained.

[0008] The sampling time when the vibration instability value exceeds the preset vibration deviation range is taken as the vibration abnormality time; the sampling time when the temperature data exceeds the preset normal temperature range is taken as the temperature abnormality time.

[0009] The abnormal interval is obtained by the time interval between the vibration abnormal moment and the temperature abnormal moment in time sequence. The boring tool instability coefficient is obtained by the proportion of vibration abnormal moments within the abnormal interval, the vibration instability value and the abnormal interval.

[0010] Adjusting boring parameters based on the boring tool instability coefficient.

[0011] Furthermore, the method for obtaining the time-domain wear confidence includes:

[0012] The vibration discrete value is determined based on the degree of fluctuation of the vibration data within the sliding window;

[0013] The vibration distribution value is determined based on the distribution pattern of the vibration data within the sliding window;

[0014] By combining the vibration discrete value and the vibration distribution value, the time-domain wear confidence level is obtained; wherein, both the vibration discrete value and the vibration distribution value are positively correlated with the time-domain wear confidence level.

[0015] Furthermore, the vibration distribution value is the sum of the absolute values ​​of kurtosis and skewness of the vibration data within the sliding window.

[0016] Furthermore, the method for obtaining the high-frequency components includes:

[0017] The vibration data within the reference window is converted into the frequency domain to obtain the dominant frequency of the reference window; the reference window is a sliding window on the time sequence of vibration data corresponding to the processing process with the highest processing accuracy in the historical processing process.

[0018] The cutoff frequency of the filter is determined based on the dominant frequency of the reference window.

[0019] Based on the filter's cutoff frequency, the high-frequency components of the sliding window are obtained.

[0020] Furthermore, the method for obtaining the vibration instability value includes:

[0021] Using any sliding window as the target window, calculate the proportion of the high-frequency components of the target window in the total energy of the target window, and obtain the proportion of high-frequency energy of the window.

[0022] By combining the time-domain wear confidence level and the high-frequency energy ratio, the vibration instability value is obtained; wherein, both the time-domain wear confidence level and the high-frequency energy ratio are positively correlated with the vibration instability value.

[0023] Furthermore, the method for obtaining the moment of abnormal vibration includes:

[0024] Real-time assessment of whether the vibration instability value of the sliding window is within the preset vibration deviation range corresponding to the sliding window;

[0025] When the vibration instability value of the sliding window exceeds the corresponding preset vibration deviation range, the sampling time corresponding to the first vibration data in the sliding window is recorded as the vibration abnormality time.

[0026] Furthermore, the method for obtaining the preset vibration deviation range corresponding to the sliding window includes:

[0027] Based on the vibration instability value of each historical sliding window, all historical sliding windows are clustered to obtain multiple clusters; the historical sliding window is a sliding window on the time sequence of historical vibration data during multiple historical normal processing processes;

[0028] Based on the difference between the mean of elements in each cluster and the vibration instability value of the real-time sliding window, the cluster to which the real-time sliding window belongs is obtained.

[0029] The maximum and minimum values ​​of vibration instability within the cluster to which the real-time sliding window belongs are used as the upper and lower limits of the preset vibration deviation range corresponding to the real-time sliding window.

[0030] Furthermore, the method for obtaining the preset normal temperature range includes:

[0031] The maximum and minimum values ​​from historical temperature data during multiple normal processing processes are used as the upper and lower limits of the preset normal temperature range.

[0032] Furthermore, the method for obtaining the instability coefficient of the boring tool includes:

[0033] By combining the proportion of abnormal vibration moments and the vibration instability value, a comprehensive abnormality characteristic is obtained;

[0034] The boring tool instability coefficient is determined based on the abnormal interval and the comprehensive abnormal characteristics. The abnormal interval is negatively correlated with the boring tool instability coefficient, while the comprehensive abnormal characteristics are positively correlated with the boring tool instability coefficient.

[0035] Furthermore, the adjustment of boring parameters based on the boring tool instability coefficient includes:

[0036] The initial boring parameters are weighted by the boring tool instability coefficient to obtain the boring parameter adjustment value;

[0037] Based on the aforementioned boring parameter adjustment value, the initial boring parameters are adjusted.

[0038] The present invention has the following beneficial effects:

[0039] This invention addresses the characteristics of boring tool wear accumulating gradually with cutting time and the continuous changes in machining conditions during precision boring. Based on historical normal machining data, it constructs preset vibration deviation ranges and preset normal temperature ranges to characterize reference standards under normal precision boring conditions. During real-time machining, a vibration instability value reflecting the degree of vibration anomaly is constructed through comprehensive analysis of vibration data in the time and frequency domains, thus quantitatively characterizing the deviation of the boring tool vibration state from normal operating conditions. Subsequently, based on the temporal evolution relationship between vibration anomalies and temperature anomalies, a boring tool instability coefficient is further constructed to characterize the intensity and speed of the evolution from vibration anomalies to thermal anomalies, thereby achieving a comprehensive assessment of the overall instability of the boring tool during precision boring. Finally, the boring parameters are dynamically adjusted using the boring tool instability coefficient, enabling the control strategy to adaptively update with changes in machining conditions. This achieves real-time closed-loop optimization control of the precision boring process, effectively extending the boring tool's service life and reducing the risk of workpiece machining deviations due to vibration instability. Attached Figure Description

[0040] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 A flowchart illustrating the steps of a step-by-step boring method for precision boring of deep elongated holes with adjustable function, provided in an embodiment of the present invention.

[0042] Figure 2 This is a flowchart of a method for obtaining abnormal vibration moments according to an embodiment of the present invention;

[0043] Figure 3 This is a flowchart illustrating a method for obtaining a preset vibration deviation range corresponding to a sliding window, as provided in an embodiment of the present invention. Detailed Implementation

[0044] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the step-deep long hole precision boring method with adjustable function proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0046] The specific solution of the step-deep long hole precision boring method with adjustable function provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0047] This invention proposes a boring method for deep, elongated holes with adjustable precision boring tools. Please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a flowchart of a method for boring deep elongated holes with an adjustable function using a precision boring tool according to an embodiment of the present invention. The method includes:

[0048] Step S101: Acquire vibration and temperature data of the boring tool in real time.

[0049] In the precision boring process of deep, elongated holes with stepped surfaces, it is necessary to collect key physical quantities that reflect the operating status of the boring tool in order to achieve real-time perception of the machining conditions and real-time adjustment of boring parameters. At the same time, before the precision boring stage, it is also necessary to complete the basic preparation of the workpiece, boring tool, and process parameters and perform rough machining to lay the foundation for subsequent stable and reliable data acquisition and analysis.

[0050] In one specific implementation of this invention, the specific steps for basic preparation and data acquisition in the processing procedure are as follows:

[0051] Workpiece preparation, boring tool installation, and boring parameter setting: Using the workpiece's cross-section and outer diameter as a reference, the workpiece is positioned using a "one-face, two-pin" method to ensure its correct position on the CNC machine tool. The workpiece is then evenly clamped with a pressure plate to ensure the coaxiality between the workpiece and the machine spindle is ≤0.05mm, preventing workpiece displacement during machining. Next, the stepped deep boring tool is installed on the CNC machine tool, with the tool tip aligned with the hole center. Finally, appropriate rough boring cutting parameters are determined based on the workpiece material and machining requirements. The basic parameters for rough boring are: cutting speed of 100-150m / min, feed rate of 0.3-0.8mm / r, and depth of cut of 2-5mm per pass.

[0052] In this embodiment, the CNC machine tool is a CK6150 CNC machine tool, the coaxiality between the workpiece and the machine tool spindle is 0.03mm, the rough boring cutting speed is 120m / min, the feed rate is 0.5mm / r, and the single cutting depth is 3mm.

[0053] Error calibration: Before machining, a trial boring is performed: On the reserved material, a process hole with the same diameter as the target hole is trial boring. The actual dimension X of the process hole diameter is measured, and the scale value Y corresponding to the boring tool diameter is recorded. The error is calculated by subtracting the scale value Y corresponding to the boring tool from the actual dimension X, and then compensated for in the subsequent target machining dimension of the process hole on the actual workpiece.

[0054] Rough boring: Rough boring has relatively low precision requirements and is a pre-process for fine boring. It controls the hole size tolerance within IT10-IT9 and the surface roughness within Ra3.2-Ra6.3. In this embodiment, the hole size tolerance is IT9 and the surface roughness is Ra6.3.

[0055] Fine boring: Appropriate fine boring cutting parameters are determined based on the workpiece material and machining requirements. The basic parameters for fine boring are: cutting speed of 60-120 m / min, feed rate of 0.1-0.2 mm / r, and depth of cut ≤ 0.5 mm. The reasonable range of vibration and temperature is determined through historical vibration and temperature data, and the fine boring cutting parameters are dynamically adjusted based on real-time data anomalies. In this embodiment, the cutting speed for fine boring is 100 m / min, the feed rate is 0.2 mm / r, and the depth of cut is 0.3 mm.

[0056] Data Acquisition: During the final step of the boring process, finish boring, a vibration acceleration sensor is used to acquire vibration data of the boring tool. Simultaneously, a wireless thermocouple sensor located inside the tool holder acquires temperature data of the boring tool during finish boring. The data acquisition frequency is 100Hz, and the acquired vibration and temperature data are synchronized in time. During acquisition, the sensors may be affected by chip splashes, leading to noise in the acquired data and affecting subsequent analysis. Therefore, in this embodiment, a filtering algorithm is used to preprocess the time-series acquired vibration and temperature data to eliminate external interference. The filtering algorithm is not limited to median filtering, wavelet filtering, or moving average filtering; this embodiment uses median filtering. Median filtering is a well-known technique and will not be elaborated upon here.

[0057] By collecting vibration and temperature data of the boring tool during precision boring in real time, data support is provided for the dynamic adjustment of subsequent boring tool cutting parameters.

[0058] Step S102: Perform window sliding on the vibration data time series. Based on the fluctuation degree and distribution pattern of the vibration data within the sliding window, obtain the time domain wear confidence of the sliding window; obtain the high frequency components of the vibration data within the sliding window, and combine the proportion of the high frequency components and the time domain wear confidence to obtain the vibration instability value.

[0059] During the precision boring of deep, elongated holes with stepped surfaces, due to the slender shape and large overhang of the boring bar, under reasonable cutting parameters and good tool conditions, the boring bar vibration is usually in a controlled and stable state. Its vibration signal exhibits characteristics of gradual amplitude changes, limited fluctuation range, and approximately stable distribution, without obvious sudden oscillations or high-frequency impact components. At this time, the vibration energy is mainly concentrated in the low-frequency range, reflecting normal cutting force fluctuations and the inherent response of the structure.

[0060] However, as the wear of the boring tool or uneven force gradually intensifies, the friction between the boring tool and the hole wall increases, and the vibration data will show an unstable fluctuation characteristic of first increasing and then decreasing in time sequence. Its dispersion and distribution pattern will change significantly. At the same time, the proportion of high-frequency vibration components will increase, reflecting the evolution of the cutting state from stable to unstable.

[0061] Therefore, by monitoring the vibration signal of the boring tool and dynamically dividing the vibration data time sequence using a sliding window method, the time-domain wear confidence of the corresponding window can be calculated by analyzing the fluctuation degree and distribution pattern of the vibration data within each sliding window. Furthermore, the high-frequency components of the vibration data within the sliding window are extracted by a filter, and combined with the proportion of the high-frequency components in the overall energy, they are fused with the time-domain wear confidence to obtain the vibration instability value used to characterize the stability of the current cutting state, thus providing a basis for subsequent tool state judgment and machining parameter adjustment.

[0062] In one specific implementation of this invention, when the cumulative sampling time of vibration data reaches 15 seconds, a sliding window with a length of 15 seconds is constructed on the time-series vibration data, and the sliding window is slid in 1-second increments so that each window analysis includes the vibration data collected in the most recent 15 seconds, which is used for subsequent calculation and analysis of vibration characteristics.

[0063] Step S103: The sampling time when the vibration instability value exceeds the preset vibration deviation range is taken as the vibration abnormality time; the sampling time when the temperature data exceeds the preset normal temperature range is taken as the temperature abnormality time.

[0064] During precision boring, vibration signals directly reflect the stress state and dynamic stability of the cutting tool. If the vibration instability value exceeds the normal fluctuation range, it indicates that the tool may have worn out. Meanwhile, abnormal temperature data is usually caused by increased friction or poor heat dissipation, indirectly reflecting energy accumulation after tool instability. Therefore, by separately recording the sampling times of vibration and temperature anomalies, a clear time reference can be provided for subsequent analysis. Furthermore, the two can complement each other, preventing false alarms or omissions from a single indicator, avoiding misjudging random fluctuations as systemic instability, and improving the accuracy of subsequent data analysis.

[0065] Step S104: Obtain the abnormal interval based on the time interval between the vibration abnormal moment and the temperature abnormal moment in the time sequence. Obtain the boring tool instability coefficient based on the proportion of vibration abnormal moments within the abnormal interval, the vibration instability value and the abnormal interval.

[0066] In real-time precision boring, when the boring tool experiences abnormal vibration due to wear or uneven cutting force, the vibration signal will deviate from the normal fluctuation state first. Temperature changes in the cutting area, influenced by heat conduction and accumulation, typically exhibit a significant hysteresis characteristic, only showing an abnormal rise after the abnormal vibration has persisted for a period. Therefore, in terms of timing, the vibration anomaly usually occurs earlier than the temperature anomaly, and the time difference between the two reflects the evolution of the boring tool's stability towards thermal anomaly.

[0067] Meanwhile, the greater the wear of the boring tool, the greater the friction it experiences. Under constant cutting speed, the faster the temperature rise rate in the cutting area, the shorter the time interval between vibration anomalies and temperature anomalies, and the higher the vibration instability value and the more frequent the occurrence of vibration anomalies within this time interval. By combining the proportion of vibration anomalies within the anomaly interval, the vibration instability value, and the anomaly interval, the boring tool instability coefficient is obtained. A larger boring tool instability coefficient indicates poorer boring tool stability under the current precision boring condition, requiring more timely adjustment of the tool's boring parameters.

[0068] In one specific implementation of this invention, an anomaly interval [y1, y2] is constructed based on the first occurrence of vibration anomaly and the first occurrence of temperature anomaly during the precision boring process. Here, y1 represents the first occurrence of vibration anomaly, and y2 represents the first occurrence of temperature anomaly. A smaller range of values ​​for the anomaly interval indicates a faster rate of tool temperature rise and a worse tool condition.

[0069] It should be noted that: the sampling time mentioned in this embodiment is the specific timestamp corresponding to the collection of vibration data and temperature data. The data collection frequency in this invention is 100Hz, which means that 100 vibration data and 100 temperature data are collected simultaneously per second. Therefore, there are 100 sampling times per second, and each abnormal time is determined based on the corresponding timestamp.

[0070] Step S105: Adjust the boring parameters based on the boring tool instability coefficient.

[0071] During boring, fixed basic parameters are typically set to achieve the machining process. However, due to dynamic factors such as tool wear and cutting force fluctuations, fixed parameters cannot respond to changes in the machining state in real time, easily leading to increased vibration and abnormal temperature rise, which in turn affects machining accuracy and surface quality. Therefore, it is necessary to dynamically adjust the boring parameters based on the real-time calculated boring tool instability coefficient, thereby actively suppressing vibration and stabilizing the cutting process in the early stages of abnormalities, ensuring the stability and consistency of machining quality.

[0072] Preferably, the method for obtaining the time-domain wear confidence score includes: determining the vibration discrete value based on the fluctuation degree of the vibration data within the sliding window; determining the vibration distribution value based on the distribution pattern of the vibration data within the sliding window; and obtaining the time-domain wear confidence score by combining the vibration discrete value and the vibration distribution value; wherein the vibration discrete value and the vibration distribution value are both positively correlated with the time-domain wear confidence score.

[0073] The larger the vibration dispersion value, the more uneven the dynamic load on the boring tool during the current cutting process, and the more likely that wear has occurred, resulting in decreased tool stability. The larger the vibration distribution value, the more unstable the distribution pattern, and the more violent the amplitude distribution fluctuation of the vibration data. This reflects the more likely there are unsteady characteristics such as intermittent impacts and sudden changes in friction state during the current cutting process, and further indicates that the tool is more likely to have worn out.

[0074] In one specific implementation of this invention, the coefficient of variation among all vibration data within the sliding window is denoted as the vibration discrete value of the sliding window to reflect the degree of fluctuation of the vibration data.

[0075] The sum of the absolute values ​​of kurtosis and skewness of the vibration data within the sliding window is recorded as the vibration distribution value of the sliding window, reflecting the distribution pattern of the vibration data.

[0076] It should be noted that in other possible implementations of the present invention, the mean difference coefficient and the root mean square error coefficient, which are dimensionless values, can be used to replace the coefficient of variation in order to calculate the discrete values ​​of vibration.

[0077] The time-domain wear confidence score is obtained by positively correlating the discrete vibration values ​​with the vibration distribution values. The higher the fluctuation of the vibration data within the sliding window and the more the distribution pattern deviates from the stable state, the greater the time-domain wear confidence score. This reflects that the boring tool is more likely to become unstable due to wear or uneven cutting force, and it is more necessary to adjust the current cutting parameters in a timely manner to suppress abnormal vibration and ensure the accuracy of the hole diameter.

[0078] In one specific implementation of this invention, taking the i-th sliding window as an example, the specific formula for the time-domain wear confidence of the i-th sliding window is expressed as follows:

[0079]

[0080] In the formula, Let be the time-domain wear confidence level for the i-th sliding window; The vibration discrete values ​​for the i-th sliding window; The vibration distribution value for the i-th sliding window; The normalization function is not limited to tanh, sigmoid, or exponential normalization functions. In this embodiment, tanh is used.

[0081] The time-domain wear confidence score reflects the fluctuation range of vibration amplitude within the sliding window through vibration discrete values, analyzing whether the boring tool is subjected to uniform force during the cutting process; it reflects the degree of severe fluctuation of vibration data through vibration distribution values, analyzing whether there are local impacts such as chipping or sudden friction changes.

[0082] Dimensional analysis: Vibration discrete values Here, is the coefficient of variation between vibration data, and is a numerical value; vibration distribution value. This is the sum of the absolute values ​​of kurtosis and skewness of the vibration data. Since both kurtosis and skewness are dimensionless values, It is also a numerical value. Therefore, it can be directly applied to... , Normalization is performed to avoid the problem of one parameter being too large, causing other parameters to have a smaller impact. Finally, the result is... It is also a numerical value, without physical dimensions.

[0083] The above formula uses addition to construct a positive correlation. In other implementations of this invention, multiplication, scaling, and weighted summation methods can also be used to construct a positive correlation, which will not be elaborated or limited here.

[0084] Preferably, the method for obtaining the high-frequency components includes: converting the vibration data within the reference window into the frequency domain to obtain the dominant frequency of the reference window; the reference window is a sliding window on the time sequence of vibration data corresponding to the processing process with the highest processing accuracy in the historical processing process; determining the cutoff frequency of the filter based on the dominant frequency of the reference window; and obtaining the high-frequency components of the sliding window based on the cutoff frequency of the filter.

[0085] During precision boring, when the boring bar is in normal cutting mode, its vibration energy is mainly distributed around the structural natural frequency and the main cutting frequency. When the tool wears or the cutting state becomes unstable, due to increased friction and changes in contact state, high-frequency components caused by impact and intermittent contact gradually appear in the vibration signal. Based on this, by extracting the high-frequency components from the vibration data, the changes in tool stability can be reflected more intuitively, thus providing stronger data support for judging the tool condition and adjusting machining parameters during precision boring.

[0086] Furthermore, since surface roughness can reflect the machining accuracy of a workpiece, the machining process with the highest machining accuracy can be selected from historical machining processes based on the surface roughness index, and this process can be used as a reference to analyze the current tool condition of the boring tool.

[0087] In one specific implementation of this invention, the historical machining process with the lowest surface roughness of the workpiece after boring is recorded as the reference process. Following the data acquisition method described above, vibration data of the boring tool during the fine boring process of the reference process is acquired, and a reference vibration sequence is constructed according to the chronological order. Then, following the aforementioned sliding window construction method and sliding step size, all sliding windows within the reference vibration sequence are acquired and recorded as reference windows.

[0088] The vibration data within each reference window are sequentially converted into frequency domain data to obtain the corresponding spectral data for each reference window. Any reference window is designated as the detection window, and the frequency corresponding to the peak value with the largest amplitude in the spectral data of the detection window is taken as the dominant frequency of the detection window. The dominant frequencies of all reference windows are obtained in the same manner.

[0089] The methods for converting time-domain data into frequency-domain data are not limited to Fourier transform and fast Fourier transform; this embodiment uses fast Fourier transform.

[0090] Taking the u-th main frequency as an example, the main frequency band range of the u-th main frequency is constructed [ ].in, Let be the frequency of the u-th main frequency, and 'a' be the degree of dispersion among the main frequencies of all reference windows, used to reflect the natural fluctuation range of the main frequency under good processing conditions. In this embodiment, the coefficient of variation is used to calculate the degree of dispersion.

[0091] The main frequency band ranges of all main frequencies are obtained in the same way, and the maximum value among all main frequency band ranges is used as the cutoff frequency of the low-pass filter. It should be noted that the cutoff frequency is only calibrated offline during the initialization phase and is not involved in online real-time calculation.

[0092] Taking the i-th sliding window as an example, the time-domain vibration data within the i-th sliding window is low-pass filtered by a low-pass filter to obtain the low-frequency components of the vibration data in the i-th sliding window.

[0093] Subsequently, the low-frequency component value of the low-pass filter output at the corresponding moment is subtracted from the value of each sampling point in the original vibration data of the i-th sliding window to obtain the high-frequency component.

[0094] High-frequency components are used to characterize the rapid fluctuation components in vibration signals relative to normal cutting conditions. Their amplitude changes can reflect whether the tool has abnormal vibration characteristics caused by wear, increased friction, or intermittent contact in the current machining stage.

[0095] Preferably, the method for obtaining the vibration instability value includes: taking any sliding window as the target window, calculating the proportion of the high-frequency component of the target window in the total energy of the target window, and obtaining the high-frequency energy proportion of the window; combining the time-domain wear confidence and the high-frequency energy proportion to obtain the vibration instability value; wherein, the time-domain wear confidence and the high-frequency energy proportion are both positively correlated with the vibration instability value.

[0096] Because relying solely on the time-domain characteristics of vibration data is insufficient to effectively distinguish between genuine anomalies caused by tool wear and false anomalies caused by non-wear factors such as sensor loosening, environmental coupling vibration, or filtering distortion, there is a risk of misjudgment. Furthermore, in actual machining, once tool wear occurs, the collected vibration data exhibits a decrease in the proportion of low-frequency energy and an increase in the energy of high-frequency components in the frequency domain. Therefore, combining the time-domain and frequency-domain characteristics of vibration data can comprehensively construct a vibration instability value, thereby improving the accuracy of tool wear identification. A larger vibration instability value indicates a greater deviation of the current vibration state from normal cutting conditions, and a higher probability of tool wear or cutting instability.

[0097] In one specific implementation of this invention, any sliding window is used as the target window. The sum of squares of the high-frequency components and the original vibration data within the target window is calculated, and the ratio of the sum of squares of the high-frequency components to the sum of squares of the original vibration data is taken as the high-frequency energy proportion of the target window. The larger the high-frequency energy proportion, the higher the concentration of vibration energy in the high-frequency components, reflecting that the impact and friction fluctuations experienced by the boring tool during the current cutting process are more obvious, and the greater the possibility of a decrease in tool stability.

[0098] Taking the i-th sliding window as an example, the specific formula for the vibration instability value is expressed as follows:

[0099]

[0100] In the formula, Let be the vibration instability value of the i-th sliding window; Let be the time-domain wear confidence level for the i-th sliding window; Let be the proportion of high-frequency energy in the i-th sliding window.

[0101] The vibration instability value reflects the degree of abnormal fluctuation of the vibration signal in the time domain and its distribution pattern change characteristics through the time domain wear confidence, and reflects the degree of concentration of vibration energy in the high frequency domain through the high frequency energy ratio. Thus, it comprehensively characterizes the stability change of the boring tool in the precision boring process from both time and frequency domain perspectives.

[0102] Dimensional analysis: Time-domain wear confidence level The value is a numerical value with a range of [0,2]; the proportion of high-frequency energy is also a numerical value with a range of (0,1); therefore, the calculated vibration instability value is also a numerical value with no physical dimension.

[0103] The above formula uses multiplication to construct a positive correlation. In other implementations of this invention, addition, scaling, and weighted summation methods can also be used to construct a positive correlation, which will not be elaborated or limited here.

[0104] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the moment of vibration abnormality is described in [reference needed]. Figure 2 The diagram illustrates a flowchart of a method for obtaining abnormal vibration moments according to an embodiment of the present invention, the method comprising:

[0105] In step S201, the vibration instability value of the sliding window is evaluated in real time to see if it is within the preset vibration deviation range corresponding to the sliding window.

[0106] In this step, the preset vibration deviation range corresponding to the sliding window is obtained.

[0107] In some possible implementations of this invention, all historical sliding windows can be clustered based on the vibration instability value of each historical sliding window to obtain multiple clusters; the historical sliding window is a sliding window on the time sequence of historical vibration data during multiple historical normal processing processes; then, based on the difference between the mean of the elements of each cluster and the vibration instability value of the real-time sliding window, the cluster to which the real-time sliding window belongs is obtained; finally, the maximum and minimum values ​​of the vibration instability values ​​within the cluster to which the real-time sliding window belongs are used as the upper and lower limits of the preset vibration deviation value range corresponding to the real-time sliding window.

[0108] In step S202, when the vibration instability value of the sliding window exceeds the corresponding preset vibration deviation range, the sampling time corresponding to the first vibration data in the sliding window is recorded as the vibration abnormality time.

[0109] During precision boring, abnormal tool vibration is a continuous process, not an instantaneous event. Calculating vibration instability values ​​using a sliding window essentially evaluates the overall vibration characteristics over a given time period. When the vibration instability value within the sliding window exceeds the corresponding vibration deviation range, it indicates that the tool vibration state has deviated from the normal pattern for the entire time period covered by the sliding window.

[0110] In one specific implementation of this invention, taking the i-th sliding window as an example, when the vibration instability value of the i-th sliding window exceeds the corresponding vibration deviation range, it indicates that the boring tool has experienced abnormal vibration. At this time, the data sampling time corresponding to the first vibration data point within the i-th sliding window is recorded as the vibration abnormality time.

[0111] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the preset vibration deviation range corresponding to the sliding window is described in [reference needed]. Figure 3 The document presents a flowchart of a method for obtaining a preset vibration deviation range corresponding to a sliding window, as provided in an embodiment of the present invention. The method includes:

[0112] In step S2011, based on the vibration instability value of each historical sliding window, all historical sliding windows are clustered to obtain multiple clusters; the historical sliding window is a sliding window on the time sequence of historical vibration data during multiple historical normal processing processes.

[0113] Since the fluctuation characteristics of vibration signals during boring directly reflect the stability of the machining process, and because the tool condition and the material of the workpiece can change, even in a normal finish boring process, the collected tool vibration data may differ. Therefore, it is first necessary to distinguish between multiple sets of historical vibration sequences.

[0114] In one specific implementation of this invention, vibration data from 100 historical normal processing operations are collected according to the above data acquisition method, and 100 sets of historical vibration sequences are constructed in chronological order.

[0115] The variance of the historical vibration data for each group is calculated and sorted according to the acquisition time sequence of the historical data to obtain a sequence A composed of variances. Since the reasonable range of the boring basic data is small, it is necessary to make a detailed judgment on the reasonable range of the vibration data fluctuation. In this embodiment, the quartile principle is used to divide the sequence A. According to the division result, the historical normal vibration data corresponding to each variance in the sequence A can be divided into four intervals, thereby realizing the coarse division of the historical vibration data.

[0116] Since the fluctuation characteristics of vibration signals during boring directly reflect the stability of the machining state, and different fluctuation levels correspond to different working conditions, the analysis window needs to be set differently according to the actual machining scenario. In the high fluctuation range, severe vibration often indicates abnormal working conditions such as accelerated tool wear, sudden changes in cutting force, or loose fixtures. In this case, a shorter window should be used to capture transient characteristics. In the low fluctuation range, stable vibration usually corresponds to normal cutting conditions, and a longer window can be used to improve analysis efficiency and reduce computational redundancy.

[0117] Based on this, this embodiment uses the quartile partitioning results and the actual sampling frequency to convert the time window into the corresponding number of sampling points for data partitioning.

[0118] Specifically: a 15-second sampling period is used as the window length to divide the historical vibration data within the first quartile interval; the second quartile interval is divided using a 30-second sampling period; the third quartile interval uses a 45-second sampling period; and the most stable fourth quartile interval is divided using a 60-second sampling period. Each sliding window resulting from this historical vibration data division is recorded as a historical sliding window.

[0119] Following the calculation steps for the vibration instability value of the sliding window described above, the vibration instability value of each historical sliding window is calculated. These values ​​are then used as a distance metric. The absolute value of the difference between the vibration instability values ​​of two historical sliding windows is taken as the distance between them. The historical sliding windows are then clustered to obtain multiple clusters. The clustering algorithm is not limited to k-means or HDBSCAN. This embodiment uses the k-means algorithm for clustering and obtains the optimal number of clusters using the elbow rule. It should be noted that the absolute value of the difference is the absolute value of the difference.

[0120] It should be noted that if the data length of the sliding window after partitioning cannot meet the number of sampling points required for the corresponding window, the mean of the existing data in the sliding window will be used to fill the missing part to ensure that the data length of the sliding window after partitioning the same set of historical data is consistent, which facilitates feature calculation and cluster analysis.

[0121] In step S2012, the cluster to which the real-time sliding window belongs is obtained based on the difference between the element mean of each cluster and the vibration instability value of the real-time sliding window.

[0122] In one specific implementation of this invention, the i-th sliding window currently being acquired in real time is still taken as an example. The absolute difference between the vibration instability value of the i-th real-time sliding window and the average vibration instability value of all sliding windows in each cluster is calculated sequentially, and the cluster corresponding to the minimum absolute difference is denoted as the cluster to which the i-th real-time sliding window belongs.

[0123] In step S2013, the maximum and minimum values ​​of vibration instability within the cluster to which the real-time sliding window belongs are used as the upper and lower limits of the preset vibration deviation range corresponding to the real-time sliding window.

[0124] In one specific implementation of this invention, the maximum and minimum values ​​of the vibration instability values ​​within the cluster to which the i-th real-time sliding window belongs are used as the upper and lower limits of the preset vibration deviation range corresponding to the i-th real-time sliding window. This obtains a dynamic judgment benchmark applicable to the current working condition, which is then used for subsequent anomaly identification and parameter adjustment. It should be noted that the vibration instability values ​​within the cluster are the vibration instability values ​​corresponding to all sliding windows within the cluster.

[0125] Preferably, the method for obtaining the preset normal temperature range includes: taking the maximum and minimum values ​​from historical temperature data during multiple historical normal processing processes as the upper and lower limits of the preset normal temperature range.

[0126] By presetting the normal temperature range, the temperature fluctuation range of the boring tool under normal machining conditions can be effectively characterized. This is beneficial for accurately identifying temperature anomalies caused by tool wear or abnormal cutting during subsequent real-time machining, improving the reliability of temperature anomaly judgment, and avoiding misjudgments caused by occasional fluctuations.

[0127] In one specific implementation of this invention, the historical temperature data of the boring tool during 100 historical normal machining operations are collected according to the above data acquisition method, and the maximum and minimum values ​​among all historical temperature data are used as the upper and lower limits of the preset normal temperature range during fine boring.

[0128] Preferably, the method for obtaining the boring tool instability coefficient includes: combining the proportion of the vibration abnormal moments and the vibration instability value to obtain comprehensive abnormal characteristics; determining the boring tool instability coefficient based on the abnormal interval and the comprehensive abnormal characteristics; the abnormal interval is negatively correlated with the boring tool instability coefficient, and the comprehensive abnormal characteristics are positively correlated with the boring tool instability coefficient.

[0129] The larger the overall anomaly characteristic, the more frequent the vibration data anomalies and the greater the fluctuation in the vibration data, indicating that the boring tool is more likely to have experienced wear failure. The smaller the range of the anomaly interval, the faster the temperature rises, and thus the more severe the wear of the boring tool may be.

[0130] Furthermore, since an abnormal interval can only be constructed when an abnormal temperature event is detected, thus enabling the analysis of the boring tool instability coefficient, in a specific implementation of this invention, taking the abnormal temperature event y2 as an example, the ratio of the number of occurrences of abnormal vibration events within the abnormal interval [y1, y2] to the total number of sliding windows within the abnormal interval is denoted as the proportion of abnormal vibration events. The mean value of the vibration instability of all sliding windows within the abnormal interval [y1, y2] is calculated and denoted as the instability mean.

[0131] The proportion of abnormal vibration moments is summed with the mean of instability, and the summation result is recorded as the comprehensive abnormal characteristic of the abnormal interval to which the temperature abnormal moment y2 belongs.

[0132] The boring tool instability coefficient at the temperature anomaly moment y2 is specifically expressed by the following formula:

[0133]

[0134] In the formula, This represents the boring tool instability coefficient at the time of temperature anomaly y2 during the precision boring process; The comprehensive abnormal characteristics of the abnormal interval to which temperature abnormality time y2 belongs; y1 is the first occurrence of vibration abnormality; y2 is the first occurrence of temperature abnormality; The timestamp length corresponding to one unit of time (second) is calculated in this embodiment by measuring the absolute difference between the timestamps corresponding to the first and last vibration data within one second in the time series; norm() is the normalization function, which is the tanh normalization function in this embodiment.

[0135] The boring tool instability coefficient, through comprehensive anomaly characteristics, can reflect the overall intensity and concentration of abnormal vibration within the abnormal interval; through... It can reflect the time scale of the evolution from vibration anomaly to temperature anomaly, thus comprehensively characterizing the degree of instability of the boring tool during the precision boring process.

[0136] Dimensional analysis: Comprehensive anomaly characteristics The sum of the proportion of vibration anomaly moments and the mean instability value is given. The proportion of vibration anomaly moments is a numerical value with a range of (0,1]. The mean instability value is the average of the vibration instability values ​​across all sliding windows, also a numerical value with a range of (0,2). Therefore, the comprehensive anomaly characteristic can be directly calculated by summing these values, and the comprehensive anomaly characteristic is also a numerical value. y1 represents the vibration anomaly moment, and y2 represents the temperature anomaly moment, and y2 represents the temperature anomaly moment, and y2 represents the temperature anomaly moment. The difference between the timestamps; This is the difference in timestamps corresponding to one unit of time (seconds). and The ratio is a numerical value, dimensionless, and can be directly normalized. Therefore, the final boring tool instability coefficient is also a numerical value, dimensionless.

[0137] Preferably, adjusting the boring parameters based on the boring tool instability coefficient includes: weighting the initial boring parameters using the boring tool instability coefficient to obtain a boring parameter adjustment value; and adjusting the initial boring parameters based on the boring parameter adjustment value.

[0138] In one specific implementation of this invention, the boring parameters that need to be adjusted for the boring tool at the abnormal temperature moment y2 include: cutting speed. Feed rate and depth of cut .

[0139] The boring parameters at the t-th time of the temperature anomaly y2 For example, for the t-th boring parameter The specific formula for adjustment is expressed as follows:

[0140]

[0141] In the formula, This represents the adjusted value of the t-th boring parameter. This represents the lower limit of the t-th boring parameter. This represents the boring parameter value before adjustment at the time of temperature anomaly y2, where the t-th boring parameter is located. This represents the adjustment factor, which prevents the boring tool instability coefficient from being too large, thus causing unreasonable boring parameter adjustments. It takes a value between [0.1, 0.5], and in this embodiment it is 0.3. This represents the boring tool instability coefficient at the time of temperature anomaly y2 during the precision boring process. `max()` is the function to find the maximum value.

[0142] In this embodiment, the lower limit of the cutting speed The lower limit of the feed rate is 60 m / min. The lower limit of the single cutting depth is 0.1 mm / r. The initial value is 0mm. The initial value of the cutting speed is 100m / min, the initial value of the feed rate is 0.2mm / r, and the initial value of the depth of cut is 0.3mm.

[0143] It should be noted that if the temperature anomaly at time y2 is the first adjustment of the boring tool's boring parameters, then the tool boring parameters at time y2 will be consistent with the initial values, i.e., the cutting speed... 100m / min, feed rate 0.2mm / r, single cutting depth It is 0.3mm.

[0144] In the above formula, optimizing the boring parameters by taking the maximum value ensures that the output result is not lower than the safety threshold, simultaneously meeting the requirements of optimization objectives and operational constraints, and guaranteeing the continuity and machining safety of the precision boring process. After the boring parameters are adjusted, the tool instability coefficient is recorded at this time; then, following the data acquisition and analysis steps described above, the tool wear state is analyzed and adjusted in real time.

[0145] During the overall data acquisition and boring parameter adjustment process, the following judgment mechanism is followed to ensure machining safety:

[0146] 1. Within the first 15 seconds of boring, check point by point whether the vibration data collected at each sampling time is within the preset normal vibration range. If the vibration data at sampling time j is within the preset normal vibration range, it is considered that the boring tool has not experienced abnormal vibration at sampling time j; otherwise, record the number of vibration data that do not meet the condition. If the number of vibration data that do not meet the condition within 1 second exceeds half of the sampling quantity within 1 second, issue a warning directly and stop machining to ensure safety.

[0147] In one specific implementation of this invention, the maximum and minimum values ​​of the historical vibration data of the boring tool during 100 historical normal machining operations are used as the upper and lower limits of the preset normal vibration value range during fine boring.

[0148] 2. If the boring tool instability coefficient can be calculated for 300 consecutive seconds, that is, if the temperature abnormality can be obtained for 300 consecutive seconds, a warning will be issued: an abnormality has occurred during the precision boring process. At this time, the processing should be stopped to ensure safety.

[0149] 3. If, after adjusting the instability coefficient of the boring tool, the following occurs... If the calculation result is accurate, the count is 1. If the count exceeds 5, a warning is issued and processing is stopped.

[0150] 4. When the single cutting depth is adjusted to the lower limit of 0, a warning will be issued directly, prompting manual inspection; if no problems are found after inspection by on-site workers, the single cutting depth will be adjusted back to the initial data.

[0151] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0152] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for boring deep, elongated holes with an adjustable precision boring tool, characterized in that, The method includes: Real-time acquisition of vibration and temperature data of the boring tool; A sliding window is performed on the vibration data time series. Based on the fluctuation degree and distribution pattern of the vibration data within the sliding window, the time-domain wear confidence of the sliding window is obtained. The high-frequency components of the vibration data within the sliding window are obtained. Combining the proportion of the high-frequency components and the time-domain wear confidence, the vibration instability value is obtained. The sampling time when the vibration instability value exceeds the preset vibration deviation range is taken as the vibration abnormality time; the sampling time when the temperature data exceeds the preset normal temperature range is taken as the temperature abnormality time. The abnormal interval is obtained by the time interval between the vibration abnormal moment and the temperature abnormal moment in time sequence. The boring tool instability coefficient is obtained by the proportion of vibration abnormal moments within the abnormal interval, the vibration instability value and the abnormal interval. Adjusting boring parameters based on the boring tool instability coefficient; The method for obtaining the time-domain wear confidence includes: The vibration discrete value is determined based on the degree of fluctuation of the vibration data within the sliding window; The vibration distribution value is determined based on the distribution pattern of the vibration data within the sliding window; By combining the vibration discrete value and the vibration distribution value, the time-domain wear confidence level is obtained; wherein, both the vibration discrete value and the vibration distribution value are positively correlated with the time-domain wear confidence level. The method for obtaining the high-frequency components includes: The vibration data within the reference window is converted into the frequency domain to obtain the dominant frequency of the reference window; the reference window is a sliding window on the time sequence of vibration data corresponding to the processing process with the highest processing accuracy in the historical processing process. The cutoff frequency of the filter is determined based on the dominant frequency of the reference window. Based on the filter's cutoff frequency, the high-frequency components of the sliding window are obtained; The method for obtaining the vibration instability value includes: Using any sliding window as the target window, calculate the proportion of the high-frequency components of the target window in the total energy of the target window, and obtain the proportion of high-frequency energy of the window. By combining the time-domain wear confidence level and the high-frequency energy ratio, the vibration instability value is obtained; wherein, both the time-domain wear confidence level and the high-frequency energy ratio are positively correlated with the vibration instability value.

2. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 1, characterized in that, The vibration distribution value is the sum of the absolute values ​​of kurtosis and skewness of the vibration data within the sliding window.

3. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 1, characterized in that, The method for obtaining the moment of vibration abnormality includes: Real-time assessment of whether the vibration instability value of the sliding window is within the preset vibration deviation range corresponding to the sliding window; When the vibration instability value of the sliding window exceeds the corresponding preset vibration deviation range, the sampling time corresponding to the first vibration data in the sliding window is recorded as the vibration abnormality time.

4. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 3, characterized in that, The method for obtaining the preset vibration deviation range corresponding to the sliding window includes: Based on the vibration instability value of each historical sliding window, all historical sliding windows are clustered to obtain multiple clusters; the historical sliding window is a sliding window on the time sequence of historical vibration data during multiple historical normal processing processes; Based on the difference between the mean of elements in each cluster and the vibration instability value of the real-time sliding window, the cluster to which the real-time sliding window belongs is obtained. The maximum and minimum values ​​of vibration instability within the cluster to which the real-time sliding window belongs are used as the upper and lower limits of the preset vibration deviation range corresponding to the real-time sliding window.

5. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 1, characterized in that, The method for obtaining the preset normal temperature range includes: The maximum and minimum values ​​from historical temperature data during multiple normal processing processes are used as the upper and lower limits of the preset normal temperature range.

6. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 1, characterized in that, The method for obtaining the instability coefficient of the boring tool includes: By combining the proportion of abnormal vibration moments and the vibration instability value, a comprehensive abnormality characteristic is obtained; The boring tool instability coefficient is determined based on the abnormal interval and the comprehensive abnormal characteristics. The abnormal interval is negatively correlated with the boring tool instability coefficient, while the comprehensive abnormal characteristics are positively correlated with the boring tool instability coefficient.

7. The method for boring deep elongated holes with adjustable function using a precision boring tool according to claim 1, characterized in that, The adjustment of boring parameters based on the boring tool instability coefficient includes: The initial boring parameters are weighted by the boring tool instability coefficient to obtain the boring parameter adjustment value; Based on the aforementioned boring parameter adjustment value, the initial boring parameters are adjusted.

Images