Automatic processing method and device for laser gyroscope cavity based on ultrasonic milling

Abstract

The present application relates to the field of automatic processing technology, and particularly relates to a laser gyroscope cavity automatic processing method and device based on ultrasonic milling and grinding. The method aims at the problems of heat dissipation failure caused by slurry accumulation in the processing of microcrystalline glass deep hole, easy to cause thermal damage and unable to directly measure temperature. Through real-time acquisition of cutting power, feed speed and resonance frequency data, the slurry accumulation ratio is determined by analyzing the resonance frequency fluctuation degree, and the heat dissipation capacity characteristic coefficient is estimated accordingly; the friction heat generation weight is extracted in combination with the correlation of cutting power and resonance frequency, and the overload friction power is calculated based on the feed speed and power deviation to generate a heat accumulation characteristic value; finally, the heat accumulation characteristic value is taken as the heat source, and the heat dissipation capacity characteristic coefficient is taken as the heat dissipation term to construct a recursive processing heat accumulation index model, and the tool is triggered to retreat and chip when the index exceeds the threshold. The method significantly improves the automation level, efficiency and reliability of processing.

Description

Technical Field

[0001] This invention relates to the field of automated machining technology, specifically to an automated machining method and apparatus for laser gyroscope cavities based on ultrasonic milling. Background Technology

[0002] As a core component of high-precision inertial navigation systems, laser gyroscopes typically use microcrystalline glass, a material with an extremely low coefficient of thermal expansion, to manufacture their cavities. Deep hole machining of microcrystalline glass presents unique technological challenges: the material is highly hard and extremely brittle, while also having a very low thermal conductivity (approaching that of an insulator). This means that the heat generated during machining is difficult to dissipate through conduction within the workpiece itself and must primarily rely on forced convection of the cutting fluid to remove it.

[0003] In deep hole machining (typically referring to holes with a depth-to-diameter ratio greater than 3), micron-sized glass-ceramic abrasive chips readily mix with cutting fluid to form a high-viscosity slurry. This slurry often accumulates in narrow chip removal channels. Without accurate temperature feedback, existing machining control methods primarily rely on single power threshold monitoring or fixed-cycle pecking drills. This approach makes automated machining of laser gyroscope cavities impossible, resulting in poor machining efficiency and reliability. Summary of the Invention

[0004] To address the technical problem of poor processing efficiency and reliability in the automated processing of laser gyroscope cavities in related technologies, this invention provides an automated processing method and apparatus for laser gyroscope cavities based on ultrasonic milling. The specific technical solution adopted is as follows:

[0005] This invention proposes an automated machining method for laser gyroscope cavities based on ultrasonic milling, the method comprising:

[0006] The cutting power, feed rate and resonant frequency data of the tool at different sampling times in the down-hole calibration interval and the deep hole interval are obtained. The hole depth in the down-hole calibration interval is smaller than that in the deep hole interval. The current sampling time is in the feeding process, and the feeding process is in the deep hole interval.

[0007] Within a preset time window prior to the current sampling time, the slurry accumulation ratio is determined based on the degree of fluctuation in the resonant frequency data; based on the slurry accumulation ratio, the heat dissipation capacity characterization coefficient of the bottom of the hole through fluid convection per unit time is estimated using the attenuation function.

[0008] Within a preset time window, correlation analysis is performed based on cutting power and resonant frequency data to determine the weight of frictional heat generation; based on feed rate and cutting power, the overload frictional power exceeding normal cutting requirements at the current sampling moment is determined; and by combining the overload frictional power and the weight of frictional heat generation, the characteristic value of heat accumulation is determined.

[0009] Based on the characteristic value of heat accumulation and the characterization coefficient of heat dissipation capacity, a thermal risk analysis is performed at the current sampling time to obtain the processing heat accumulation index at the current sampling time. Based on the processing heat accumulation index, the processing tool retraction control is performed.

[0010] Furthermore, determining the slurry accumulation ratio based on the degree of fluctuation in the resonant frequency data includes:

[0011] Calculate the numerical standard deviation of the resonant frequency data at all sampling times within the downhole calibration interval, and use it as a benchmark for the degree of fluctuation.

[0012] Calculate the standard deviation of the resonant frequency data at all sampling times within the preset time window, as the degree of window fluctuation;

[0013] The window fluctuation degree is numerically compared with the baseline fluctuation degree to determine the slurry accumulation ratio, which is not less than 1.

[0014] Furthermore, the heat dissipation capacity characterization coefficient based on the slurry accumulation ratio and using an attenuation function to estimate the heat loss from the bottom of the borehole through fluid convection per unit time includes:

[0015] The difference between the slurry accumulation ratio and the unit value of 1 is calculated as the blockage coefficient;

[0016] Based on the preset baseline cooling rate and preset blockage sensitivity, the blockage coefficient is analyzed to obtain the heat dissipation capacity characterization coefficient. Specifically, the product of the preset blockage sensitivity and the blockage coefficient is calculated, and the sum of the unit value 1 and the product value is used as the denominator, while the preset baseline cooling rate is used as the numerator. The fraction is then used to calculate the heat dissipation capacity characterization coefficient.

[0017] Furthermore, the correlation analysis based on cutting power and resonant frequency data to determine the weight of frictional heat generation includes:

[0018] Pearson correlation calculations were performed on the cutting power and resonant frequency data within a preset time window to obtain the Pearson correlation coefficient.

[0019] When the Pearson correlation coefficient is less than the preset correlation threshold, the frictional heat generation weight is set to 0. When the Pearson correlation coefficient drifts in the positive direction beyond the preset correlation threshold, it is determined that there is abnormal friction. The Pearson correlation coefficient is normalized and mapped to the value range of [0,1] to obtain the frictional heat generation weight.

[0020] Furthermore, determining the overload friction power exceeding normal cutting requirements at the current sampling moment based on feed rate and cutting power includes:

[0021] In the down-the-hole calibration range, the standard energy consumption per unit depth is calculated based on the total energy consumed in the range and the total effective cutting stroke.

[0022] Calculate the product of the feed rate and the standard energy consumption per unit depth at the current sampling moment, and use it as the ideal power;

[0023] The total power of the tool during the feeding process is obtained under the sampling period, as well as the no-load power of the tool in the initial stage of idling before it contacts the workpiece. The overload friction power is obtained by combining the difference between the total power, the ideal power, and the no-load power. When the difference is less than 0, the overload friction power is set to 0.

[0024] Furthermore, the calculation of the standard energy consumption per unit depth based on the total energy consumed in the interval and the total effective cutting stroke includes:

[0025] The ratio of total energy consumption to effective cutting stroke is calculated and used as the standard energy consumption per unit depth.

[0026] Furthermore, the determination of the heat accumulation characteristic value by combining the overload frictional power and the frictional heat generation weight includes:

[0027] The frictional heat generation weights are weighted and corrected based on a preset frictional gain constant to obtain the frictional characteristic coefficients.

[0028] The product of the friction characteristic coefficient and the overload friction power is calculated as the thermal accumulation characteristic value.

[0029] Furthermore, the thermal risk analysis performed on the current sampling moment based on the thermal accumulation characteristic value and the heat dissipation capacity characterization coefficient to obtain the processing thermal accumulation index at the current sampling moment includes:

[0030] The attenuation coefficient is determined based on the heat dissipation capacity characterization coefficient, where the larger the value of the heat dissipation capacity characterization coefficient, the smaller the value of the attenuation coefficient.

[0031] The product of the thermal accumulation characteristic value and the sampling period duration is used as the thermal energy index of ineffective friction;

[0032] Set the processing heat accumulation index to 0 at the first sampling moment during the feeding process;

[0033] Calculate the product of the processing heat accumulation index and the attenuation coefficient at the previous sampling time, and use it as the residual risk energy index at the previous sampling time.

[0034] The sum of the residual risk energy index at the previous sampling time and the thermal energy index at the corresponding sampling time is used as the thermal energy index at the corresponding sampling time. This process is repeated until the processing heat accumulation index at the current sampling time is obtained.

[0035] Furthermore, the machining retraction control based on the machining heat accumulation index includes:

[0036] When the processing heat accumulation index is greater than or equal to a preset risk threshold, it is determined that the microcrystalline glass processing has a risk of thermal damage, and a tool retraction and chip removal command is generated to achieve tool retraction control.

[0037] On the other hand, it also includes an automated machining device for laser gyroscope cavities based on ultrasonic milling, the device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method as described in any of the foregoing.

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

[0039] This invention effectively solves the technical problem of uncontrollable thermal damage caused by slurry accumulation in deep hole machining of microcrystalline glass by constructing a dynamic thermal risk assessment system that integrates cooling state perception and decoupling of frictional heat sources. The slurry accumulation ratio is quantified using the degree of resonant frequency fluctuation, and a heat dissipation capacity characterization coefficient is estimated accordingly, enabling online characterization of the degradation of convection heat dissipation capacity at the bottom of the hole. Simultaneously, the correlation analysis between cutting power and resonant frequency extracts the weight of frictional heat generation, and the overload frictional power is determined by combining feed rate and power deviation, thereby generating a heat accumulation characteristic value that only reflects the contribution of harmful friction. Based on this, a recursive machining heat accumulation index model is established, using the heat accumulation characteristic value as the heat source and the heat dissipation capacity characterization coefficient as the heat dissipation term, to achieve cumulative quantification of the unmeasurable heat accumulation process at the bottom of the hole. This method abandons the traditional extensive control strategy that relies on fixed power thresholds or periodic pecking drills, and can accurately identify real thermal risks, triggering tool retraction and chip removal when necessary. This significantly improves automated machining efficiency and process reliability while ensuring the machining quality of the cavity without microcracks. 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 The flowchart illustrates an automated machining method for laser gyroscope cavities based on ultrasonic milling, as provided in one embodiment of the present invention. Detailed Implementation

[0042] 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 an automated processing method and apparatus for laser gyroscope cavities based on ultrasonic milling 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.

[0043] 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.

[0044] As a core component of high-precision inertial navigation systems, laser gyroscopes typically use microcrystalline glass, a material with an extremely low coefficient of thermal expansion, to manufacture their cavities. Deep hole machining of microcrystalline glass presents unique technological challenges: the material is highly hard and extremely brittle, while also having a very low thermal conductivity (approaching that of an insulator). This means that the heat generated during machining is difficult to dissipate through conduction within the workpiece itself and must primarily rely on forced convection of the cutting fluid to remove it.

[0045] In deep hole machining (typically referring to holes with a depth-to-diameter ratio greater than 3), micron-sized glass-ceramic abrasive chips readily mix with cutting fluid to form a high-viscosity slurry. This slurry often accumulates in narrow chip removal channels. Without accurate temperature feedback, existing machining control methods primarily rely on single power threshold monitoring or fixed-cycle pecking drills. This approach makes automated machining of laser gyroscope cavities impossible, resulting in poor machining efficiency and reliability.

[0046] To address the aforementioned technical problems, the following detailed description, in conjunction with the accompanying drawings, illustrates a specific solution for an automated machining method for laser gyroscope cavities based on ultrasonic milling, provided by this invention.

[0047] Please see Figure 1 The diagram illustrates a flowchart of an automated machining method for a laser gyroscope cavity based on ultrasonic milling, according to an embodiment of the present invention. The method includes:

[0048] S101: Obtain the cutting power, feed rate and resonant frequency data of the tool at different sampling times in the down-hole calibration interval and the deep hole interval. The hole depth in the down-hole calibration interval is smaller than that in the deep hole interval. The current sampling time is during the feed process, and the feed process is in the deep hole interval.

[0049] Before performing deep hole machining on the laser gyroscope cavity (microcrystalline glass), it is necessary to first establish a machining benchmark that is compatible with the current "machine tool-tool-workpiece" system. Since microcrystalline glass is a typical hard and brittle material, different batches of blanks may have differences in microscopic hardness, and diamond grinding rods will inevitably wear during machining. This means that fixed process parameters cannot adapt to dynamically changing machining conditions. Therefore, dynamic machining adjustments are required.

[0050] Tool cutting mainly includes a shallow hole calibration range and a deep hole range. In the shallow hole calibration range, such as the 0.5mm-2.5mm range, the selection of this range should meet two conditions: First, the maximum depth should be less than the critical deep hole limit (usually three times the hole diameter) to ensure that the external cutting fluid nozzle can directly inject coolant into the cutting area, the chip removal channel is unobstructed, and there is no thermal resistance caused by slurry accumulation; Second, this range is located in the stable wear stage of the tool life cycle. If a new tool is used, a brief pre-break-in cutting can be performed first. In the embodiment of this invention, the feed process is mainly in the deep hole range, such as the depth range after 2.5mm. When in the shallow hole calibration range, the machining heat accumulation index is forcibly set to 0 to prevent accidental tool retraction due to unstable data at the initial stage of system startup.

[0051] The system can sample every 10 milliseconds to obtain a sampling time. During the initialization phase, before the tool contacts the workpiece, the system controls the spindle to start rotating and activates the ultrasonic generator, causing it to operate at a preset machining speed (e.g., The system remains idle at a preset ultrasonic amplitude (e.g., 0.005 mm). The system operates within a preset time window (e.g., ...). Within ) the sampling period (e.g. The system continuously acquires the output power sequence of the ultrasonic generator and calculates its arithmetic mean to obtain the system's no-load ultrasonic power. Then, the tool is fed.

[0052] While the tool is in the real-time vertical depth of cutter during the feed process, the total power of the ultrasonic generator, the ultrasonic resonant frequency, and the real-time feed speed of the machine tool's Z-axis are simultaneously acquired at discrete sampling times. The difference between the total power of the ultrasonic generator and the system's idle ultrasonic power is used as the cutting power for cutting.

[0053] S102: Within a preset time window prior to the current sampling time, determine the slurry accumulation ratio based on the degree of fluctuation of the resonant frequency data; based on the slurry accumulation ratio, use the attenuation function to estimate the heat dissipation capacity characterization coefficient of the bottom of the hole through fluid convection per unit time.

[0054] During the ultrasonic milling of deep holes for machining microcrystalline glass cavities, as the hole depth increases, high-viscosity abrasive slurry easily accumulates in the narrow chip removal channel, severely hindering the cooling effect of the cutting fluid on the bottom of the hole. The dynamic decay of this cooling capacity cannot be directly measured. If it cannot be quantified in time, it will lead to uncontrolled heat accumulation and irreversible thermal damage.

[0055] To address the challenge of the "unobservable" cooling state, this invention proposes to indirectly characterize the degree of slurry accumulation by analyzing the numerical fluctuation of the resonant frequency signal within a preset time window before the current sampling time, and to calculate a slurry accumulation ratio that reflects the actual decrease in heat dissipation capacity. Furthermore, based on this ratio, an attenuation function is constructed to estimate the heat dissipation capacity characterization coefficient of heat loss from the bottom of the hole through convection per unit time.

[0056] The acquisition of heat dissipation capacity characterization coefficients enables online and quantitative perception of the critical but implicit cooling efficiency degradation process in deep hole machining, providing indispensable dynamic parameters on the heat dissipation side for subsequent thermal risk assessment, and fundamentally solving the problem of lagging thermal damage prevention and control caused by the unmeasurable cooling failure.

[0057] The preset time window is a window for data analysis at the current sampling time. In this embodiment of the invention, the preset time window can be specifically, for example, […]. Two seconds before the current sampling time (including the current sampling time) is used as the preset time window for the current sampling time. It should be noted that for cold start processes where the feeding process is less than two seconds, this invention can use a preset initial value or the value of the first sampling time of the feeding process to supplement the previously missing window data.

[0058] Furthermore, in some embodiments of the present invention, determining the slurry accumulation ratio based on the degree of numerical fluctuation of the resonant frequency data includes: calculating the numerical standard deviation of the resonant frequency data at all sampling times within the down-the-hole calibration interval as a reference fluctuation degree; calculating the numerical standard deviation of the resonant frequency data at all sampling times within a preset time window as a window fluctuation degree; and comparing the window fluctuation degree with the reference fluctuation degree to determine the slurry accumulation ratio, wherein the slurry accumulation ratio is not less than 1.

[0059] In a specific embodiment of the present invention, determining the slurry accumulation ratio based on the degree of fluctuation of the resonant frequency data specifically includes the following steps:

[0060] First, the resonant frequency signal output by the ultrasonic power supply is continuously acquired throughout the entire processing. The standard deviation of the resonant frequency data at the sampling time of the down-the-hole calibration interval is calculated as the benchmark fluctuation degree reflecting the overall vibration stability of the system. Second, for the current sampling time, a preset time window is selected, and the standard deviation of the resonant frequency data at all sampling times within the window is also calculated as the window fluctuation degree characterizing the local dynamic characteristics. Finally, the window fluctuation degree is compared with the benchmark fluctuation degree, and the resulting ratio is the slurry accumulation ratio.

[0061] It should be noted that when the baseline fluctuation value is 0, it indicates that the resonant frequency data values ​​are stable at all sampling times within the down-the-hole calibration interval, meaning there is no frequency change. Therefore, the corresponding window fluctuation value is also 0. Consequently, no ratio calculation is performed, and the slurry accumulation ratio can be directly recorded as a value of 1. Furthermore, when the ratio of the window fluctuation to the baseline fluctuation is less than 1, the slurry accumulation ratio is directly set to 1 for ease of subsequent analysis and calculation.

[0062] Because the gradual blockage of the chip removal channel during deep hole machining exacerbates the impedance mismatch in the ultrasonic system, leading to a significant increase in resonant frequency fluctuations, this ratio is typically not less than 1. A higher value indicates more severe slurry accumulation and a more pronounced deterioration in cooling conditions. This quantitative indicator effectively captures the degradation trend of cooling efficiency, providing crucial input for subsequent thermal risk modeling.

[0063] Furthermore, in some embodiments of the present invention, the heat dissipation capacity characterization coefficient of the bottom heat dissipation through fluid convection per unit time is estimated using an attenuation function based on the slurry accumulation ratio. This includes: calculating the difference between the slurry accumulation ratio and the unit value 1 as the blockage coefficient; analyzing the blockage coefficient according to a preset reference cooling rate and a preset blockage sensitivity to obtain the heat dissipation capacity characterization coefficient, wherein the product of the preset blockage sensitivity and the blockage coefficient is calculated, and the sum of the unit value 1 and the product value is used as the denominator, and the preset reference cooling rate is used as the numerator to calculate the fraction to obtain the heat dissipation capacity characterization coefficient.

[0064] In a specific embodiment of the present invention, the method of estimating the heat dissipation capacity of the bottom of the hole through fluid convection per unit time based on the slurry accumulation ratio using the decay function includes the following steps: First, the determined slurry accumulation ratio is calculated by difference with the unit value 1 to obtain the blockage coefficient characterizing the degree of blockage of the chip removal channel. This blockage coefficient reflects the accumulation increment under the current processing state relative to the ideal unobstructed working condition.

[0065] Subsequently, two process experience coefficients obtained in advance through offline process calibration are introduced: preset baseline cooling rate and preset blockage sensitivity. The preset baseline cooling rate characterizes the maximum convective heat dissipation capacity per unit time under ideal conditions without sludge buildup, while the preset blockage sensitivity reflects the sensitivity of the tool-material-ultrasonic system combination to cooling degradation caused by slurry buildup.

[0066] The preset baseline cooling rate and blockage sensitivity coefficient are preset empirical coefficients. They are obtained by: conducting destructive processing experiments on workpieces of the same material beforehand, recording the critical moment when thermal damage microcracks occur; and adjusting the above coefficients through data playback so that the processing heat accumulation index calculated by the above recursive integral model reaches the preset thermal damage threshold at the critical moment. In a typical embodiment, the preset baseline cooling rate can be taken as 0.1, and the preset blockage sensitivity coefficient can be taken as 0.5.

[0067] Based on this, the product of the preset blockage sensitivity and the blockage coefficient is calculated, and the unit value 1 is added to this product value as the denominator; finally, the preset benchmark cooling rate is used as the numerator.

[0068]

[0069] In the formula, This represents the heat dissipation capacity characterization coefficient at the k-th sampling time. This represents the preset baseline cooling rate, which can be set to 0.1. This indicates the preset blocking sensitivity, which can be set to 0.5. This represents the slurry accumulation ratio at the k-th sampling time. This represents the blocking coefficient.

[0070] This decay function reflects the physical law that cooling efficiency decreases nonlinearly with the increase of slurry accumulation: when the slurry accumulation ratio approaches 1 (i.e. no accumulation), the heat dissipation capacity characterization coefficient approaches the preset benchmark cooling rate; as the accumulation ratio increases, the denominator increases, and the heat dissipation capacity characterization coefficient decreases accordingly, thereby quantitatively characterizing the real-time degradation level of the convection heat dissipation capacity at the bottom of the hole, and providing accurate heat dissipation side parameters for the dynamic integration of the processing heat accumulation index in subsequent steps.

[0071] S103: Within a preset time window, perform correlation analysis based on cutting power and resonant frequency data to determine the weight of frictional heat generation; based on feed rate and cutting power, determine the overload frictional power that exceeds normal cutting requirements at the current sampling time; and combine the overload frictional power and frictional heat generation weight to determine the characteristic value of heat accumulation.

[0072] During ultrasonic milling of deep holes in microcrystalline glass, the load on the tool includes both normal material removal cutting force and abnormal frictional resistance caused by slurry accumulation. Both will cause an increase in power, but only the latter is the main cause of thermal damage. If these two types of power components cannot be distinguished and the total power is used as the criterion, it is easy to misjudge high-efficiency cutting as dangerous friction, or fail to identify it in time when real friction occurs.

[0073] To address this issue of "confusing heat source attributes," this invention proposes to extract the frictional heat generation weight that reflects abnormal frictional characteristics by performing correlation analysis on cutting power and resonant frequency data within a preset time window; simultaneously, by combining feed rate and cutting power, the overload frictional power exceeding normal cutting energy consumption at the current moment is quantified; finally, the two are fused to generate a heat accumulation characteristic value that characterizes the actual thermal risk contribution.

[0074] The embodiments of the present invention achieve effective decoupling of the cutting effect and the friction effect in the total power, enabling the assessment of heat input to change from extensive total monitoring to precise component identification, fundamentally improving the accuracy and reliability of thermal damage early warning.

[0075] Furthermore, in some embodiments of the present invention, the friction heat generation weight is determined by performing correlation analysis based on cutting power and resonant frequency data, including: performing Pearson correlation calculation on cutting power and resonant frequency data within a preset time window to obtain the Pearson correlation coefficient; when the Pearson correlation coefficient is less than a preset correlation threshold, the friction heat generation weight is set to 0; when the Pearson correlation coefficient drifts positively beyond the preset correlation threshold, it is determined that there is abnormal friction; the Pearson correlation coefficient is normalized and mapped to the value range of [0,1] to obtain the friction heat generation weight.

[0076] Among them, the cutting power and resonant frequency data at different sampling times within the preset time window are consistent, which indicates that the chip removal channel is clogged and the tool is subjected to abnormal friction in the high viscosity slurry. Therefore, correlation analysis can be achieved.

[0077] Within a preset time window prior to the current sampling time, the cutting power sequence of the electric spindle and the resonant frequency sequence of the ultrasonic power supply are acquired synchronously; Pearson correlation is calculated on these two time-series signals to obtain the Pearson correlation coefficient, which characterizes the degree of linear correlation between the two.

[0078] According to the mechanism of ultrasonic milling, during normal cutting, the rigid impact between the tool and the workpiece will cause the cutting power to increase while the resonant frequency decreases, showing a negative correlation. However, when the chip removal channel is clogged or the tool experiences abnormal friction in high-viscosity slurry, the frictional resistance increases, causing the cutting power to rise. At the same time, the ultrasonic vibration is blocked, causing the resonant frequency to also rise or fluctuate violently, thus showing a positive or weak correlation.

[0079] Therefore, in this embodiment of the invention, a preset correlation threshold (e.g., -0.3) is set to determine the friction-dominant state: when the Pearson correlation coefficient is less than the threshold, it is determined that the current load is mainly caused by normal cutting, and the friction heat generation weight is reset to 0; when the Pearson correlation coefficient is greater than or equal to the threshold, it indicates that there is a significant abnormal friction component. At this time, the Pearson correlation coefficient is linearly normalized (e.g., maximum and minimum value normalization, the maximum and minimum values ​​are set according to the specific scenario and data numerical characteristics, and the maximum and minimum values ​​are adjusted, calibrated or optimized, which does not constitute a limitation of this invention), and mapped to the value range of [0,1]. The result is the friction heat generation weight.

[0080] This weighted analysis of frictional heat generation quantitatively reflects the proportion of frictional heat generation in the total power under the current processing conditions, providing a key basis for the accurate calculation of subsequent heat accumulation characteristic values.

[0081] Furthermore, in some embodiments of the present invention, determining the overload friction power exceeding normal cutting requirements at the current sampling moment based on feed rate and cutting power includes: calculating the standard energy consumption per unit depth based on the total energy consumed in the down-the-hole calibration interval and the total effective cutting stroke; calculating the product of the feed rate and the standard energy consumption per unit depth at the current sampling moment as the ideal power; obtaining the total power of the tool during the feed process under the sampling period, and the idle power of the tool in the initial stage of idling before contacting the workpiece; combining the difference between the total power, the ideal power, and the idle power to obtain the overload friction power; and setting the overload friction power to 0 when the difference is less than 0.

[0082] Based on the sampling period duration, corresponding cutting power, and feed rate between adjacent sampling moments, the ratio of the total energy consumed by the tool during the effective cutting phase to the actual cutting stroke is calculated to obtain the standard energy consumption per unit depth, reflecting the energy consumption characteristics of material removal under the current working conditions. The feed rate at the current sampling moment is multiplied by the standard energy consumption per unit depth to obtain the ideal power required to complete the current feed under ideal frictionless conditions. At the same time, the total power consumed by the tool during the feed process within the current sampling period is acquired in real time, and the total power is baseline-corrected by combining the no-load power measured during the machining initialization phase (i.e., the tool has not yet contacted the workpiece and is only in an idling state).

[0083] Based on this, the difference between the corrected effective power (i.e., total power minus no-load power) and the ideal power is calculated as a potential overload component (total power minus no-load power, then subtracting the ideal power). If the difference is greater than zero, it is identified as overload friction power caused by slurry accumulation or abnormal friction; if the difference is less than or equal to zero, it indicates that the current load does not exceed the normal cutting energy consumption range, and the overload friction power is set to 0. Through the above method, the quantitative separation of unnecessary friction energy consumption is achieved, effectively distinguishing the reasonable power required for material removal from the redundant power that leads to thermal accumulation, laying the foundation for accurate modeling of thermal accumulation characteristic values.

[0084] Furthermore, in some embodiments of the present invention, the standard energy consumption per unit depth is calculated based on the total energy consumed in the interval and the total effective cutting stroke, including: calculating the ratio of the total energy consumed to the effective cutting stroke as the standard energy consumption per unit depth.

[0085] Specifically, data sampling can also be performed in the down-the-hole calibration interval. For the continuous sampling sequence in the down-the-hole calibration interval, the energy consumed in each sampling period can be calculated by numerical integration using the time interval between adjacent sampling times (i.e., the sampling period duration) and the cutting power at the corresponding time. The total energy consumed in the entire down-the-hole calibration interval can then be obtained by summing the results.

[0086] In other words, in this embodiment of the invention, the cutting power at each sampling moment in the downhole calibration interval is calculated as the product of the sampling time and the sampling period of the previous sampling moment, and is used as the energy consumed at the corresponding sampling moment. It should be noted that the energy consumed at the first sampling moment is 0.

[0087] Then, the sum of the energy consumed at all sampling moments is calculated as the total energy consumed. The effective cutting stroke is calculated in a similar manner, that is, the product of the tool feed rate at each sampling moment within the down-the-hole calibration interval and the sampling period between that sampling moment and the previous sampling moment is calculated as the cutting stroke at that sampling moment. The sum of the cutting strokes at all sampling moments within the down-the-hole calibration interval is then taken as the effective cutting stroke. Both methods use numerical integration to perform continuous analysis on adjacent sampling time intervals based on discrete time data.

[0088] Finally, the ratio of total energy consumption to effective cutting stroke is used as the standard energy consumption per unit depth. It should be noted that the effective cutting stroke value is not zero. If the effective cutting stroke is detected as zero within the down-the-hole calibration range, it indicates that the tool is not performing cutting work. In this case, an error can be reported directly, and a system check can be performed.

[0089] Standard energy consumption per unit depth, expressed as the average energy consumption per unit axial cutting depth, is used to characterize the baseline energy consumption level of the current material-tool combination under normal cutting conditions. This parameter serves as a dynamic reference benchmark, adaptively reflecting the impact of different machining stages or localized material hardness changes on the cutting load, providing a reliable basis for the accurate identification of subsequent overload friction power.

[0090] The heat accumulation characteristic value is determined by combining the overload friction power and the friction heat generation weight, including: weighting the friction heat generation weight based on the preset friction gain constant to obtain the friction characteristic coefficient; and calculating the product of the friction characteristic coefficient and the overload friction power as the heat accumulation characteristic value.

[0091] In this embodiment of the invention, the preset friction gain constant can be obtained through offline process calibration. This constant is used to characterize the amplification effect of abnormal friction behavior being converted into effective heat input under a specific tool-material-ultrasonic system combination. Specifically, the calibration example value is 2.0.

[0092] By introducing a preset friction gain constant to weight the frictional heat generation, a friction characteristic coefficient is obtained, which is used to weight the overload power: when the signal characteristics show strong fluid friction characteristics (i.e., the friction characteristic coefficient is large), the heat conversion effect of this part of the power will be significantly amplified, thereby more sensitively capturing the risk of heat accumulation caused by slurry accumulation.

[0093] Multiplying the friction characteristic coefficient by the overload friction power calculated at the current sampling time yields the thermal accumulation characteristic value. This thermal accumulation characteristic value is not a simple electrical power value, but a quantitative index that comprehensively considers three factors: the probability of friction (weight), the system's thermal conversion characteristics (gain constant), and redundant mechanical power consumption (overload power). It contributes to the actual thermal accumulation at the bottom of the hole, thus providing a high-fidelity heat source input for the dynamic modeling of the processing thermal accumulation index.

[0094] S104: Based on the thermal accumulation characteristic value and the heat dissipation capacity characterization coefficient, perform thermal risk analysis at the current sampling time to obtain the processing thermal accumulation index at the current sampling time, and perform processing retraction control based on the processing thermal accumulation index.

[0095] During the ultrasonic milling of deep holes in microcrystalline glass, heat continuously accumulates at the bottom of the hole and cannot be directly measured. If control is based solely on instantaneous power or temperature threshold, it is difficult to reflect the gradual and historical dependence of thermal damage, which can easily lead to response lag or malfunction.

[0096] To address the aforementioned issues, this invention proposes using the thermal accumulation characteristic value as a dynamic heat source term and the heat dissipation capacity characterization coefficient as a time-varying heat dissipation term to construct a recursive evolution thermal risk analysis model. This model performs thermal risk integral evaluation at the current sampling moment, generating a processing thermal accumulation index. Based on the processing thermal accumulation index, it makes real-time decisions to control the machining tool retraction.

[0097] It enables online, cumulative, and closed-loop risk quantification and proactive intervention for the unmeasurable heat accumulation process at the bottom of the hole, fundamentally solving the problem of thermal damage prevention and control failure caused by the lack of thermal history perception in traditional methods, and significantly improving the safety and reliability of the processing.

[0098] Furthermore, in some embodiments of the present invention, based on the thermal accumulation characteristic value and the heat dissipation capacity characterization coefficient, a thermal risk analysis is performed on the current sampling moment to obtain the processing thermal accumulation index at the current sampling moment, including: determining the attenuation coefficient based on the heat dissipation capacity characterization coefficient, wherein the larger the value of the heat dissipation capacity characterization coefficient, the smaller the value of the attenuation coefficient; using the product of the thermal accumulation characteristic value and the sampling period duration as the thermal energy index of invalid friction; setting the processing thermal accumulation index at the first sampling moment during the feeding process to 0; calculating the product of the processing thermal accumulation index and the attenuation coefficient at the previous sampling moment as the residual risk energy index remaining at the previous sampling moment; and using the sum of the residual risk energy index at the previous sampling moment and the thermal energy index at the corresponding sampling moment as the thermal energy index at the corresponding sampling moment, performing recursive analysis until the processing thermal accumulation index at the current sampling moment is obtained.

[0099] In a specific embodiment of the present invention, firstly, an attenuation coefficient is determined based on the heat dissipation capacity characterization coefficient. This attenuation coefficient is used to characterize the proportion of thermal risk remaining from the previous moment to the current moment. Its value is negatively correlated with the heat dissipation capacity characterization coefficient. That is, the larger the heat dissipation capacity characterization coefficient (indicating that the current heat dissipation capacity is stronger), the smaller the attenuation coefficient, and the less residual thermal risk. Conversely, when the cooling efficiency is severely degraded, the attenuation coefficient approaches 1, and the thermal risk is almost completely accumulated.

[0100] Specifically, the difference between the unit value 1 and the heat dissipation capacity characterization coefficient is calculated as the attenuation coefficient.

[0101] Multiplying the thermal accumulation characteristic value at the current sampling moment by the corresponding sampling period duration yields an index characterizing the thermal energy generated by ineffective friction during that period. Subsequently, the processing thermal accumulation index at the start of the feed process (i.e., the first effective cutting sampling point) is set to 0 as the initial condition for integration. Based on this, a recursive method is used to calculate the processing thermal accumulation index at each moment: for any current sampling moment, the product of the processing thermal accumulation index at the previous sampling moment and the attenuation coefficient is first calculated as the residual risk energy index at the previous moment; then, this residual risk energy index is added to the thermal energy index at the current sampling moment, and the sum is the processing thermal accumulation index at the current sampling moment. Through this dynamic integral model, the system can continuously track the accumulation and dissipation process of thermal risk at the bottom of the hole, achieving a quantitative characterization of the trend of unmeasurable thermal damage, and providing a high-confidence decision-making basis for precise tool retraction control.

[0102] In other words, the processing heat accumulation index at the first sampling moment during the feeding process is set to 0. Then, the processing heat accumulation index at the second sampling moment is calculated. Based on the processing heat accumulation index at the second sampling moment, the processing heat accumulation index at the third sampling moment is calculated. This process is repeated recursively until the current sampling moment is reached, thus obtaining the processing heat accumulation index at the current sampling moment.

[0103] It should be noted that when the processing heat accumulation index at the current sampling moment is detected to be greater than or equal to a preset risk threshold, it is determined that the microcrystalline glass processing has a risk of thermal damage, and a tool retraction and chip removal command is generated to achieve tool retraction control. The preset risk threshold is the processing heat accumulation index value corresponding to the critical moment of thermal damage, obtained through destructive testing in S102. Specifically, in the offline process calibration experiment, the critical moment when the workpiece develops thermal damage microcracks is recorded, and the processing heat accumulation index corresponding to that moment is calculated and recorded as the critical damage value (e.g., a value of 1000). To ensure processing safety, the preset risk threshold is set as a safe proportion of this critical damage value (e.g., 80%~90% of the critical damage value).

[0104] After the tool is retracted and chips are removed, a new feed process can continue, thereby recalculating the machining heat accumulation index at each new sampling moment and achieving automated thermal risk control.

[0105] This invention effectively solves the technical problem of uncontrollable thermal damage caused by slurry accumulation in deep hole machining of microcrystalline glass by constructing a dynamic thermal risk assessment system that integrates cooling state perception and decoupling of frictional heat sources. The slurry accumulation ratio is quantified using the degree of resonant frequency fluctuation, and a heat dissipation capacity characterization coefficient is estimated accordingly, enabling online characterization of the degradation of convection heat dissipation capacity at the bottom of the hole. Simultaneously, the correlation analysis between cutting power and resonant frequency extracts the weight of frictional heat generation, and the overload frictional power is determined by combining feed rate and power deviation, thereby generating a heat accumulation characteristic value that only reflects the contribution of harmful friction. Based on this, a recursive machining heat accumulation index model is established, using the heat accumulation characteristic value as the heat source and the heat dissipation capacity characterization coefficient as the heat dissipation term, to achieve cumulative quantification of the unmeasurable heat accumulation process at the bottom of the hole. This method abandons the traditional extensive control strategy that relies on fixed power thresholds or periodic pecking drills, and can accurately identify real thermal risks, triggering tool retraction and chip removal when necessary. This significantly improves automated machining efficiency and process reliability while ensuring the machining quality of the cavity without microcracks.

[0106] On the other hand, an automated machining device for laser gyroscope cavities based on ultrasonic milling is also provided. The device includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the methods described above.

[0107] 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.

[0108] 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. An automatic processing method of a laser gyro cavity based on ultrasonic milling, characterized in that, The method comprises: Obtaining cutting power, feed speed and resonance frequency data of the tool at different sampling time points in the shallow hole calibration interval and the deep hole interval, wherein the hole depth of the shallow hole calibration interval is less than that of the deep hole interval, the current sampling time point is located in the feeding process, and the feeding process is in the deep hole interval; Within a preset time window before the current sampling time point, determining a slurry accumulation ratio of slurry accumulation according to the numerical fluctuation degree of the resonance frequency data; based on the slurry accumulation ratio, using a decay function to estimate a heat dissipation capacity representation coefficient of heat dissipation of the hole bottom heat through fluid convection per unit time; Within the preset time window, performing correlation analysis based on the cutting power and the resonance frequency data to determine a friction heat generation weight; based on the feed speed and the cutting power, determining an overload friction power of the current sampling time point exceeding normal cutting requirements; combining the overload friction power and the friction heat generation weight to determine a heat accumulation characteristic value; Based on the heat accumulation characteristic value and the heat dissipation capacity representation coefficient, performing heat risk analysis on the current sampling time point to obtain a machining heat accumulation index of the current sampling time point, and performing machining relief control based on the machining heat accumulation index; The method for determining the heat dissipation capacity representation coefficient comprises: Calculating the difference between the slurry accumulation ratio and a unit value 1 as an obstruction coefficient; According to a preset reference cooling rate and a preset obstruction sensitivity, analyzing the obstruction coefficient to obtain the heat dissipation capacity representation coefficient, wherein the product value of the preset obstruction sensitivity and the obstruction coefficient is calculated, the sum value of the unit value 1 and the product value is taken as the denominator, the preset reference cooling rate is taken as the numerator, and the heat dissipation capacity representation coefficient is calculated by the fraction; The method for determining the heat accumulation characteristic value comprises: Based on a preset friction gain constant, the friction heat generation weight is weighted and corrected to obtain a friction characteristic coefficient; The product value of the friction characteristic coefficient and the overload friction power is calculated as the heat accumulation characteristic value.

2. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 1, characterized in that, The method for determining the slurry accumulation ratio of slurry accumulation according to the numerical fluctuation degree of the resonance frequency data comprises: The numerical standard deviation of the resonance frequency data at all sampling time points in the shallow hole calibration interval is calculated as a reference fluctuation degree; The numerical standard deviation of the resonance frequency data at all sampling time points in the preset time window is calculated as a window fluctuation degree; The window fluctuation degree is compared with the reference fluctuation degree to determine the slurry accumulation ratio of slurry accumulation, and the slurry accumulation ratio is not less than 1.

3. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 1, characterized in that, The method for determining the friction heat generation weight by correlation analysis based on the cutting power and the resonance frequency data comprises: Pearson correlation calculation is performed on the cutting power and the resonance frequency data in the preset time window to obtain a Pearson correlation coefficient; When the Pearson correlation coefficient is less than a preset correlation threshold, the friction heat generation weight is set to 0; when the Pearson correlation coefficient drifts to the positive direction and exceeds the preset correlation threshold, it is determined that there is abnormal friction, the Pearson correlation coefficient is normalized to map to the value range interval [0, 1] to obtain the friction heat generation weight.

4. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 1, characterized in that, The method for determining the overload friction power of the current sampling time point exceeding normal cutting requirements based on the feed speed and the cutting power comprises: In the shallow hole calibration interval, the unit depth standard energy consumption is calculated according to the total energy consumption and the total effective cutting stroke. The product value of the feeding speed at the current sampling moment and the unit depth standard energy consumption is calculated as the ideal power; The total power of the cutter in the feeding process and the idling power of the cutter in the initialization stage without contacting the workpiece are obtained under the sampling period length, and the overload friction power is obtained by combining the difference between the total power and the ideal power and the idling power. When the difference is less than 0, the overload friction power is set to 0.

5. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 4, characterized in that, The unit depth standard energy consumption is calculated according to the total consumed energy and the total effective cutting stroke in the interval, including: The ratio of the total consumed energy and the effective cutting stroke is calculated as the unit depth standard energy consumption.

6. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 1, characterized in that, The thermal risk analysis is performed on the current sampling moment based on the thermal accumulation characteristic value and the heat dissipation ability representation coefficient to obtain the machining thermal accumulation index of the current sampling moment, including: The attenuation coefficient is determined based on the heat dissipation ability representation coefficient, wherein the larger the value of the heat dissipation ability representation coefficient is, the smaller the value of the attenuation coefficient is; The product value of the thermal accumulation characteristic value and the sampling period length is taken as the thermal energy index of the invalid friction; The machining thermal accumulation index of the first sampling moment in the feeding process is set to 0; The product value of the machining thermal accumulation index of the last sampling moment and the attenuation coefficient is calculated as the residual risk energy index of the last sampling moment; The sum of the residual risk energy index of the last sampling moment and the thermal energy index of the corresponding sampling moment is taken as the thermal energy index of the corresponding sampling moment, and the recursive analysis is performed until the machining thermal accumulation index of the current sampling moment is obtained.

7. The method for automatic processing of a laser gyroscope cavity based on ultrasonic milling according to claim 1, characterized in that, The machining relief control is performed based on the machining thermal accumulation index, including: When the machining thermal accumulation index is greater than or equal to the preset risk threshold, it is determined that the glass ceramic cutting tool has the thermal damage risk, the relief chip removal instruction is generated, and the relief control is realized.

8. An apparatus for automated processing of a laser gyroscope cavity based on ultrasonic milling, the apparatus comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to realize the steps of the method of any one of claims 1-7. The processor executes the computer program to realize the steps of the method of any one of claims 1-7.

Images