End mill tool bit parameter optimization method

Abstract

The invention relates to the technical field of numerical control machining, and discloses an end mill tool bit parameter optimization method which comprises the following steps: step 1, collecting cutting force data, cutting temperature data and machining vibration data in a machining process in real time through a sensor group integrated on an end mill tool bit; the sensor group comprises a strain sensor, a temperature sensor and a vibration sensor; 2, the collected cutting force data, cutting temperature data and machining vibration data are subjected to wireless processing; according to the technical scheme, a closed-loop control system is constructed, the feeding speed, the cutting angle and the cutting depth are automatically adjusted according to dynamically-changed cutting force and vibration data, it is ensured that the machining process is always kept within a stable interval, the machining precision and the surface quality of a workpiece are effectively improved, and the machining efficiency is improved. Compared with the technical scheme that fixed preset parameters are adopted for machining in the prior art, the defect that the precision and quality of the workpiece are finally damaged due to process instability caused by the fact that adaptive adjustment cannot be achieved is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of numerical control machining, in particular to a method for optimizing the parameters of a ball end mill head. BACKGROUND

[0002] In the field of aerospace and high-end equipment manufacturing, high-temperature alloys or titanium alloys are often used to manufacture key load-bearing and heat-resistant components. Ball end mills play a crucial role in the cutting process of such materials.

[0003] Currently, when cutting the above-mentioned difficult-to-machine materials, the commonly used technical method is to set a fixed group of cutting parameters before starting the machining, including feed speed, cutting angle, and cutting depth. The parameters are provided by the experience values in the process manual or through offline simulation. During the machining process, the numerical control machine strictly follows the pre-set fixed parameters without adjustment.

[0004] The above-mentioned technical method using fixed cutting parameters has the following significant technical defects:

[0005] The existing technology lacks real-time monitoring of tool wear when machining high-hardness materials, which inevitably leads to rapid and abnormal tool wear, resulting in a shortened tool life cycle and a significant increase in manufacturing costs.

[0006] The existing cutting method uses pre-set fixed machining parameters, which cannot be adjusted adaptively according to the dynamically changing cutting force and vibration during machining, thereby causing poor machining process stability and ultimately affecting the machining precision and surface quality of the workpiece.

[0007] The existing technology relies on fixed empirical parameters for machining, which makes it difficult to maintain the tool in the best cutting state during the machining cycle. To ensure machining stability, operators have to use conservative machining parameters, resulting in low overall machining efficiency and failing to meet the needs of efficient production.

[0008] Therefore, the present application proposes a method for optimizing the parameters of a ball end mill head to solve the above-mentioned problems. SUMMARY

[0009] To overcome the shortcomings of the prior art, the present application provides a method for optimizing the parameters of a ball end mill head to solve the problems mentioned in the background art.

[0010] To achieve the above-mentioned purposes, the present application is implemented by the following technical solution: a method for optimizing the parameters of a ball end mill head, comprising:

[0011] Step one, through the sensor group integrated on the end mill cutter head, real-time acquisition of cutting force data, cutting temperature data and machining vibration data in the machining process; the sensor group includes strain sensor, temperature sensor and vibration sensor;

[0012] Step two, the collected cutting force data, cutting temperature data and machining vibration data are wirelessly sent to the main control calculation unit, and the main control calculation unit uses a preset artificial intelligence algorithm to analyze and predict the current wear state of the cutter according to the real-time acquisition of cutting force data, cutting temperature data and machining vibration data in the machining process;

[0013] Step three, according to the current wear state of the cutter predicted in step two, the main control calculation unit dynamically calculates and generates a set of optimal cutting parameters, including feed speed, cutting angle and cutting depth, with the optimization goal of maximizing material removal rate and minimizing tool wear rate;

[0014] In the step three, further comprising:

[0015] Substep According to the current wear state of the cutter determined in step two, the corresponding weight factor is selected from the preset weight strategy library And A multi-objective utility function for parameter optimization is constructed The expression of the multi-objective utility function is as follows:

[0016] ,

[0017] Among them, is the feed speed, is the cutting angle, is the cutting depth, is the material removal rate, is the preset maximum material removal rate, is the tool wear rate predicted based on the current working condition, is the preset maximum allowable tool wear rate, is the efficiency weight factor, is the life weight factor;

[0018] Substep Based on the multi-objective utility function , a constraint optimization problem containing boundary conditions and performance constraints is established, which aims to solve a set of optimal cutting parameters This problem is subject to the following constraint condition set:

[0019] ≤ ≤ ,

[0020] ≤ ≤ ,

[0021] ≤ ≤ ,

[0022] ≤ ,

[0023] wherein, and are the minimum and maximum feed speeds allowed by the machine tool, and are the minimum and maximum cutting angles allowed by the process, and are the minimum and maximum cutting depths allowed by the process, is the predicted machining vibration amplitude, is the preset vibration safety threshold value;

[0024] sub-step , using an intelligent optimization algorithm, iteratively optimizing the constructed multi-objective utility function within the defined set of constraints until the function converges to a maximum value, and the corresponding set of at the time of obtaining the maximum value is output as the optimal cutting parameters for use in step four for automatic adjustment;

[0025] Step four, the main control calculation unit sends the optimal cutting parameter instructions generated in step three to the numerical control machine tool control system, the control system automatically adjusts the machining parameters of the end mill, and continuously repeats steps one to four to form a closed loop control.

[0026] Preferably, in step one, the cutting force data, cutting temperature data and machining vibration data collected during the machining process are digitized before being sent.

[0027] Preferably, in step two, the artificial intelligence algorithm analyzes and predicts the received cutting force data, cutting temperature data and machining vibration data before performing data fusion processing.

[0028] Preferably, in step two, the current tool wear state relied on by the artificial intelligence algorithm is obtained by calculating a pre-constructed mathematical model, which is used to describe the functional relationship between cutting force, cutting temperature, machining vibration and tool wear.

[0029] Preferably, in the step three, the optimization target comprises maintaining the machining vibration data within a preset stable threshold range.

[0030] In the step three, the process of dynamically calculating to generate a set of optimal cutting parameters is continuously performed within a machining cycle.

[0031] Preferably, the automatic adjustment in the step four is triggered and performed when the predicted current tool wear state exceeds a preset wear threshold.

[0032] The automatic adjustment in the step four is triggered and performed when the collected cutting force data fluctuates beyond a preset range.

[0033] Preferably, the end mill head parameter optimization method is applied to the cutting machining scene of high-temperature alloy materials.

[0034] Preferably, in the step one, further comprising:

[0035] Sub-step , acquiring analog voltage signals representing cutting force , analog voltage signals representing cutting temperature , and analog voltage signals representing machining vibration , wherein, t is time;

[0036] Sub-step , sampling and digitizing the acquired analog voltage signals , , to obtain discrete digital signal sequences, and then processing the digital signal sequences through a moving average filtering algorithm to generate filtered cutting force data , filtered cutting temperature data , and filtered machining vibration data , the calculation method of the moving average filtering algorithm being:

[0037] ,

[0038] wherein, is any filtered data, is the original digital signal at the current and historical time, is the index of the sampling time, is the size of the filtering window;

[0039] Sub-step , generating the filtered cutting force data , filtered cutting temperature data the filtered machining vibration data When all of the following conditions are met, the set of data is determined as valid data and is packaged as a data frame for subsequent wireless transmission:

[0040] ≤ ≤ ,

[0041] ≤ ≤ ,

[0042] ≤ ≤ ,

[0043] wherein, and are the lower limit and the upper limit of the effective range of cutting force, and are the lower limit and the upper limit of the effective range of cutting temperature, and are the lower limit and the upper limit of the effective range of machining vibration.

[0044] Preferably, in step two, further comprising:

[0045] Sub-step The master control unit receives the data frame sent in step one, and performs normalization processing on the filtered cutting force data , filtered cutting temperature data and filtered machining vibration data in the data frame to eliminate dimensional differences and generate a normalized feature vector The calculation method of the normalization processing is:

[0046] ,

[0047] wherein, is any normalized data component, is the corresponding filtered data, and are the preset reference minimum value and the reference maximum value, and the normalized feature vector generated finally is ;

[0048] Sub-step The generated normalized feature vector is taken as input and substituted into the pre-constructed and trained multiple nonlinear regression mathematical model to calculate the quantitative tool wear prediction value , the function expression of the multivariate nonlinear regression mathematical model is:

[0049] ,

[0050] wherein, is the tool wear prediction value at the time point, and is a component in the normalized feature vector , , and is the model regression coefficient obtained by training historical experimental data;

[0051] Substep , the calculated quantitative tool wear prediction value is compared with a set of preset wear state judgment thresholds to determine the current wear state of the tool, and the current wear state is the direct basis for the optimization decision in subsequent step three, and the judgment condition is:

[0052] If , it is determined that the current wear state is normal wear;

[0053] If ≤ , it is determined that the current wear state is moderate wear;

[0054] If ≥ , it is determined that the current wear state is severe wear;

[0055] wherein, is the first wear threshold for distinguishing between normal wear and moderate wear, is the second wear threshold for distinguishing between moderate wear and severe wear.

[0056] Preferably, in step four, further comprising:

[0057] Substep , the main control calculation unit encapsulates the output optimal cutting parameters into a standardized numerical control instruction data packet , and sends the numerical control instruction data packet to the numerical control machine tool control system through a preset communication protocol;

[0058] Substep , after sending the numerical control instruction data packet , the main control calculation unit queries the execution state feedback code from the numerical control machine tool control system, and determines the execution state feedback code ​The value of the instruction is determined whether the instruction is successfully received and executed, and the determination condition is:

[0059] If , it is determined that the instruction is successfully executed, and an execution success flag is generated; wherein is a predefined feedback code indicating successful execution of the instruction;

[0060] The main control unit checks the generated execution success flag , and at the same time acquires a program end flag indicating whether the machining task is completed When the following logical conditions are met, the system control flow returns to step one to collect machining data under new parameters, realizing closed-loop control:

[0061] ,

[0062] wherein is a logical AND operation, is a logical NOT operation, and the condition indicates that when the instruction is successfully executed and the machining task has not ended, the optimization loop continues.

[0063] The present application provides a method for optimizing the parameters of an end mill head. The method has the following advantages:

[0064] 1. The present application adopts a technical solution of constructing a closed-loop control system, automatically adjusting the feed speed, cutting angle and cutting depth according to the dynamically changing cutting force and vibration data, ensuring that the machining process is always maintained within a stable interval, effectively improving the machining precision and surface quality of the workpiece, compared with the technical solution of using fixed preset parameters for machining in the prior art, solving the problem of process instability caused by the inability to adaptively adjust, and ultimately damaging the precision and quality of the workpiece.

[0065] 2. The present application adopts a technical solution of integrating a sensor group to collect cutting force, temperature and vibration data in real time, and using artificial intelligence algorithm to predict the current wear state of the tool, realizing active intervention and control of the tool wear process, significantly prolonging the effective service life of the tool, compared with the technical solution lacking real-time monitoring means in the prior art, solving the problem of rapid abnormal wear of the tool, leading to shortened service life and increased manufacturing cost.

[0066] 3. The application adopts a technical scheme of dynamically solving by establishing a multi-objective utility function with the target of maximizing material removal rate and minimizing tool wear rate, combined with an intelligent optimization algorithm, so that the tool can work under the optimal cutting parameters in the machining life cycle, greatly improving the overall machining efficiency. Compared with the prior art which relies on fixed empirical parameters and selects conservative machining parameters to ensure stability, the application solves the problem of low overall machining efficiency caused by the inability of the tool to maintain the best cutting state. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 The flowchart of the application. DETAILED DESCRIPTION

[0068] To enable those skilled in the art to understand the application scheme, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the application.

[0069] The application will be described in detail below with reference to the drawings:

[0070] Embodiment:

[0071] Please refer to the accompanying Figure 1 The embodiment of the application provides a method for optimizing the parameters of a ball end mill head, comprising:

[0072] Step one, through the sensor group integrated on the ball end mill head, real-time collection of cutting force data, cutting temperature data and machining vibration data in the machining process; the sensor group includes a strain sensor, a temperature sensor and a vibration sensor;

[0073] Step two, wireless transmission of the collected cutting force data, cutting temperature data and machining vibration data to the main control computing unit, the main control computing unit uses a preset artificial intelligence algorithm to analyze and predict the current wear state of the tool according to the real-time collected cutting force data, cutting temperature data and machining vibration data in the machining process;

[0074] Step three, the main control computing unit dynamically calculates and generates a set of optimal cutting parameters according to the predicted current wear state of the tool in step two, with the optimization target of maximizing material removal rate and minimizing tool wear rate, the optimal cutting parameters including feed speed, cutting angle and cutting depth;

[0075] Step four, the master control unit sends the optimal cutting parameter instruction generated in step three to the numerical control machine tool control system, the control system automatically adjusts the machining parameters of the end mill, and steps one to four are repeatedly continued to form a closed loop control.

[0076] Step one further comprises:

[0077] Sub-step , using strain sensors, temperature sensors and vibration sensors to obtain analog voltage signals representing cutting force , analog voltage signals representing cutting temperature And analog voltage signals representing machining vibration , wherein, is time;

[0078] Sub-step , the obtained analog voltage signals , , are sampled and digitized to obtain discrete digital signal sequences, and then the digital signal sequences are processed by a moving average filtering algorithm to generate filtered cutting force data , filtered cutting temperature data And filtered machining vibration data The calculation method of the moving average filtering algorithm is:

[0079] ,

[0080] Wherein, is any filtered data, is the original digital signal at the current and historical time, is the index of the sampling time, is the size of the filter window;

[0081] Sub-step , compare the generated filtered cutting force data , filtered cutting temperature data And filtered machining vibration data With the preset effective data threshold range, when all the following conditions are met, the data set is determined as valid data and encapsulated as a data frame for subsequent wireless transmission:

[0082] ≤ ≤ ,

[0083] ≤ ≤ ,

[0084] ≤ ≤ ,

[0085] wherein, and are lower and upper limits of the effective range of cutting force, and are lower and upper limits of the effective range of cutting temperature, and are lower and upper limits of the effective range of machining vibration.

[0086] In step two, further comprising:

[0087] Sub-step , the master control unit receives the data frame sent in step one, and normalizes the filtered cutting force data , filtered cutting temperature data and filtered machining vibration data in the data frame to eliminate dimensional differences, generating a normalized feature vector , the calculation method of normalization processing is:

[0088] ,

[0089] wherein, is any normalized data component, is the corresponding filtered data, and are the preset reference minimum value and reference maximum value, and the finally generated normalized feature vector is ;

[0090] Sub-step , the generated normalized feature vector is taken as input and substituted into the pre-constructed and trained multiple nonlinear regression mathematical model to calculate the quantitative tool wear prediction value , the function expression of the multiple nonlinear regression mathematical model is:

[0091] ,

[0092] wherein, is the tool wear prediction value at moment, and are components in the normalized feature vector , , and The regression coefficient of the model trained by the historical experimental data;

[0093] Sub-step The calculated quantitative tool wear prediction value is compared with a set of preset wear state judgment thresholds to determine the current wear state of the tool, which is the direct basis for the optimization decision in subsequent step three, and the judgment condition is:

[0094] If , it is determined that the current wear state is normal wear;

[0095] If ≤ , it is determined that the current wear state is moderate wear;

[0096] If ≥ , it is determined that the current wear state is severe wear;

[0097] Wherein, is the first wear threshold for distinguishing between normal wear and moderate wear, is the second wear threshold for distinguishing between moderate wear and severe wear.

[0098] In step three, further comprising:

[0099] Sub-step According to the current wear state of the tool determined in step two, select the corresponding weight factor from the preset weight strategy library , construct a multi-objective utility function for parameter optimization , the multi-objective utility function is expressed as:

[0100] ,

[0101] Wherein, is the feed speed, is the cutting angle, is the cutting depth, is the material removal rate, is the preset maximum material removal rate, is the tool wear rate predicted based on the current working condition, is the preset maximum allowable tool wear rate, is the efficiency weight factor, is the life weight factor;

[0102] Sub-step Based on the multi-objective utility function , a constrained optimization problem is established including boundary conditions and performance constraints, the constrained optimization problem aims to solve a set of optimal cutting parameters , the problem is limited by the following constraint set:

[0103] ≤ ≤ ,

[0104] ≤ ≤ ,

[0105] ≤ ≤ ,

[0106] ≤ ,

[0107] wherein, and are the minimum and maximum feed speeds allowed by the machine tool, and are the minimum and maximum cutting angles allowed by the process, and are the minimum and maximum cutting depths allowed by the process, is the predicted machining vibration amplitude, is the preset vibration safety threshold value;

[0108] Substep , an intelligent optimization algorithm is used to perform iterative optimization calculation on the constructed multi-objective utility function within the defined constraint set, until the function converges to obtain the maximum value, and the corresponding set of at the time of obtaining the maximum value is taken as the optimal cutting parameters output, for automatic adjustment in step four.

[0109] In step four, further comprising:

[0110] Substep , the main control calculation unit encapsulates the output optimal cutting parameters into a standardized numerical control instruction data packet , and sends the numerical control instruction data packet to the numerical control machine tool control system through a preset communication protocol;

[0111] Substep , after sending the numerical control instruction data packet , the main control calculation unit queries the numerical control machine tool control system for an execution state feedback code , and determines whether the execution state feedback code The value of the success flag is determined by the value of the feedback code received from the tool. The success flag is set to 1 if the feedback code is equal to the predefined code, otherwise it is set to 0.

[0112] If the success flag is equal to 1, the system determines that the instruction has been successfully received and executed, and the process proceeds to step 3. If the success flag is equal to 0, the system determines that the instruction has not been successfully received and executed, and the process proceeds to step 2. wherein the feedback code is a predefined code indicating successful execution of the instruction.

[0113] In step 2, the system generates a new instruction based on the current state of the tool and the machining task, and sends the new instruction to the tool. In step 3, the system checks the success flag generated in step 2 and the program end flag indicating whether the machining task has been completed. If the success flag is equal to 1 and the program end flag is equal to 0, the system returns to step 1 to collect machining data under new parameters, achieving closed-loop control.

[0114]

[0115] wherein the logical AND operation and the logical NOT operation indicate that the optimization loop continues when the instruction is successfully executed and the machining task has not been completed. By integrating sensors directly into the tool head at the cutting zone, combined with moving average filtering and effective data threshold judgment, the real-time, high-fidelity, and correlation of the collected cutting force, temperature, and vibration data are ensured. This fundamentally ensures the quality of the source data for the analysis and decision-making of the closed-loop control system, providing accurate and noise-free reliable input for subsequent artificial intelligence algorithms, and is a necessary data basis for intelligent adaptive machining.

[0116] By introducing artificial intelligence algorithms, the multi-dimensional physical data collected at the front end is successfully converted into a direct and quantitative understanding of the tool health status. This step uses normalization to eliminate the dimensional influence between data, and then uses a pre-trained multivariate nonlinear regression model to accurately predict the current wear state of the tool. Finally, through hierarchical judgment, clear and explicit scientific basis is provided for subsequent optimization decisions, and the original sensor signal is upgraded to intelligent insight with guiding significance.

[0117] By constructing a multi-objective utility function that can balance between machining efficiency and tool life, the dynamic balance of the machining target is achieved. This step can adaptively adjust the optimization strategy according to the real-time health status of the tool, and under the constraints of the physical capabilities of the machine tool and the stability of the machining process, the optimal machining parameter combination under the current state is solved through intelligent optimization algorithms, ensuring the scientificity, safety, and maximization of benefits of the final decision.

[0118]

[0119] ​​​​​By seamlessly transforming the optimized decision into specific execution actions of the CNC machine tool, and establishing a feedback mechanism for instruction execution confirmation and task state checking, a complete automatic closed-loop control is finally formed. This step ensures that the optimized instructions generated by the system can be executed stably and reliably, and through the continuous repetition of the "perception-analysis-decision-execution" process, the system has continuous self-optimization capability, ensuring that the end mill is always in the best working state dynamically in the machining life cycle.

[0120] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely divergences of the principles and spirit of the application and that numerous modifications, changes, substitutions, and alterations can be made thereto without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. A method for optimizing end mill head parameters, characterized in that, include: Step 1: Real-time acquisition of cutting force data, cutting temperature data, and machining vibration data during the machining process using a sensor group integrated on the end mill head; the sensor group includes a strain gauge sensor, a temperature sensor, and a vibration sensor. Step 2: The collected cutting force data, cutting temperature data, and machining vibration data are wirelessly transmitted to the main control computing unit. The main control computing unit uses a preset artificial intelligence algorithm to analyze and predict the current wear state of the tool based on the real-time collected cutting force data, cutting temperature data, and machining vibration data during the machining process. Step 3: Based on the current tool wear state predicted in Step 2, the main control calculation unit dynamically calculates and generates a set of optimal cutting parameters with the optimization objectives of maximizing material removal rate and minimizing tool wear rate. The optimal cutting parameters include feed rate, cutting angle and depth of cut. Step three further includes: Sub-step Based on the current tool wear state determined in step two, the corresponding weighting factor is selected from the preset weighting strategy library. and Construct a multi-objective utility function for parameter optimization. The multi-objective utility function The expression is: ， in, For feed rate, For cutting angle, For cutting depth, For material removal rate, The preset maximum material removal rate, The tool wear rate is predicted based on the current operating conditions. The preset maximum allowable tool wear rate, As an efficiency weighting factor, Lifetime weighting factor; Sub-step Based on the multi-objective utility function A constrained optimization problem is established, which includes boundary conditions and performance constraints. The goal of this constrained optimization problem is to find a set of optimal cutting parameters. The problem is subject to the following set of constraints: ≤ ≤ ， ≤ ≤ ， ≤ ≤ ， ≤ ， in, and These are the minimum and maximum allowable feed rates of the machine tool. and These are the minimum and maximum cutting angles allowed by the process. and The minimum and maximum depths of cut allowed by the process. For the predicted processing vibration amplitude, To preset the vibration safety threshold; Sub-step Using an intelligent optimization algorithm, within the defined set of constraints, the constructed multi-objective utility function is optimized. Perform iterative optimization calculations until the function converges and reaches its maximum value, and then collect a set of values ​​corresponding to the maximum value. As the optimal cutting parameters Output for automatic adjustment in step four; Step four: The main control computing unit sends the optimal cutting parameter command generated in step three to the CNC machine tool control system. The control system automatically adjusts the machining parameters of the end mill and continuously repeats steps one to four to form a closed-loop control.

2. The method for optimizing end mill head parameters according to claim 1, characterized in that, In step one, the cutting force data, cutting temperature data, and processing vibration data collected in real time during the processing are digitized before being sent.

3. The method for optimizing end mill head parameters according to claim 1, characterized in that, In step two, before the artificial intelligence algorithm analyzes and predicts, it needs to perform data fusion processing on the received cutting force data, cutting temperature data, and machining vibration data.

4. The method for optimizing end mill head parameters according to claim 3, characterized in that, In step two, the current wear state of the tool, which is the basis of the artificial intelligence algorithm, is calculated by a pre-built mathematical model. The mathematical model is used to describe the functional relationship between cutting force, cutting temperature, machining vibration and tool wear.

5. The method for optimizing end mill head parameters according to claim 1, characterized in that, In step three, the optimization objective includes maintaining the processing vibration data within a preset stable threshold range; In step three, the process of dynamically calculating and generating a set of optimal cutting parameters is performed continuously within the machining cycle.

6. The method for optimizing end mill head parameters according to claim 1, characterized in that, The automatic adjustment in step four is triggered when the predicted current wear state of the tool exceeds the preset wear threshold. The automatic adjustment in step four is triggered when the collected cutting force data fluctuates beyond a preset range.

7. The method for optimizing end mill head parameters according to claim 1, characterized in that, The method for optimizing end mill head parameters is applied to the cutting and machining of high-temperature alloy materials.

8. The method for optimizing end mill head parameters according to claim 1, characterized in that, Step one further includes: Sub-step Using the strain gauge sensor, temperature sensor, and vibration sensor, an analog voltage signal characterizing the cutting force is acquired. Analog voltage signal characterizing cutting temperature Analog voltage signal characterizing processing vibration ,in, For time; Sub-step The acquired analog voltage signal , , Sampling and digitization are performed to obtain discrete digital signal sequences. These sequences are then processed using a moving average filtering algorithm to generate filtered cutting force data. Filtered cutting temperature data Compared with filtered processing vibration data The calculation method of the moving average filtering algorithm is as follows: ， in, For any filtered data, The raw digital signals for the current and historical moments. This is the index of the sampling time. This refers to the size of the filter window; Sub-step The generated filtered cutting force data Filtered cutting temperature data Compared with filtered processing vibration data The data is compared with a preset valid data threshold range. If all of the following conditions are met, the data set is determined to be valid and encapsulated into a data frame for subsequent wireless transmission: ≤ ≤ ， ≤ ≤ ， ≤ ≤ ， in, and These represent the lower and upper limits of the effective range of cutting force. and These represent the lower and upper limits of the effective cutting temperature range. and These are the lower and upper limits of the effective range of processing vibration.

9. The method for optimizing end mill head parameters according to claim 1, characterized in that, Step two further includes: Sub-step The main control computing unit receives the data frame sent in step one and processes the filtered cutting force data within the data frame. Filtered cutting temperature data Compared with filtered processing vibration data Normalization is performed to eliminate dimensional differences and generate normalized feature vectors. The normalization calculation method is as follows: ， in, For any normalized data component, For the corresponding filtered data, and The final normalized feature vector is generated based on the preset reference minimum and reference maximum values. for ; Sub-step The generated normalized feature vector As input, the pre-built and trained multivariate nonlinear regression mathematical model is substituted to calculate the quantified tool wear prediction value. The functional expression of the multivariate nonlinear regression mathematical model is: ， in, In order to be in Predicted tool wear value at any given time. and Normalized eigenvectors The components in , and These are the regression coefficients of the model obtained by training with historical experimental data; Sub-step The quantified tool wear prediction value obtained through calculation The tool's current wear state is determined by comparing it with a set of preset wear state judgment thresholds. This current wear state is the direct basis for the optimization decision in the subsequent step three. The judgment conditions are as follows: like The current wear condition is determined to be normal wear. like ≤ The current wear condition is determined to be moderate wear. like ≥ The current wear condition is determined to be severe wear; in, The first wear threshold is used to distinguish between normal wear and moderate wear. A second wear threshold is used to distinguish between moderate and severe wear.

10. The method for optimizing end mill head parameters according to claim 1, characterized in that, Step four further includes: Sub-step The main control computing unit will output the optimal cutting parameters. Encapsulated into standardized CNC instruction data packages The numerical control instruction data packet The data is sent to the CNC machine tool control system via a preset communication protocol. Sub-step The main control computing unit sends the numerical control instruction data packet. Then, query the execution status feedback code from the CNC machine tool control system. And based on the execution status feedback code The value determines whether the instruction was successfully received and executed. The determination condition is: like If the instruction is executed successfully, a success flag is generated. ;in, A predefined feedback code indicating successful execution of an instruction; Sub-step The main control computing unit checks the generated execution success flag. At the same time, it acquires the program end flag indicating whether the processing task is completed. When the following logical conditions are met, the system control flow will return to step one to collect processing data under the new parameters and achieve closed-loop control: ， in, For logical AND operation, The condition is a logical NOT operation, indicating that the optimization loop continues if the instruction is executed successfully and the processing task has not yet ended.

Images