Numerical control lathe groove turning control method and system

Abstract

The invention relates to the technical field of numerical control lathes, in particular to a numerical control lathe groove turning machining control method and system.The numerical control lathe groove turning machining control method comprises the steps that S1, a numerical control system controls a numerical control lathe to conduct groove turning machining, various sensors are installed on the numerical control lathe, various data are collected in real time, and the data are transmitted to a cloud platform through the Internet of Things technology to be stored and analyzed; s2, the cloud platform uses an artificial intelligence algorithm to identify abnormal conditions in the machining process according to the various data, when the method is used, cutting parameters are adjusted in real time, cutting force fluctuation is reduced, and the machining precision and the surface quality are improved conveniently; cutting parameters are dynamically optimized, so that tool abrasion is conveniently reduced, and the service life is prolonged; according to real-time data in the machining process, cutting parameters are optimized, and the machining efficiency is improved; and by combining an artificial intelligence algorithm and an internet of things technology, self-learning is facilitated, processing characteristics of different materials can be adapted, and the intelligent level of the system can be improved.

Description

Technical Field

[0001] This invention relates to the field of CNC lathe technology, specifically to a method and system for controlling the machining of grooves on a CNC lathe. Background Technology

[0002] CNC lathe grooving refers to a modern machining process that uses a CNC lathe, an automated machining equipment, to cut grooves with specific dimensions, shapes, and precision requirements on the surface of a rotating workpiece (outer circle, inner hole, end face, etc.) by programming and controlling specially designed grooving tools according to preset machining paths and process parameters.

[0003] Essentially, it is a process that automates, achieves high precision, and repeatability in the traditional manual or ordinary lathe groove machining process through digital program control. It is a specialized process for groove features in CNC turning.

[0004] With the widespread application of composite materials in manufacturing, CNC lathes often face problems such as large variations in cutting forces and severe tool wear when machining multi-material composite workpieces. Traditional machining control methods are insufficient to address these challenges.

[0005] The patent with publication number CN118707895A discloses in its specification a "CNC lathe grooving control method and system, relating to the field of CNC lathe technology. This system achieves high-precision machining of complex contour grooves through intelligent path planning, machining parameter optimization, and real-time vibration monitoring, solving the problem of insufficient accuracy caused by unreasonable path planning in traditional systems. The system utilizes 3D model analysis and path optimization algorithms to automatically generate the optimal machining path, improving machining accuracy. Simultaneously, through intelligent algorithms combined with real-time vibration monitoring, it dynamically adjusts cutting speed, feed rate, and depth, solving the vibration and tool wear problems caused by traditional parameter settings relying on experience, ensuring smooth machining and improved surface quality. For multiple processes, the system integrates intelligent tool changing and adaptive positioning technologies, achieving automatic workpiece identification and precise positioning, and determining the tool changing timing based on real-time monitored tool wear data, significantly improving production efficiency and machining continuity."

[0006] While the existing technical solutions have the advantages mentioned above, their disadvantages are that traditional control methods cannot adjust cutting parameters in real time when machining multi-material composite workpieces. This results in large fluctuations in cutting force and severe tool wear during the machining process, which in turn affects machining quality and efficiency.

[0007] In conclusion, developing a control method and system for machining grooves on CNC lathes remains a critical issue that urgently needs to be addressed in the field of CNC lathe technology. Summary of the Invention

[0008] The purpose of this invention is to solve the problem that traditional control methods in the prior art cannot adjust cutting parameters in real time when machining multi-material composite workpieces, resulting in large fluctuations in cutting force and severe tool wear during the machining process, which in turn affects machining quality and efficiency. This invention provides a CNC lathe groove machining control method and system.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] This invention provides a method for controlling the machining of grooves on a CNC lathe, comprising:

[0011] S1. The CNC lathe is controlled by the CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things technology.

[0012] S2. The cloud platform uses artificial intelligence algorithms to identify abnormal situations in the processing based on the various data, predict tool wear and processing quality, and generate corresponding processing suggestions.

[0013] S3. Based on the processing suggestions, the cutting parameters in the CNC system, including feed rate, depth of cut, and cutting speed, are automatically adjusted in real time.

[0014] S4. The adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The artificial intelligence algorithm is further optimized after collecting the machining results.

[0015] Furthermore, in step S1, the CNC lathe is controlled by a CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to a cloud platform for storage and analysis via Internet of Things (IoT) technology.

[0016] The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established.

[0017]

[0018] In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ;

[0019] The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference:

[0020]

[0021] In the formula, The original voltage signal from the sensor. To calibrate the cutting force,

[0022] The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning.

[0023] The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data, and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to a cloud platform for storage and analysis using Internet of Things (IoT) technology.

[0024] Furthermore, in step S2, the cloud platform utilizes artificial intelligence algorithms to identify anomalies in the machining process based on the various data, predicts tool wear and machining quality, and generates corresponding processing suggestions. The method is as follows:

[0025] The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window;

[0026] The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as:

[0027]

[0028] In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis:

[0029]

[0030] In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions.

[0031]

[0032] In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ;

[0033] The processing characteristics of the multi-material transition region are analyzed by the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

[0034] Furthermore, in step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows:

[0035] The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters:

[0036]

[0037] In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time.

[0038] Furthermore, in step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows:

[0039] The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is:

[0040]

[0041] In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model.

[0042] In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes stopping machining immediately when a serious abnormality is detected; the gradual adjustment includes adjusting the cutting parameters gradually with small steps for minor abnormalities.

[0043] Furthermore, in step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows:

[0044] After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is:

[0045]

[0046] In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force.

[0047] Furthermore, in step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows:

[0048] A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index:

[0049]

[0050] In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters:

[0051]

[0052] In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

[0053] On the other hand, the present invention also provides a CNC lathe grooving machining control system, comprising:

[0054] The data acquisition module controls the CNC lathe to perform groove machining through the CNC system, and installs various sensors on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things (IoT) technology.

[0055] The anomaly identification module utilizes the cloud platform and artificial intelligence algorithms to identify anomalies in the machining process based on various data, predict tool wear and machining quality, and generate corresponding processing suggestions.

[0056] An automatic adjustment module is used to automatically adjust the cutting parameters of the CNC system, including feed rate, depth of cut, and cutting speed, in real time according to the processing suggestions.

[0057] The feedback optimization module is used to apply the adjusted cutting parameters to the CNC system to control the CNC lathe to perform groove machining, continuously monitor the machining status, and further optimize the artificial intelligence algorithm by collecting the machining results.

[0058] Furthermore, the operation process of the data acquisition module includes:

[0059] The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established.

[0060]

[0061] In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ;

[0062] The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference:

[0063]

[0064] In the formula, The original voltage signal from the sensor. To calibrate the cutting force,

[0065] The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning.

[0066] The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to the cloud platform for storage and analysis using Internet of Things (IoT) technology.

[0067] Furthermore, the operation process of the anomaly detection module includes:

[0068] The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window;

[0069] The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as:

[0070]

[0071] In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis:

[0072]

[0073] In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions.

[0074]

[0075] In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ;

[0076] The processing characteristics of the multi-material transition region are analyzed by the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

[0077] Furthermore, the operation process of the automatic adjustment module includes:

[0078] The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters:

[0079]

[0080] In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time;

[0081] The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is:

[0082]

[0083] In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model.

[0084] In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes immediately stopping the machining when a serious abnormality is detected; the gradual adjustment includes gradually adjusting the cutting parameters in small steps for minor abnormalities.

[0085] Furthermore, the operation process of the feedback optimization module includes:

[0086] After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is:

[0087]

[0088] In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force;

[0089] A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index:

[0090]

[0091] In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters:

[0092]

[0093] In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

[0094] Beneficial effects

[0095] Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects:

[0096] When in use, this invention reduces cutting force fluctuations by adjusting cutting parameters in real time, thereby improving machining accuracy and surface quality; it also reduces tool wear and extends tool life by dynamically optimizing cutting parameters; it optimizes cutting parameters based on real-time data during machining, thereby improving machining efficiency; and by combining artificial intelligence algorithms and Internet of Things technology, it facilitates self-learning and adaptation to the machining characteristics of different materials, which helps to improve the intelligence level of the system. Attached Figure Description

[0097] Figure 1 This is a flowchart of a CNC lathe groove machining control method according to the present invention;

[0098] Figure 2 This is a system diagram of a CNC lathe groove machining control system according to the present invention. Detailed Implementation

[0099] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0100] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0101] The present invention will now be described in further detail with reference to the accompanying drawings:

[0102] Example 1:

[0103] like Figure 1 As shown, the present invention provides a method for controlling the machining of grooves on a CNC lathe, comprising:

[0104] S1. The CNC lathe is controlled by the CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things technology.

[0105] Furthermore, in step S1, the CNC lathe is controlled by a CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to a cloud platform for storage and analysis via Internet of Things (IoT) technology.

[0106] The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established.

[0107]

[0108] In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ;

[0109] The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference:

[0110]

[0111] In the formula, The original voltage signal from the sensor. To calibrate the cutting force,

[0112] The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning.

[0113] The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to the cloud platform for storage and analysis using Internet of Things (IoT) technology.

[0114] In this embodiment, the machining of the sealing groove of a new energy vehicle motor shaft with a composite structure of 45# steel shaft body and copper alloy conductive ring is taken as an example. The groove is 5mm wide and 3mm deep, with a tolerance of ±0.01mm. The present invention first obtains the CAD model of the motor shaft and the machining process requirements, and establishes a partitioning function for multiple material properties:

[0115]

[0116] Based on this, initial cutting parameters were set: 45# steel region: cutting speed 120 m / min, feed rate 0.12 mm / r; copper alloy region: cutting speed 200 m / min, feed rate 0.15 mm / r. Tool paths were generated using quintic B-spline curves. A 9-DOF piezoelectric force sensor was installed on the turret of the CNC lathe, and an infrared temperature sensor was used to monitor the cutting area. A three-axis vibration sensor was installed on the spindle. The 9-DOF piezoelectric force sensor collected three-dimensional cutting forces, and a calibration matrix was used to eliminate signal cross-interference, obtaining calibrated cutting force data. Simultaneously, real-time data was collected. Temperature and vibration data are collected and transmitted to the cloud platform at a frequency of 1000Hz via 5G IoT technology. The cloud platform is equipped with a three-layer AI algorithm: an anomaly detection model, a tool wear prediction model, and a parameter optimization model, enabling real-time data storage and intelligent analysis. This improves the adaptability of initial parameters when modeling multi-material partitions, avoiding machining deviations caused by misjudgment of material properties. It also improves the accuracy of cutting force data after calibration of the 9-DOF force sensor and shortens the anomaly detection response time. 5G transmission ensures data real-time performance, improves the pass rate of motor shaft sealing groove machining, and reduces tool wear from trial cutting.

[0117] S2. The cloud platform uses artificial intelligence algorithms to identify abnormal situations in the processing based on the various data, predict tool wear and processing quality, and generate corresponding processing suggestions.

[0118] Furthermore, in step S2, the cloud platform utilizes artificial intelligence algorithms to identify anomalies in the machining process based on the various data, predicts tool wear and machining quality, and generates corresponding processing suggestions. The method is as follows:

[0119] The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window;

[0120] The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as:

[0121]

[0122] In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis:

[0123]

[0124] In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions.

[0125]

[0126] In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ;

[0127] The processing characteristics of the multi-material transition region are analyzed using the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

[0128] In this embodiment, the machining of the tenon groove of a titanium alloy-high temperature alloy composite blade in the aerospace field is used as an example. The groove width is 12mm, the depth is 8mm, the tolerance is ±0.008mm, and the surface roughness is Ra0.6μm. This invention first targets the real-time acquisition of the three-dimensional cutting force sequence. Using a sliding window Calculate the mean Standard deviation Peak factor Coefficient of variation Construct feature vectors Input the deep learning neural network anomaly detection model, in or temperature First, it identifies anomalies in real time and triggers early warnings; second, it establishes a wear-sensitive feature set. Dimensionality reduction through kernel principal component analysis—using radial basis functions Calculate feature similarity, then... Extract key dimensionality-reduced features and input them into a tool wear prediction model based on a recurrent neural network to accurately predict the wear amount on the tool flank. The prediction error is ≤0.005mm; finally, the processing characteristics of the titanium alloy-high temperature alloy transition region are analyzed through parameter optimization model, combined with feature set. Mutations: The value was abruptly changed from 1.2 to 1.8, identifying the boundary of material change and generating processing suggestions such as "reducing the cutting speed in the transition area from 150m / min to 100m / min and adjusting the feed rate from 0.1mm / r to 0.08mm / r". This helps improve the accuracy of anomaly detection and reduce the scrap rate caused by cutting force fluctuations and high temperature. The tool wear prediction lead time is 5-8 minutes, extending tool life. It also improves the machining accuracy in multi-material transition areas, and the surface roughness is stabilized at Ra0.5μm.

[0129] S3. Based on the processing suggestions, the cutting parameters in the CNC system, including feed rate, depth of cut, and cutting speed, are automatically adjusted in real time.

[0130] Furthermore, in step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows:

[0131] The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters:

[0132]

[0133] In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time.

[0134] Furthermore, in step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows:

[0135] The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is:

[0136]

[0137] In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model.

[0138] In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes immediately stopping the machining when a serious abnormality is detected; the gradual adjustment includes gradually adjusting the cutting parameters in small steps for minor abnormalities.

[0139] In this embodiment, the machining of the bearing mounting groove of an automotive gearbox gear shaft using a 45# steel body and a portion of GCr15 bearing steel is taken as an example. The bearing mounting groove is 8mm wide and 5mm deep, with a tolerance of ±0.015mm. This invention is based on the material properties: 45# steel GCr15 steel Tool parameters: Carbide tools mm, construct a dynamic threshold model for cutting force ,in, As a correction factor, when the real-time cutting force data For example, when processing the GCr15 region Press immediately , , Adjust the feed rate and depth of cut to avoid cutting force overload; at the same time, consider the heat resistance of the tool: cemented carbide. Oxidation temperature of material: 45# steel Construct a dynamic temperature threshold After detection ,pass ,in, Optimize cutting speed and according to ,in, Increase cooling flow; in the transition zone from 45# steel to GCr15 steel, this invention automatically identifies the boundary based on the material partitioning function, switching the cutting speed from 120m / min to 100m / min and the feed rate from 0.15mm / r to 0.12mm / r; simultaneously, based on tool wear prediction results: ,according to Adjust parameters in advance and adopt a tiered control strategy—in When an emergency adjustment is triggered, processing should be stopped immediately. The process employs small-step incremental adjustments, with each adjustment not exceeding 5% of the initial value; this facilitates improving machining accuracy from ±0.02mm to ±0.008mm, stabilizing surface roughness at Ra0.8μm; extending tool life and reducing tool change frequency; improving grooving efficiency and reducing scrap rate.

[0140] S4. The adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The artificial intelligence algorithm is further optimized after collecting the machining results.

[0141] Furthermore, in step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows:

[0142] After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is:

[0143]

[0144] In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force.

[0145] Furthermore, in step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows:

[0146] A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index:

[0147]

[0148] In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters:

[0149]

[0150] In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

[0151] In this embodiment, a 0.5mm wide groove is machined in the precision mid-frame of a mobile phone made of aluminum alloy and stainless steel composite material, with a tolerance of ±0.005mm. As an example, in step S4, intelligently adjusted cutting parameters, such as a feed rate of 0.08mm / r and a cutting speed of 120m / min for the aluminum alloy area, and a feed rate of 0.05mm / r and a cutting speed of 80m / min for the stainless steel area, are applied to the CNC system to control the CNC lathe for groove machining. Simultaneously, the cutting force, temperature, and vibration are continuously monitored. After collecting the machining results, based on 200 newly acquired sets of normal machining data... The Bayesian update method is used to optimize the dynamic threshold of abnormal cutting force—the cutting force is set to follow a normal distribution. The prior distribution is ,in, , ,in, After calculating the posterior distribution, the final new threshold is taken as the 99.7% upper confidence limit of the posterior distribution. This improves the accuracy of anomaly identification; simultaneously, a processing technology knowledge base is established, and the groove width error is detected after processing according to... Calculation process capability index , ,satisfy And through surface roughness model Actual measurement The system completes quality assessments and generates comprehensive reports including tool life and machining efficiency. It stores data such as cutting force, temperature, and parameters from this machining process in a knowledge base, providing accurate references for subsequent machining of similar workpieces. Through online learning and optimization of artificial intelligence algorithms, it facilitates improved anomaly detection accuracy and reduced tool wear prediction errors.

[0152] Example 2:

[0153] like Figure 2 As shown, Embodiment 2 provides a CNC lathe grooving machining control system, including:

[0154] The data acquisition module controls the CNC lathe to perform groove machining through the CNC system, and installs various sensors on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things (IoT) technology.

[0155] The anomaly identification module utilizes the cloud platform and artificial intelligence algorithms to identify anomalies in the machining process based on various data, predict tool wear and machining quality, and generate corresponding processing suggestions.

[0156] An automatic adjustment module is used to automatically adjust the cutting parameters of the CNC system, including feed rate, depth of cut, and cutting speed, in real time according to the processing suggestions.

[0157] The feedback optimization module is used to apply the adjusted cutting parameters to the CNC system to control the CNC lathe to perform groove machining, continuously monitor the machining status, and further optimize the artificial intelligence algorithm by collecting the machining results.

[0158] Furthermore, the operation process of the data acquisition module includes:

[0159] The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established.

[0160]

[0161] In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ;

[0162] The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference:

[0163]

[0164] In the formula, The original voltage signal from the sensor. To calibrate the cutting force,

[0165] The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning.

[0166] The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to the cloud platform for storage and analysis using Internet of Things (IoT) technology.

[0167] Furthermore, the operation process of the anomaly detection module includes:

[0168] The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window;

[0169] The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as:

[0170]

[0171] In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis:

[0172]

[0173] In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions.

[0174]

[0175] In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ;

[0176] The processing characteristics of the multi-material transition region are analyzed by the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

[0177] Furthermore, the operation process of the automatic adjustment module includes:

[0178] The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters:

[0179]

[0180] In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time;

[0181] The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is:

[0182]

[0183] In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model.

[0184] In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes immediately stopping the machining when a serious abnormality is detected; the gradual adjustment includes gradually adjusting the cutting parameters in small steps for minor abnormalities.

[0185] Furthermore, the operation process of the feedback optimization module includes:

[0186] After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is:

[0187]

[0188] In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force;

[0189] A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index:

[0190]

[0191] In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters:

[0192]

[0193] In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

[0194] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling the machining of grooves on a CNC lathe, characterized in that, include: S1. The CNC lathe is controlled by the CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things technology. S2. The cloud platform uses artificial intelligence algorithms to identify abnormal situations in the processing based on the various data, predict tool wear and processing quality, and generate corresponding processing suggestions. S3. Based on the processing suggestions, the cutting parameters in the CNC system, including feed rate, depth of cut, and cutting speed, are automatically adjusted in real time. S4. The adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The artificial intelligence algorithm is further optimized after collecting the machining results.

2. The method for controlling the machining of grooves on a CNC lathe according to claim 1, characterized in that, In step S1, the CNC lathe is controlled by a CNC system to perform groove machining, and various sensors are installed on the CNC lathe to collect various data in real time. The data is then transmitted to a cloud platform for storage and analysis via Internet of Things (IoT) technology. The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established. In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ; The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference: In the formula, The original voltage signal from the sensor. To calibrate the cutting force, The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning. The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data, and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to a cloud platform for storage and analysis using Internet of Things (IoT) technology.

3. The method for controlling the machining of grooves on a CNC lathe according to claim 2, characterized in that, In step S2, the cloud platform uses artificial intelligence algorithms to identify abnormalities in the machining process based on the various data, predict tool wear and machining quality, and generate corresponding processing suggestions. The method is as follows: The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window; The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as: In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis: In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions. In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ; The processing characteristics of the multi-material transition region are analyzed by the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

4. The CNC lathe grooving control method according to claim 3, characterized in that, In step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows: The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters: In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time.

5. The method for controlling the machining of grooves on a CNC lathe according to claim 4, characterized in that, In step S3, the method for automatically adjusting the cutting parameters, including feed rate, depth of cut, and cutting speed, in the CNC system in real time according to the processing suggestion is as follows: The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is: In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model. In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes stopping machining immediately when a serious abnormality is detected; the gradual adjustment includes adjusting the cutting parameters gradually with small steps for minor abnormalities.

6. The method for controlling the machining of grooves on a CNC lathe according to claim 5, characterized in that, In step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows: After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is: In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force.

7. The method for controlling the machining of grooves on a CNC lathe according to claim 6, characterized in that, In step S4, the adjusted cutting parameters are applied to the CNC system to control the CNC lathe to perform groove machining, and the machining status is continuously monitored. The method for further optimizing the artificial intelligence algorithm by collecting machining results is as follows: A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index: In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters: In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

8. A CNC lathe grooving machining control system, based on the CNC lathe grooving machining control method according to any one of claims 1-7, characterized in that, include: The data acquisition module controls the CNC lathe to perform groove machining through the CNC system, and installs various sensors on the CNC lathe to collect various data in real time. The data is then transmitted to the cloud platform for storage and analysis through Internet of Things (IoT) technology. The anomaly identification module utilizes the cloud platform and artificial intelligence algorithms to identify anomalies in the machining process based on various data, predict tool wear and machining quality, and generate corresponding processing suggestions. An automatic adjustment module is used to automatically adjust the cutting parameters of the CNC system, including feed rate, depth of cut, and cutting speed, in real time according to the processing suggestions. The feedback optimization module is used to apply the adjusted cutting parameters to the CNC system to control the CNC lathe to perform groove machining, continuously monitor the machining status, and further optimize the artificial intelligence algorithm by collecting the machining results.

9. A CNC lathe grooving control system according to claim 8, characterized in that, The operation process of the data acquisition module includes: The grooving process includes acquiring the CAD model of the workpiece and the machining requirements, setting initial cutting parameters based on the workpiece material properties and groove shape parameters, and performing machining path planning and tool trajectory generation. For multi-material composite workpieces, a material partitioning function is established. In the formula, The material partitioning function for multi-material composite workpieces is used to describe the workpiece in spatial coordinates. The material composition and physical properties at that location, For material partitioning indicator functions, This area indicates that it is a material. Otherwise, it is 0. For the first A set of physical parameters for a material, including hardness. thermal conductivity Yield strength ; The various sensors include force sensors, temperature sensors, and vibration sensors; the various data include cutting force data, temperature data, and vibration data; the force sensor is a 9-DOF piezoelectric force sensor that collects three-dimensional cutting force data. Through calibration matrix Eliminate cross-interference: In the formula, The original voltage signal from the sensor. To calibrate the cutting force, The cloud platform includes artificial intelligence algorithms, including an anomaly detection model, a tool wear prediction model, and a parameter optimization model. The anomaly detection model uses a deep learning neural network to identify machining anomalies such as fluctuations in cutting force data, abnormal temperature data, and excessive vibration data. The tool wear prediction model is based on a recurrent neural network to predict the remaining tool life. The parameter optimization model uses a reinforcement learning algorithm to find the optimal combination of cutting parameters through trial and error learning. The CNC lathe is controlled by a CNC system to perform groove machining. Cutting force data, temperature data and vibration data during the groove machining process are collected in real time by multiple sensors. The collected data is then transmitted to the cloud platform for storage and analysis using Internet of Things (IoT) technology. The operation process of the anomaly detection module includes: The anomaly detection model in the aforementioned artificial intelligence algorithm identifies abnormal situations during the machining process, including cutting force fluctuations and excessively high temperatures; and it analyzes the real-time acquired cutting force data sequence. ,in, Indicates time The changes are respectively along , , Real-time cutting force data in the direction is obtained using a sliding window. Calculating time-domain features includes the mean. Standard deviation Peak factor and coefficient of variation Construct feature vectors of cutting force data In the formula, , The extreme value within the window; The tool wear prediction model in the aforementioned artificial intelligence algorithm predicts the tool wear state based on real-time data; it establishes a mapping relationship between cutting force, vibration, temperature, and wear amount, and defines a wear-sensitive feature set, expressed as: In the formula, It is a wear-sensitive feature set. It is the characteristic frequency of the vibration signal. It is the highest temperature in the cutting zone. Indicates time Real-time data of changing vibration acceleration. It is the energy integral of vibrational acceleration. It is the cutting speed. The fundamental frequency of the main axis Given the initial temperature, the dimensionality reduction expression is obtained through kernel principal component analysis: In the formula, These are the dimensionality reduction features after kernel principal component analysis. For vectors Find the parameter that maximizes the following expression. It is a wear-sensitive feature set exist Variance in direction, These are constraints; the kernel function uses radial basis functions. In the formula, It's a kernel function. It is a sample and The square of the Euclidean distance, ; The processing characteristics of the multi-material transition region are analyzed by the parameter optimization model in the artificial intelligence algorithm to identify the material change boundary.

10. A CNC lathe grooving control system according to claim 9, characterized in that, The operation process of the automatic adjustment module includes: The anomaly detection model detects cutting force data. Exceeding the dynamic threshold of abnormal cutting force At the same time, adjust the feed rate and depth of cut, and construct a dynamic threshold model of cutting force based on material properties and tool parameters: In the formula, Based on the threshold, For material coefficients, steel ,aluminum , Indicates time The degree of tool wear varies. For the maximum permissible wear, the wear tolerance for carbide cutting tools is 0.3 mm. For a moment Temperature of the cutting zone at that time For ambient temperature, The heat resistance limit of the material, , As a correction factor, when the real-time cutting force data When this occurs, a mechanism is triggered to automatically adjust the cutting parameters in the CNC system in real time; The anomaly detection model detects temperature data exceeding a dynamic threshold for abnormal temperatures. At the same time, optimize the cutting speed and cooling parameters; combining the tool's heat resistance and the material's oxidation temperature, the expression is: In the formula, The heat resistance temperature of the cemented carbide tool is set at 800℃. The oxidation temperature of the material. To accumulate cutting time, Indicates time The varying degree of tool wear in multi-material transition regions allows for automatic switching of cutting parameters based on material properties; and the cutting parameters are adjusted in advance based on the tool wear prediction results output by the tool wear prediction model. In addition, the adjustment of cutting parameters adopts a graded control strategy, including emergency adjustment and gradual adjustment; the emergency adjustment includes immediately stopping the machining when a serious abnormality is detected; the gradual adjustment includes gradually adjusting the cutting parameters in small steps for minor abnormalities. The operation process of the feedback optimization module includes: After collecting machining results, an online learning mechanism is used to further optimize the artificial intelligence algorithm, updating the dynamic thresholds for abnormal temperature, abnormal cutting force, and tool wear prediction model parameters; based on the newly collected data... Group normal processing data The Bayesian update method is used to update the dynamic threshold of abnormal cutting forces, and the cutting force data is set to follow a normal distribution. The prior distribution is , Then the posterior distribution is: In the formula, These are cutting force data. It's temperature data. It is the standard deviation of the cutting force data. It is the prior mean of the average cutting force. It is the prior standard deviation of the standard deviation of the cutting force. It is an inverse gamma distribution Shape parameters and scale parameters, It is the sample mean of the newly collected cutting force data, and the new threshold is taken from the posterior distribution. Confidence ceiling: In the formula, It is the updated dynamic threshold for abnormal cutting force. It is the posterior mean of the average cutting force. It is the posterior standard deviation of the standard deviation of the cutting force; A machining process knowledge base is established. After the groove machining is completed, the machining quality is inspected and evaluated. The groove width / depth error of the machined groove follows a normal distribution. Calculation process capability index: In the formula, It is a process capability index. It is the standard deviation of the groove width / groove depth error. It is a process capability index. It is the average value of the groove width / groove depth error. , For the upper and lower limits of tolerance, the requirements are: Surface roughness Relationship model with cutting parameters: In the formula, It is an empirical constant. It's the feed rate. It is the cutting speed. It is the depth of cut. This involves measuring the wear on the tool's flank face, generating a machining quality report and a tool usage report; and storing the various data and cutting parameters from this grooving process into the machining process knowledge base for reference in subsequent machining.

Images