Intelligent processing control method and electronic equipment for CNC lathe

Abstract

The present invention discloses an intelligent processing control method and electronic equipment for a CNC lathe, and relates to the field of data processing technology. The method comprises: obtaining first operating state information when the CNC lathe processes a workpiece; identifying the first operating state information through a preset vibration recognition model to obtain a first vibration type recognition result of the CNC lathe; when the first vibration type recognition result is a first abnormal vibration type, determining a processing parameter adjustment strategy according to the first abnormal vibration type; based on the processing parameter adjustment strategy and the first operating state information, determining a first adjustment parameter of the CNC lathe and a second adjustment parameter of a mechanical arm; and sending the first adjustment parameter and the second adjustment parameter to the CNC lathe and the mechanical arm, respectively. Based on the characteristics of different abnormal vibration types, the present invention specifically adjusts and controls the CNC lathe and the mechanical arm to more comprehensively control and reduce vibration, reduce processing errors caused by vibration, and improve processing accuracy.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an intelligent processing control method and electronic equipment for a CNC lathe. Background Art

[0002] With the rapid development of manufacturing industry, CNC (Computerized Numerical Control) lathes play an increasingly important role in the field of precision machining.

[0003] CNC machining is a very important machining process. The process of CNC machining using CNC lathes in related technologies is as follows: First, the robot arm is responsible for accurately placing the workpiece to be processed on the workbench of the CNC lathe to realize automatic loading and unloading and improve production efficiency. Then the CNC lathe accurately controls the movement of the tool through a preset program to ensure the accuracy and stability of the machining. The robot arm works closely with the CNC lathe to adjust the position of the workpiece in real time to ensure that the tool can penetrate into the hole for fine cutting.

[0004] However, during the CNC machining process, the lathe spindle, tool, fixture and other components will inevitably produce some vibration. Abnormal vibration will affect the machining accuracy of the CNC lathe. Summary of the invention

[0005] In view of the above-mentioned technical problems and defects, the purpose of the present invention is to provide an intelligent processing control method and electronic equipment for CNC lathes, which can identify the vibration type and adaptively suppress the vibration during CNC processing, reduce the impact of vibration on processing accuracy, and improve processing accuracy.

[0006] In order to achieve the above-mentioned purpose, in a first aspect, the present invention provides an intelligent processing control method for a CNC lathe, comprising: obtaining first operating state information of the CNC lathe when processing a workpiece, the first operating state information including at least one of lathe vibration data, sound signal, feed amount and cutting force data; identifying the first operating state information through a preset vibration recognition model to obtain a first vibration type recognition result of the CNC lathe; when the first vibration type recognition result is a first abnormal vibration type, determining a processing parameter adjustment strategy according to the first abnormal vibration type; based on the processing parameter adjustment strategy and the first operating state information, determining a first adjustment parameter of the CNC lathe and a second adjustment parameter of the robot arm, the first adjustment parameter being used for the CNC lathe itself to perform vibration control operations, and the second adjustment parameter being used for the robot arm to perform vibration control operations on the CNC lathe; and sending the first adjustment parameter and the second adjustment parameter to the CNC lathe and the robot arm, respectively.

[0007] This embodiment can fully understand the operation status of the machine tool by acquiring the first operation status information of the CNC lathe when processing the workpiece in real time. This meticulous monitoring is the premise for ensuring the stability of the processing process and predictive maintenance. By analyzing these operation status information using a preset vibration recognition model, the vibration type of the lathe can be effectively identified, especially the abnormal vibration, which is crucial for preventing machine tool failures and ensuring processing quality. When the abnormal vibration type is identified, the corresponding processing parameter adjustment strategy can be immediately determined and executed according to the specific characteristics of the abnormal vibration type. This strategy includes not only adjusting the parameters of the CNC lathe itself, but also adjusting the operating parameters of the auxiliary mechanical arm to control and reduce vibration more comprehensively and finely. By accurately adjusting the operating parameters of the CNC lathe and the mechanical arm, the processing process can be optimized, the error caused by vibration can be reduced, and the processing accuracy can be improved. This embodiment can also significantly reduce the failure rate of the machine tool, extend the service life of the machine tool and the tool, improve the processing efficiency and product quality, and reduce the production cost and maintenance cost.

[0008] In some embodiments, the vibration recognition model calculates the first abnormal vibration type through a vibration recognition formula, and the vibration recognition formula includes:

[0009] F = [f, A, S, v, P];

[0010] Z = φ(F);

[0011]

[0012] Among them, f represents the vibration frequency, A represents the vibration amplitude, S represents the characteristics of the sound signal, v represents the feed rate, P represents the tool cutting force, Z is the nonlinear mapping function of F, y is the output result of the classifier, indicating the abnormal vibration type, α i is the Lagrange multiplier obtained through training, y i is the category label of the i-th training sample, K is the kernel function used to calculate the similarity between two data points in the original feature space, b is the bias term, and N is the number of training samples.

[0013] The vibration identification formula of this embodiment is adopted, and the data in the original feature space is mapped to a higher-dimensional space by combining multi-dimensional physical parameters such as vibration frequency, amplitude, sound signal characteristics, feed rate and cutting force, so as to facilitate the application of support vector machine (SVM) algorithm to distinguish different types of vibrations. The kernel function used in the model can flexibly capture the complex relationship between data points, and the optimization of Lagrange multipliers and bias terms ensures that the model can achieve high precision on the training data while controlling the complexity of the model through regularization terms, effectively avoiding overfitting. In addition, the constraints of the model allow a certain degree of error to exist, which increases the robustness of the model. Overall, the model constructed by this formula can effectively identify and classify abnormal vibrations of CNC lathes during processing, and has significant beneficial effects on improving processing accuracy, preventing mechanical failures, extending equipment life and optimizing production efficiency.

[0014] In some embodiments, after the step of sending the first adjustment parameter and the second adjustment parameter to the CNC lathe and the robotic arm respectively, it also includes: after the CNC lathe and the robotic arm perform corresponding vibration control operations, obtaining second operating status information of the CNC lathe; identifying the second operating status information through a vibration recognition model to obtain a second vibration type recognition result of the CNC lathe; when the second vibration type recognition result is a second abnormal vibration type, generating an inspection instruction according to the second abnormal vibration type; sending an inspection instruction to the robotic arm so that the robotic arm collects inspection information of the target inspection area of ​​the CNC lathe; and judging whether the CNC lathe has a fault according to the inspection information.

[0015] By adopting this embodiment, the operation status information is collected again after the CNC lathe and the robot arm complete the vibration control operation, so as to realize continuous monitoring of the lathe status. The newly collected data is analyzed by the vibration recognition model to obtain the second vibration type recognition result. If this result indicates an abnormality, targeted inspection instructions can be generated intelligently. These instructions are then sent to the robot arm to guide it to accurately inspect the key parts of the lathe and collect information on the target inspection area. Finally, based on the inspection information collected by the robot arm, it can be determined whether the lathe has a fault, so that maintenance or adjustment measures can be taken in time. This process not only improves the accuracy and response speed of fault detection, but also reduces human errors and machine downtime, ensuring the continuity of production and processing quality.

[0016] In some embodiments, the step of generating an inspection instruction according to the second abnormal vibration type includes: determining a target inspection area and inspection items of the CNC lathe according to the second abnormal vibration type; determining a moving path of the robot arm according to the target inspection area and inspection items; and generating an inspection instruction according to the moving path.

[0017] By adopting this embodiment, the target inspection area and inspection items of the CNC lathe are accurately determined through intelligent analysis of the second abnormal vibration type, thereby realizing targeted fault diagnosis. This precise positioning greatly improves the efficiency and accuracy of the inspection. Subsequently, the system intelligently plans the optimal movement path of the robot arm based on these target information to ensure that the robot arm can reach the inspection position safely and quickly, reducing unnecessary movement and waiting time. Finally, detailed inspection instructions are automatically generated according to the planned path to guide the robot arm to perform the inspection task, further improving the level of automation and response speed. This integrated intelligent processing flow from abnormal vibration identification to path planning and then to instruction generation not only improves the operating reliability and processing accuracy of the CNC lathe, but also reduces maintenance costs and operating difficulties, providing strong technical support for intelligent manufacturing.

[0018] In some embodiments, the first operating state information also includes tool temperature and lathe current. After the step of identifying the first operating state information through a preset vibration recognition model to obtain a first vibration type recognition result of the CNC lathe, it also includes: extracting a vibration spectrum, tool temperature change, sound characteristics and lathe current waveform from the first operating state information; determining a target fault prediction model from multiple fault prediction models according to the vibration type in the first vibration type recognition result; inputting the vibration spectrum, tool temperature change, sound characteristics and lathe current waveform into the target fault prediction model to obtain the probability of failure of the CNC lathe, the fault prediction model is based on the fault prediction formula to calculate the fault probability, and the mathematical expression of the fault prediction formula is:

[0019]

[0020] X(t)=[V(t),T(t),S(t),C(t)],

[0021] Where P(t) represents the probability of failure at time t; σ is the Sigmoid function, which is used to map the model output to the interval [0,1], which is expressed as the probability of failure; β0, β1, β2, β3, β4 are model parameters; V(t), T(t), S(t), C(t) represent the vibration spectrum, tool temperature change, sound characteristics and lathe current waveform vector at time t respectively; α j is the characteristic function φ j The relevant weight; X j (t) is a time series-based feature vector, representing the jth feature input at time t; k is the number of feature functions; φ j is the jth characteristic function; ∈(t) is the random error term at time t.

[0022] By adopting this embodiment, the first operating state information of the CNC lathe is deeply analyzed, and key physical parameters are extracted from it: vibration spectrum, tool temperature change, sound characteristics and lathe current waveform, which provides a comprehensive quantitative description of the operating state of the machine tool. Combined with these parameters, the system can select the most suitable model from multiple preset fault prediction models to predict the fault probability according to the identified vibration type. This method not only improves the accuracy of fault prediction, but also enhances the adaptability and flexibility of the system. After the extracted features are input into the target model, the system can calculate the probability of failure of the lathe and provide forward-looking maintenance suggestions for the operator. Among them, using the fault prediction formula, the model can capture the complex relationship between the input features and the fault probability, which is conducive to the output of the model to reflect the fault probability as close to the real as possible. Such an intelligent prediction mechanism makes preventive maintenance possible, effectively avoids unexpected downtime, reduces maintenance costs, and improves production efficiency and processing quality.

[0023] In some embodiments, before the step of obtaining the first operating status information when the CNC lathe processes the workpiece, it also includes: determining the expected pressure information of the workstation on the machine according to the processing progress of the CNC lathe, the workstation is used to place the workpiece, and a pressure sensor is provided on the workstation; judging whether the processing flow of the workpiece is normal based on the current workstation pressure information and the expected pressure information detected by the pressure sensor.

[0024] By adopting this embodiment, accurate monitoring of the workpiece placement status is achieved by monitoring the processing progress of the CNC lathe and determining the expected pressure information of the workstation on the machine. Using the pressure sensor configured on the workstation, the system can detect the pressure information of the current workstation in real time and compare it with the expected value. This comparison provides a direct and effective method to determine whether the workpiece has problems such as improper placement, missing or over-placement, thereby ensuring the normal progress of the processing flow. Timely detection and correction of these potential problems can avoid processing errors, reduce material waste, prevent damage to machine tools and workpieces, and improve processing efficiency and product quality. In addition, this intelligent monitoring mechanism also helps to achieve preventive maintenance, reduce emergency downtime, reduce operating costs, and enhance production flexibility and the ability to respond to market changes.

[0025] In some embodiments, before the step of obtaining the first operating status information when the CNC lathe processes the workpiece, it also includes: obtaining the workpiece placement image of the machine; identifying the workpiece placement image through computer vision technology to obtain the number and placement position of the workpiece on the machine; and judging whether the processing flow of the workpiece is normal based on the processing progress of the CNC lathe, and the number and placement position of the workpiece on the machine.

[0026] By adopting this embodiment, by acquiring the workpiece placement image of the machine and applying computer vision technology, the automatic detection of the number and position of the workpieces on the machine is realized. This automated image recognition technology can accurately identify the workpiece status at each workstation, including whether the workpiece exists, whether the position is accurate, and whether the quantity is consistent with the processing progress. Combined with the real-time processing progress information of the CNC lathe, the system can promptly determine whether the current processing flow is normal. Once an abnormal workpiece placement is found, such as missing workpieces, wrong position or inconsistent quantity, the system will quickly issue an alarm and provide feedback, prompting the operator to intervene, thereby avoiding potential processing errors and machine tool failures. This method significantly improves the efficiency and accuracy of production monitoring, reduces the need for manual inspection, improves the automation level and response speed of the production line, and ensures the continuity of the processing process and the stability of the processing quality.

[0027] In a second aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0028] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on an electronic device, enables the electronic device to execute the method described in the first aspect and any possible implementation of the first aspect.

[0029] In a fourth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on an electronic device, enables the electronic device to execute the method described in the first aspect and any possible implementation of the first aspect.

[0030] It is understandable that the electronic device provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0031] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0032] 1. By acquiring the first operating status information of the CNC lathe when processing the workpiece in real time, the operating status of the machine tool can be fully understood. This meticulous monitoring is the premise for ensuring the stability of the processing process and predictive maintenance. By analyzing these operating status information using the preset vibration recognition model, the vibration type of the lathe can be effectively identified, especially the abnormal vibration, which is crucial for preventing machine tool failures and ensuring processing quality. When the system identifies the abnormal vibration type, it can immediately determine and execute the corresponding processing parameter adjustment strategy based on the specific characteristics of the type. This strategy includes not only adjusting the parameters of the CNC lathe itself, but also adjusting the operating parameters of the auxiliary mechanical arm to more comprehensively control and reduce vibration. By accurately adjusting the operating parameters of the CNC lathe and the mechanical arm, the processing process can be optimized, the error caused by vibration can be reduced, and the processing accuracy can be improved. This embodiment can significantly reduce the failure rate of the machine tool, extend the service life of the machine tool and the tool, improve the processing efficiency and product quality, and reduce the production cost and maintenance cost.

[0033] 2. By collecting the operating status information again after the CNC lathe and the robotic arm complete the vibration control operation, continuous monitoring of the lathe status is achieved. The newly collected data is analyzed using the vibration recognition model to obtain the second vibration type recognition result. If this result indicates an abnormality, targeted inspection instructions can be intelligently generated. These instructions are then sent to the robotic arm to guide it to accurately inspect the key parts of the lathe and collect information on the target inspection area. Finally, based on the inspection information collected by the robotic arm, it can be determined whether the lathe has a fault, so that maintenance or adjustment measures can be taken in time. This process not only improves the accuracy and response speed of fault detection, but also reduces human errors and machine downtime, ensuring production continuity and processing quality.

[0034] 3. Through intelligent analysis of the second abnormal vibration type, the target inspection area and inspection items of the CNC lathe are accurately determined, thereby achieving targeted fault diagnosis. This precise positioning greatly improves the efficiency and accuracy of the inspection. Subsequently, the system intelligently plans the optimal movement path of the robot arm based on these target information to ensure that the robot arm can reach the inspection position safely and quickly, reducing unnecessary movement and waiting time. Finally, detailed inspection instructions are automatically generated according to the planned path to guide the robot arm to perform inspection tasks, further improving the level of automation and response speed. This integrated intelligent processing flow from abnormal vibration identification to path planning and then to instruction generation not only improves the operating reliability and processing accuracy of the CNC lathe, but also reduces maintenance costs and operating difficulties, providing strong technical support for intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings herein are incorporated into and constitute a part of the specification, showing embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0036] Figure 1 It is an architecture diagram of the intelligent processing system in the embodiment of the present application;

[0037] Figure 2 is a flow chart of an intelligent processing control method according to an embodiment of the present application;

[0038] Figure 3 is a flow chart of another intelligent processing control method according to an embodiment of the present application;

[0039] Figure 4 It is a schematic diagram of the architecture of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0040] The terms used in the following embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to be limiting of the present invention. As used in the specification of the present invention, the singular expressions "a", "a", "above", "the" and "this" are intended to also include plural expressions, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present invention refers to any or all possible combinations comprising one or more of the listed items.

[0041] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "plurality" is two or more.

[0042] In the CNC machining process of CNC lathes, vibration is a factor that cannot be ignored, and its influence on machining accuracy is crucial. First, vibration will directly affect the cutting performance of the tool, causing fluctuations in cutting force, which in turn affects machining accuracy. Secondly, vibration will change the contact state between the tool and the workpiece, causing deviation during the cutting process, thus affecting machining accuracy. In addition, vibration will also cause the surface quality of the workpiece to deteriorate, affecting machining accuracy.

[0043] Therefore, the embodiment of the present invention provides an intelligent processing control method for a CNC lathe, which can fully understand the operating status of the machine tool by acquiring the first operating state information of the CNC lathe when processing a workpiece in real time. This meticulous monitoring is a prerequisite for ensuring the stability of the processing process and predictive maintenance. By analyzing these operating state information using a preset vibration recognition model, the vibration type of the lathe can be effectively identified, especially abnormal vibration, which is crucial for preventing machine tool failures and ensuring processing quality. When the system identifies the abnormal vibration type, it can immediately determine and execute the corresponding processing parameter adjustment strategy based on the specific characteristics of the type. This strategy includes not only adjusting the parameters of the CNC lathe itself, but also adjusting the operating parameters of the auxiliary mechanical arm to more comprehensively control and reduce vibration. By accurately adjusting the operating parameters of the CNC lathe and the mechanical arm, the processing process can be optimized, the error caused by vibration can be reduced, and the processing accuracy and surface finish can be improved. This embodiment can significantly reduce the failure rate of the machine tool, extend the service life of the machine tool and the tool, improve the processing efficiency and product quality, and reduce the production cost and maintenance cost.

[0044] The intelligent processing control method of this embodiment is executed by the control end of the intelligent processing system. Figure 1 As shown, the intelligent processing system also includes a CNC lathe and a robotic arm, and the control end is communicatively connected with the CNC lathe and the robotic arm respectively.

[0045] In this embodiment, the control end is the core of the entire CNC lathe intelligent processing system, responsible for coordinating and executing various complex control tasks. The control end is usually composed of one or more high-performance industrial computers, which are equipped with professional software systems to receive and process data from machine tools and sensors, and execute algorithms such as vibration identification, fault prediction, and parameter optimization. The control end has powerful data processing capabilities, can analyze the operating status of the machine tool in real time, and make rapid decision responses.

[0046] The software system at the control end usually includes components such as user interface (UI), database management system, data analysis module, machine learning model, communication interface and security protocol. The user interface enables operators to easily monitor the status of machine tools, adjust processing parameters, view alarm information and historical records. The database management system is responsible for storing a large amount of processing data and historical maintenance records. The data analysis module uses advanced statistical methods and machine learning algorithms to extract valuable information from the data for optimizing the processing process and predicting potential failures.

[0047] The control end also has the ability to integrate with other enterprise management software such as cloud platforms, MES systems, ERP systems, etc., to achieve seamless data flow and resource sharing. In addition, the control end is also responsible for communicating with other automation equipment such as robotic arms, automatic tool changing systems, and workpiece conveying systems to achieve collaborative operations of the entire processing line.

[0048] In terms of security, the control end adopts multi-level security measures, including physical security, network security, data security and application security, to ensure the stable operation of the system and the security of data. Through the powerful functions of the control end, the CNC lathe of this solution can achieve highly automated and intelligent processing, significantly improve production efficiency and processing quality, and reduce labor costs and equipment failure risks.

[0049] The robot arm in this embodiment can be a multi-axis robot arm, such as a four-axis robot arm, a six-axis robot arm, an eight-axis robot arm and other multi-axis robots. Taking the six-axis robot arm as an example, the six-axis robot arm is a highly flexible automation device with six rotational degrees of freedom, which can simulate the complex movements of human arms. Each axis represents a rotational joint, which together allow the robot arm to perform a variety of operations in three-dimensional space. The six-axis robot arm usually includes a base, a shoulder, an elbow, a wrist and an end effector, such as a gripper or a tool head. This design enables the six-axis robot arm to perform precise positioning tasks, such as grasping, placing, assembling, welding, spraying and other complex operations. In addition, the advanced control system of the six-axis robot arm can receive real-time data, execute pre-programmed commands, or respond to real-time instructions to perform adaptive motion adjustments. In the intelligent processing system of the CNC lathe, the six-axis robot arm can not only be used for automatic tool replacement and rapid loading and unloading of workpieces, but also for precise inspection and maintenance work, significantly improving production efficiency and processing accuracy, while reducing manual intervention and improving the safety of operation.

[0050] The CNC lathe in this embodiment is an advanced computer numerical control machine tool, which is specially designed for efficient and accurate processing of rotating bodies. The machine tool integrates a high-performance control system that can interpret complex program codes and control the machine tool to perform a variety of precision machining operations, such as turning, milling, drilling and tapping. The CNC lathe is equipped with precision servo motors and drivers to ensure that the relative movement between the tool and the workpiece is accurate and correct. In addition, the machine tool is equipped with a variety of sensors to monitor the processing status in real time, including vibration, sound, temperature and cutting force, etc., which are crucial for the intelligent processing system. The design of the CNC lathe takes into account rigidity, precision and durability, and is suitable for mass production and precision processing. In this solution, the CNC lathe realizes real-time monitoring, fault prediction, adaptive parameter adjustment and automated maintenance of the processing process through integration with the intelligent processing system, which significantly improves processing efficiency and product quality, while reducing operating costs and equipment downtime.

[0051] Combine the following Figure 2 The intelligent processing control method of this embodiment is specifically introduced, and the method specifically comprises the following steps:

[0052] Step 101, obtaining first operating state information of a CNC lathe when processing a workpiece.

[0053] The first operation status information includes at least one of lathe vibration data, sound signal, feed rate and cutting force data. The feed rate refers to the displacement of the tool of the CNC lathe relative to the workpiece in the feed motion direction.

[0054] Specifically, the first operating status information can be obtained through various sensors installed on the CNC lathe, which can monitor and record key parameters of the lathe during the processing in real time. Specifically, the following sensors may be included:

[0055] Vibration sensor: used to measure the vibration of the lathe during the processing. The vibration data can reflect the stability of the lathe and the abnormal vibration that may occur during the processing.

[0056] Sound sensor: By capturing the sound signals generated when the lathe is working, the noise characteristics of the processing process can be analyzed. Abnormal sounds may indicate potential problems.

[0057] Torque sensor: Installed on the cutting tool or spindle to measure the cutting force, which helps to understand the load conditions during the cutting process.

[0058] Encoder: Monitors the lathe's feed and spindle speed, providing precise information about the lathe's motion control.

[0059] The data collected by these sensors, combined with advanced signal processing technology and data analysis algorithms, can accurately obtain the first operating status information of the CNC lathe, providing a basis for subsequent vibration type identification and parameter adjustment. Real-time acquisition of this information is crucial to ensure the stability of the machining process and improve machining quality.

[0060] Step 102: Identify first operating state information using a preset vibration identification model to obtain a first vibration type identification result of the CNC lathe.

[0061] The first vibration type recognition result includes whether the current vibration is normal or abnormal, and if it is abnormal, which type of abnormal vibration it is.

[0062] In this embodiment, the vibration types include normal vibration and abnormal vibration. Normal vibration is the normal vibration generated when the CNC lathe is working, which generally does not cause too much impact on CNC machining. Abnormal vibration includes but is not limited to forced vibration, self-excited vibration and resonance.

[0063] Forced vibration is a vibration caused by an external periodic force. In the CNC machining system process, this vibration may be caused by uneven cutting forces, defects in the machine tool structure, or dynamic interaction between the tool and the workpiece.

[0064] Self-excited vibration is a type of vibration that does not require external excitation to continue. In CNC lathes, self-excited vibration can be caused by instabilities in the cutting process, such as improper cutting parameter settings or tool wear. This type of vibration is characterized by some part of the machine tool starting to vibrate, and the vibration amplitude increases over time until it reaches a steady state. Self-excited vibration can lead to deterioration of the machined surface quality or even damage to the tool.

[0065] Resonance is the phenomenon that the vibration amplitude of a system increases dramatically when the frequency of the external excitation matches the natural frequency of the system. In CNC lathes, resonance can be caused by the natural frequency of the machine tool components matching the operating frequency. In resonance, even a small external excitation can cause huge vibrations, which can cause deformation of the machine tool structure, tool breakage or reduced machining accuracy.

[0066] The control end uses advanced data analysis and machine learning technology through the vibration recognition model to process and interpret the data collected from the lathe vibration sensor. The vibration data contains rich information such as the frequency, amplitude and pattern of vibration, which is crucial for identifying the operating status of the lathe.

[0067] The vibration recognition model first preprocesses the collected vibration signals, including filtering and denoising, to improve the quality of the signals. Subsequently, through feature extraction techniques, the model extracts key features from the signals, such as peak frequency, energy distribution, and time-frequency characteristics. These features are then input into a trained classifier, such as a support vector machine (SVM), neural network, or random forest, to distinguish between different vibration types.

[0068] The classifier matches the real-time vibration data with the known vibration types based on the learned vibration patterns to identify the current vibration type. This may include normal vibration, forced vibration, self-excited vibration, or resonance, etc. Ultimately, the model outputs a vibration type identification result, indicating whether the current operation status of the lathe is normal, or whether there is abnormal vibration that may affect the processing quality.

[0069] This process not only improves the monitoring accuracy of the lathe's operating status, but also enables the system to adjust processing parameters in a timely manner and optimize the processing process through real-time feedback, thereby improving processing efficiency and product quality while reducing potential risks of mechanical damage.

[0070] In this embodiment, the construction and training of the vibration recognition model is a complex process involving multiple steps, including data collection, preprocessing, feature extraction, model selection, training and verification. The following is a detailed description of the various stages of this process:

[0071] 1. Data collection: collect the vibration signals of the target lathe in different vibration types in time series, including sound signals, feed rate, cutting force, etc. under different vibration types. Each vibration signal, sound signal, feed rate, and cutting force data has a corresponding vibration type label.

[0072] 2. Data Preprocessing

[0073] Filtering: Removing noise and unwanted frequency components, such as using low-pass, high-pass, or band-pass filters.

[0074] Denoising: Use methods such as wavelet denoising to reduce random noise in the signal.

[0075] Normalization: Scaling the data to a uniform range, such as 0 to 1 or -1 to 1, to improve the generalization ability of the model.

[0076] 3. Feature extraction

[0077] Time domain analysis: Calculate the statistical characteristics of the signal, such as mean, variance, skewness, peak, etc.

[0078] Frequency domain analysis: Convert the signal from the time domain to the frequency domain through Fourier transform and analyze the frequency components, such as power spectral density.

[0079] Time-frequency analysis: Use short-time Fourier transform or wavelet transform to analyze the time-frequency characteristics of the signal.

[0080] Feature selection: Select the most representative and discriminative features to reduce the complexity of the model.

[0081] 4. Model selection

[0082] Classifier selection: Depending on the nature of the problem and the characteristics of the data, choose an appropriate machine learning algorithm, such as support vector machine (SVM), random forest, neural network, k-nearest neighbor (k-NN), etc.

[0083] Model architecture: For neural networks, design the network structure, including the number of layers, number of neurons, and activation functions.

[0084] 5. Training

[0085] Training set division: The collected data is divided into training set and test set, and cross-validation is usually used to improve the generalization ability of the model.

[0086] Loss Function: Define the loss function, which measures the difference between the model prediction and the actual value.

[0087] Optimization algorithm: Select an optimization algorithm, such as gradient descent, Adam, etc., to minimize the loss function.

[0088] Hyperparameter tuning: Adjust the model’s hyperparameters, such as learning rate, regularization parameters, etc., to achieve the best performance.

[0089] 6. Validation and test model evaluation: Use the test set to evaluate the performance of the model. Commonly used indicators include accuracy, recall, F1 score, etc.

[0090] Performance Analysis: Analyze the model’s performance on different types of vibration signals and identify the model’s strengths and limitations.

[0091] Model improvement: Based on the validation results, adjust the model structure or parameters and perform multiple iterations until satisfactory performance is achieved.

[0092] 7. Deployment

[0093] System integration: Integrate the trained model into the intelligent control system of the CNC lathe.

[0094] Real-time monitoring: The model receives vibration data in real time and identifies the vibration type.

[0095] Feedback control: Based on the recognition results, the system automatically adjusts the processing parameters to achieve closed-loop control.

[0096] The construction and training of vibration recognition models is an iterative process that requires continuous adjustment and optimization to adapt to various situations in practical applications. In addition, the interpretability and robustness of the model are also important factors to be considered during design.

[0097] In some embodiments, the vibration recognition model calculates the first abnormal vibration type through a vibration recognition formula, and the vibration recognition formula includes:

[0098] F = [f, A, S, v, P];

[0099] Z = φ(F);

[0100]

[0101] Among them, f represents the vibration frequency, A represents the vibration amplitude, S represents the characteristics of the sound signal, v represents the feed rate, P represents the tool cutting force, Z is the nonlinear mapping function of F, y is the output result of the classifier, indicating the abnormal vibration type, α i is the Lagrange multiplier obtained through training, y i is the category label of the i-th training sample, K is the kernel function used to calculate the similarity between two data points in the original feature space, b is the bias term, and N is the number of training samples.

[0102] Furthermore, a regularization term Ω is introduced to prevent the model from overfitting:

[0103]

[0104] where γ is the regularization parameter.

[0105] Next, train the model by solving the following optimization problem:

[0106]

[0107]

[0108] Among them, C is the penalty parameter, ξi is the slack variable, and j is an iteration variable used to traverse all training samples in the optimization problem in order to calculate the decision function and satisfy the constraints of the support vector machine.

[0109] The goal of the optimization problem is to find the optimal Lagrange multiplier α and bias term b to minimize the error of the model on the training set, while controlling the model complexity through the regularization term Ω(α) to avoid overfitting. The slack variable ξi is used as a constraint to ensure that the model is consistent in the classification of training samples, while allowing some training samples to have small errors.

[0110] The following example illustrates the application process of the vibration recognition model in this embodiment:

[0111] Assume the following conditions and data:

[0112] A series of vibration data of CNC lathes under different vibration conditions were collected, and each sample included vibration frequency (f), vibration amplitude (A), sound signal characteristics (S), feed rate (v) and cutting force (P).

[0113] The dataset is labeled into four vibration types: normal vibration (M1), forced vibration (M2), self-excited vibration (M3), and resonance (M4).

[0114] Train the SVM model:

[0115] A SVM classifier is trained using the collected data. During the training process, a suitable kernel function is selected, such as a radial basis function (RBF) kernel.

[0116] Solve the optimization problem and find the optimal Lagrange multiplier (α) and bias term (b).

[0117] Formula application:

[0118] For a new vibration sample F′=[f′,A′,S′,v′,P′], firstly map it to a high-dimensional space through a nonlinear mapping function φ, and obtain Z′=φ(F′). Substitute Z′ into the decision function and calculate y′:

[0119]

[0120] Classification judgment:

[0121] Assume that the calculation result of y′ corresponds to the set of labels in our training data, for example:

[0122] If 'y' corresponds to the label "M1", it is determined that the current vibration type is normal vibration.

[0123] If y′ corresponds to the label “M2”, the current vibration type is determined to be forced vibration.

[0124] If y′ corresponds to the label “M3”, the current vibration type is determined to be self-excited vibration.

[0125] If y′ corresponds to the label “M4”, the current vibration type is determined to be resonance.

[0126] Example scenario:

[0127] Assume that a new vibration data sample F' is obtained, and y'=-1 is calculated by the SVM model (in practical applications, y' will be a specific class label or the distance of the decision boundary). According to the trained model, this may mean that the sample belongs to self-excited vibration (M3).

[0128] Practical Application:

[0129] In actual application, the vibration recognition model will receive the vibration data of the CNC lathe in real time and quickly identify the vibration type through the above steps. Once the abnormal vibration type is identified, the system will immediately take corresponding adjustment strategies, such as adjusting cutting parameters or machine tool settings, to maintain the stability and efficiency of the machining process.

[0130] The vibration identification formula of this embodiment combines multi-dimensional physical parameters such as vibration frequency, amplitude, sound signal characteristics, feed rate and cutting force, and uses a nonlinear mapping function to map the data in the original feature space to a higher-dimensional space, so as to facilitate the application of the support vector machine (SVM) algorithm to distinguish different types of vibrations. The kernel function used in the model can flexibly capture the complex relationship between data points, and the optimization of Lagrange multipliers and bias terms ensures that the model can achieve high precision on the training data while controlling the complexity of the model through the regularization term, effectively avoiding overfitting. In addition, the constraints of the model allow a certain degree of error to exist, which increases the robustness of the model. Overall, the model constructed by this formula can effectively identify and classify abnormal vibrations of CNC lathes during processing, and has significant beneficial effects on improving processing accuracy, preventing mechanical failures, extending equipment life, and optimizing production efficiency.

[0131] Step 103: When the first vibration type identification result is the first abnormal vibration type, a processing parameter adjustment strategy is determined according to the first abnormal vibration type.

[0132] The first abnormal vibration type may be one of forced vibration, self-excited vibration and resonance.

[0133] Specifically, the control end will determine the corresponding machining parameter adjustment strategy based on the specific characteristics and properties of this abnormal vibration type. This may include adjusting the cutting speed, feed rate, cutting depth or geometric parameters of the tool to reduce or eliminate the impact of abnormal vibration. Through real-time feedback and intelligent adjustment, the strategy aims to optimize the machining process and ensure machining quality, while protecting the machine tool and tool to avoid damage or failure due to abnormal vibration, thereby improving overall production efficiency and machining safety.

[0134] For example, assuming that the first abnormal vibration type is forced vibration, the corresponding processing parameter adjustment strategy is as follows:

[0135] Cutting speed: Adjust the cutting speed to avoid matching the natural frequency of the machine tool and reduce the impact of forced vibration.

[0136] Feed rate: Change the feed rate to change the periodic variation of cutting force and reduce forced vibration.

[0137] Tool stiffness: Use a more rigid tool or adjust the tool clamping method to increase the stiffness of the tool system and resist the influence of external coercive forces.

[0138] When the first abnormal vibration type is self-excited vibration, the corresponding processing parameter adjustment strategy is as follows:

[0139] Cutting parameters: Reduce cutting speed, feed rate or cutting depth to reduce cutting force, thereby reducing the possibility of self-excited vibration.

[0140] Tool wear monitoring: Implement tool wear monitoring and replace worn tools in time to avoid self-excited vibration caused by tool wear.

[0141] Cutting fluid: Use cutting fluid to improve cutting conditions, reduce friction and heat, and reduce vibration.

[0142] When the first abnormal vibration type is resonance, the corresponding processing parameter adjustment strategy is as follows:

[0143] Machine tool structure optimization: Optimize the machine tool structure, increase its natural frequency, and avoid resonance caused by being close to the operating frequency.

[0144] Damping treatment: Add dampers or use damping materials to reduce the vibration amplitude during resonance.

[0145] Dynamic stiffness adjustment: Adjust the dynamic stiffness of the machine tool, such as by adjusting bearing preload or using adjustable supports.

[0146] This embodiment takes into account that the causes and influencing mechanisms behind each type of abnormal vibration are different. For example, forced vibration is usually caused by periodic external forces, and it may be necessary to avoid matching the frequency of the excitation source by changing the cutting speed or feed rate; self-excited vibration is often related to the instability of the cutting process, and it may require reducing cutting parameters or replacing worn tools to restore stability; and resonance is caused by the natural frequency of the machine tool being close to the excitation frequency, which may need to be solved through structural optimization, adjusting the lathe counterweight or increasing damping.

[0147] Therefore, only through customized adjustment strategies can various vibration problems be effectively solved, the stability of the processing process and product quality be ensured, while minimizing damage to machine tools and cutting tools, and improving production efficiency and economic benefits. This embodiment customizes the processing parameter adjustment strategy for different types of abnormal vibrations, so that it can accurately identify and respond to various abnormal vibrations encountered by CNC lathes during processing. This targeted adjustment not only improves the accuracy and efficiency of vibration control, but also enhances the stability and safety of the processing process.

[0148] It should be noted that when implementing the above adjustment strategy, various factors such as the specific conditions of the lathe and tool, the material characteristics of the workpiece, and the processing environment can also be considered. The formulation of the adjustment strategy should be based on an in-depth analysis of the vibration identification results, combined with actual processing experience and machine tool performance. In addition, when adjusting the processing parameters, other possible influences such as tool life, processing efficiency, and processing costs should also be considered to achieve the best comprehensive benefits.

[0149] Step 104 : determining a first adjustment parameter of the CNC lathe and a second adjustment parameter of the robot arm based on the machining parameter adjustment strategy and the first operating state information.

[0150] The first adjustment parameter is used for the CNC lathe itself to perform vibration control operations, and the second adjustment parameter is used for the robot arm to perform vibration control operations on the CNC lathe.

[0151] Specifically, the first adjustment parameter directly acts on the CNC lathe itself, which may include but is not limited to cutting speed, feed rate, cutting depth, cutting force, etc. The adjustment of these parameters directly affects the mechanical and thermodynamic behaviors during the machining process, thereby having a direct impact on the vibration. For example, when the system identifies forced vibration, it may be necessary to reduce the impact of periodic external forces on the machine tool by reducing the cutting speed or adjusting the feed rate. In the face of self-excited vibration, it may be necessary to optimize the combination of cutting depth and speed to avoid instability during the cutting process.

[0152] Secondly, the second adjustment parameter is used to guide the robot arm to perform vibration control operations. The robot arm can know the current state and vibration characteristics of the CNC lathe according to the second adjustment parameter and dynamically adjust its support or intervention strategy.

[0153] For example, when a lathe resonance is detected, the robot arm can change the counterweight on the CNC lathe to adjust the dynamic balance of the machine tool. Proper counterweight helps reduce vibration caused by imbalance and improves the stability of the machining process. The robot arm can also reduce vibration caused by improper positioning of the workpiece by fine-tuning the position of the workpiece in the fixture. Correct fixture position helps ensure even distribution of cutting forces, thereby reducing vibration. The robot arm can also adjust the parameters of the lathe's attached damper to increase the damping ratio and help absorb and reduce vibration energy. Especially in the case of resonance, proper damping can significantly reduce the amplitude of vibration.

[0154] In some embodiments, the robot arm will also automatically identify the need for tool replacement based on the second adjustment parameter, and perform a series of precise actions to replace the tool of the lathe. First, the robot arm receives the second adjustment parameter sent by the control end, which indicates the wear state of the current tool or the abnormal vibration during the processing, indicating that the tool needs to be replaced to maintain the processing accuracy and efficiency. Subsequently, the robot arm moves to the CNC lathe tool position, uses its precise clamping device to safely unload the current tool, and places it in the designated storage location. Next, the robot arm selects a suitable new tool from the tool library, performs precise loading and calibration to ensure the correct position and cutting parameters of the new tool. This automated tool replacement process not only improves the safety and efficiency of the operation, but also reduces the downtime of the machine tool, enhances the intelligence level and adaptability of the entire processing system, and ensures the continuity of the processing process and product quality.

[0155] When determining the first adjustment parameter and the second adjustment parameter, multiple factors need to be considered comprehensively, including but not limited to the current vibration type, vibration intensity, characteristics of the processed material, the status of the tool, the structural characteristics of the machine tool, etc. Through advanced data fusion technology and intelligent algorithms, the system can comprehensively analyze these complex information to obtain the optimal adjustment strategy.

[0156] In addition, this dynamic adjustment strategy based on real-time operating status information can not only improve the stability and quality of the machining process, but also extend the service life of machine tools and cutting tools and reduce maintenance costs. At the same time, it can also improve production efficiency and reduce downtime caused by vibration problems. Ultimately, this intelligent vibration control solution is an important step in realizing intelligent manufacturing and improving the level of production automation, and is of great significance to improving the competitiveness of the entire manufacturing industry.

[0157] Step 105: Send the first adjustment parameter and the second adjustment parameter to the CNC lathe and the robot arm respectively.

[0158] After comprehensively analyzing the first operating state information and the output of the vibration identification model, the control end generates two sets of fine adjustment parameters: the first adjustment parameters and the second adjustment parameters. The first adjustment parameters are specifically used to directly adjust the internal processing parameters of the CNC lathe, such as cutting speed, feed rate and cutting depth, to optimize the lathe's own vibration control and processing performance. The second adjustment parameters are designed for the cooperating robot arm to guide it to perform operations such as adjusting the counterweight, optimizing the fixture position, adjusting the damping, dynamic support, etc., to assist the lathe in reducing vibration and improving the overall processing stability.

[0159] Subsequently, the control end sends these two sets of parameters to the control systems of the CNC lathe and the robotic arm respectively in a wireless or wired manner through the internal communication network or external interface. In this way, the CNC lathe and the robotic arm can work together and make intelligent adjustments based on real-time feedback to ensure the efficiency, stability and accuracy of the processing process, thereby improving product quality and production efficiency.

[0160] After receiving the first adjustment parameter and the second adjustment parameter from the control end, the CNC lathe and the robot arm will respectively execute their own adjustment strategies: the CNC lathe will optimize the internal processing parameters according to the first adjustment parameter, which is the parameter directly related to the machine tool processing behavior, such as cutting speed, feed rate and cutting depth, to improve the processing process and reduce the vibration caused by the machine tool itself. At the same time, the robot arm will perform auxiliary adjustments to the CNC lathe according to the second adjustment parameter, such as adjusting the counterweight, changing the fixture position or adjusting the damping, etc., to provide additional stability support and vibration control.

[0161] This collaborative execution strategy enables the CNC lathe and robotic arm to accurately respond to the output of the vibration identification model and achieve intelligent optimization during the machining process, thereby improving machining accuracy, ensuring product quality, and enhancing the stability and adaptability of the entire machining system.

[0162] In order to explain the intelligent processing control method of the embodiment of the present invention in more detail, Figure 3 To further illustrate, the method of this embodiment may further specifically include the following steps:

[0163] Step 201, determining whether the processing flow of the workpiece is normal.

[0164] In some embodiments, the method specifically includes the following steps:

[0165] First, according to the processing progress of the CNC lathe, the expected pressure information of the workstation on the machine is determined. The workstation is used to place the workpiece and is equipped with a pressure sensor.

[0166] Specifically, assume that the machine has 100 workstations, which are numbered from 1 to 100, and each workstation is placed with a workpiece, and the processing order is in the order from 1 to 100. When the nth workpiece is processed, according to the processing schedule of the CNC lathe, the 1st to nth workstations should all be empty, and the pressure sensors on the workstations should not detect the pressure from the workpiece; that is, the expected pressure information is that the expected pressure of the 1st to nth workstations is 0, or less than a certain set pressure threshold.

[0167] Then, based on the current workstation pressure information and expected pressure information detected by the pressure sensor, it is determined whether the processing flow of the workpiece is normal.

[0168] For example, the current CNC lathe processing progress is to process the 16th workpiece. Then, on the machine, under normal circumstances, the pressure sensors at stations 1 to 16 should not detect the gravity pressure of the workpiece. If, at this time, the pressure sensor at station 15 detects pressure, it means that there is a workpiece to be processed at station 15, and the workpiece at station 15 has been missed. This indicates that the processing flow is abnormal. At this time, the control end can stop the work of the CNC lathe and the robot arm, and issue an alarm message to prompt the staff to check and adjust the processing flow.

[0169] Otherwise, it is determined that the processing flow of the workpiece is normal, and the process goes to step 202 .

[0170] In some embodiments, this step may also specifically include:

[0171] (1) Obtain the workpiece placement image of the machine.

[0172] In the automated production process of CNC lathes, real-time monitoring of the placement of workpieces on the machine is essential to ensure the accuracy of the machining process. To this end, a high-definition camera installed in the machining area can be used to capture images on the machine in real time. These images not only reflect the immediate placement of the workpiece, but also provide the necessary visual data for subsequent automated analysis. Image acquisition is the first step in the automated monitoring system, which requires the camera to operate stably under various working conditions, ensure the clarity and real-time nature of the image, and lay a solid foundation for subsequent workpiece identification and position verification.

[0173] (2) Use computer vision technology to identify the workpiece placement image and obtain the number and placement of the workpieces on the machine.

[0174] The acquired workpiece placement images are then input into a system equipped with advanced computer vision technology. The system uses machine learning and pattern recognition algorithms to conduct in-depth analysis of the images to identify the number of workpieces and their specific placement on the machine. Computer vision technology can process complex image information, including different lighting conditions, the diversity of workpieces, and possible occlusions, so as to accurately identify the outline, size, and position of the workpieces. This step is the core link in realizing automated monitoring of the processing process, and it provides accurate data support for subsequent process judgment and decision-making.

[0175] (3) Determine whether the processing flow of the workpiece is normal based on the processing progress of the CNC lathe, as well as the number and placement of the workpieces on the machine.

[0176] After obtaining the accurate quantity and location information of the workpieces, these data are compared and analyzed with the processing schedule of the CNC lathe. According to the processing schedule and the real-time status of the workpieces, it is possible to intelligently determine whether the current processing process is normal. If the quantity or location of the workpieces does not match the plan, such as missing, misplaced or unauthorized changes of the workpieces, the system will identify it as an abnormal situation.

[0177] The following is a specific example to illustrate:

[0178] Assume that a CNC lathe is processing a batch of standardized parts. Each part needs to go through three main processing stages: roughing, semi-finishing, and finishing. The processing schedule is as follows: Hour 1 to Hour 3: Roughing, with an estimated completion rate of 20 parts per hour. Hour 3 to Hour 5: Semi-finishing, with an estimated completion rate of 15 parts per hour. Hour 5 to Hour 7: Finishing, with an estimated completion rate of 10 parts per hour. Real-time monitoring: During the processing, the control end monitors the processing progress and the status of the workpiece in real time through sensors and vision systems integrated with the CNC lathe. Data collection: Collect information such as the processing time of each stage, the number of completed parts, and the placement of the workpiece on the machine. Progress comparison: Compare the actual processing progress with the planned progress to check whether there is a delay or ahead of schedule. Example analysis: Time point 1 (Hour 3): It was found that 60 parts should have been completed in the roughing stage, but only 50 were actually completed. At the same time, the vision system identified that 5 parts on the machine were placed incorrectly, which may affect the subsequent semi-finishing.

[0179] Time point 2 (5th hour): According to the plan, roughing and semi-finishing should be completed at this time, totaling 75 parts. The control end confirms that the actual number of completed parts is 70, and all workpieces are placed correctly, and can continue with finishing.

[0180] Time point 3 (7th hour): At the end of the finishing phase, it was found that only 8 parts had been finished, far less than the planned 30. The vision system also found that 2 parts were missing on the machine, which may mean that the workpiece was dropped or lost during the processing.

[0181] Judgment and response:

[0182] At time point 1, the visual system determines that the machining process is abnormal, which may be due to improper placement of the workpiece, resulting in reduced machining efficiency. The control end then notifies the operator to adjust the workpiece position and check the machining parameters.

[0183] At time point 2, although the progress is slightly delayed, the vision system determines that the machining process is basically normal and can continue to perform finishing.

[0184] At time point 3, it is determined that the machining process is seriously abnormal, the finishing progress is seriously delayed, and workpieces are lost. The control end automatically pauses the machining and issues an alarm, while generating fault diagnosis and inspection instructions to guide the operator to conduct inspections and repairs.

[0185] Through this real-time monitoring and intelligent analysis, the control end can promptly detect anomalies in the processing flow and take corresponding measures to ensure the continuity of the processing process and the stability of the processing quality. This intelligent judgment is of great significance for preventing processing errors, reducing material waste and avoiding equipment damage. Once an abnormality is found, the operation of the robot arm can be adjusted in time, the technician can be notified for inspection, or the processing plan can be automatically rearranged to ensure the continuity of the processing flow and the processing quality of the product.

[0186] Step 202, obtaining first operating state information when the CNC lathe processes the workpiece.

[0187] This step may refer to step 101 and will not be described in detail here.

[0188] Step 203: Identify the first operating state information by using a preset vibration identification model to obtain a first vibration type identification result of the CNC lathe.

[0189] This step may refer to step 102 and will not be described in detail here.

[0190] Step 204: when the first vibration type identification result is the first abnormal vibration type, determining a processing parameter adjustment strategy according to the first abnormal vibration type.

[0191] This step may refer to step 103 and will not be described in detail here.

[0192] Step 205 : determining a first adjustment parameter of the CNC lathe and a second adjustment parameter of the robot arm based on the machining parameter adjustment strategy and the first operating state information.

[0193] This step may refer to step 104 and will not be described in detail here.

[0194] Step 206: Send the first adjustment parameter and the second adjustment parameter to the CNC lathe and the robot arm respectively.

[0195] This step may refer to step 105 and will not be described in detail here.

[0196] Step 207: After the CNC lathe and the robot arm perform corresponding vibration control operations, second operation status information of the CNC lathe is obtained.

[0197] Specifically, after the CNC lathe and the robotic arm complete their respective vibration control operations, the control end will enter the second operation status information collection phase to check the effects of the vibration control operations of the CNC lathe and the robotic arm to determine whether the abnormal vibration of the CNC lathe is eliminated or mitigated.

[0198] To this end, the control end will once again use sensors installed on the CNC lathe, such as vibration sensors, sound sensors, torque sensors, etc., to monitor and record the operating status of the lathe in real time. These data will fully reflect the vibration characteristics of the lathe after the adjustment, including the frequency, amplitude, sound signal, feed rate and cutting force of the vibration. After obtaining this information, the control end will pre-process the data, such as filtering and normalization, to facilitate subsequent analysis and identification.

[0199] Step 208: Identify the second operating state information through the vibration identification model to obtain a second vibration type identification result of the CNC lathe.

[0200] The second vibration type recognition result includes whether the current vibration is normal or abnormal, and if it is abnormal, which type of abnormal vibration it is.

[0201] The specific identification process refers to the process and principle of identifying the first abnormal vibration type in the previous text, which will not be repeated here.

[0202] Step 209: When the second vibration type identification result is a second abnormal vibration type, generating an inspection instruction according to the second abnormal vibration type.

[0203] Specifically, when the second vibration type recognition result of the vibration recognition model shows that the lathe has the second abnormal vibration type, the system will generate corresponding inspection instructions according to this result. These instructions will guide the operator or the automation system to perform detailed inspection and maintenance on the CNC lathe.

[0204] For example, if a resonance problem is identified, the inspection instructions may include checking the fixation of the machine tool structure, calibrating the dynamic stiffness of the machine tool, etc. If it is self-excited vibration, it may be necessary to check the wear state of the tool, adjust the cutting parameters, etc. The generated inspection instructions will specifically solve the current abnormal vibration problem to restore the normal operation of the lathe and prevent potential mechanical damage.

[0205] In some embodiments, step 209 may specifically include:

[0206] (1) Determine the target inspection area and inspection items of the CNC lathe according to the second abnormal vibration type.

[0207] When the second vibration type identification result of the CNC lathe shows abnormal vibration, the target inspection area and inspection items of the lathe can be determined based on the preset knowledge base and the specific attributes of the abnormal vibration type. For example, if a resonance problem is identified, the target inspection area may include the various connecting parts and supporting structures of the machine tool, and the inspection items may involve the looseness of fasteners and the balance of the structure. In the case of self-excited vibration, the inspection focus may be tool wear, cutting parameters or workpiece clamping status.

[0208] (2) Determine the movement path of the robot arm based on the target inspection area and inspection items.

[0209] After the target inspection area and inspection items are identified, an optimal movement path will be planned for the robot. This path takes into account the robot's range of motion, the layout of the workpiece and machine tool, and the need for safe obstacle avoidance, ensuring that the robot can accurately and efficiently reach each predetermined inspection location. The planning of the movement path aims to minimize the robot's movement distance and time, while avoiding collisions with machine tools or other obstacles, and improving the efficiency and safety of the inspection process.

[0210] For example, suppose a CNC lathe is detected to have abnormal vibration during machining. After analysis by the vibration identification model, it is diagnosed as self-excited vibration possibly caused by tool wear or unstable tool clamping. Based on this diagnosis result, the following target inspection areas and inspection items are determined:

[0211] Tool clamping area: Check that the tool is firmly clamped and not loose.

[0212] Tool wear: Check the degree of tool wear to determine whether it needs to be replaced or sharpened.

[0213] Spindle bearings: Check the spindle bearings for abnormalities such as overheating or noise.

[0214] Next, we will plan the movement path for the robot arm:

[0215] Planning start: The robot starts in its standby position (e.g. a safe location near a CNC lathe).

[0216] Arriving at the tool clamping area: The robot arm first moves to the tool clamping area and performs visual and tactile checks to confirm that the tool is clamped.

[0217] Tool wear inspection: After checking the clamping condition, the robot arm adjusts its posture and uses high-definition cameras and sensors to inspect the tool wear.

[0218] Move to the spindle area: After confirming the tool status, the robot arm moves to the spindle bearing area and uses an infrared thermometer to check the bearing temperature to confirm whether there is overheating.

[0219] Return to standby position: After completing all inspection items, the robot arm cleans and records the inspection data, and then safely returns to its standby position.

[0220] (3) Generate inspection instructions based on the moving path.

[0221] Based on the robot's movement path, a series of detailed inspection instructions will be generated. These instructions include specific movement instructions for the robot to reach each inspection point, and the inspection operations to be performed, such as visual inspection, torque inspection of fasteners, or evaluation of tool status. The inspection instructions will also include how to record and report the inspection results, as well as corrective measures to be taken when problems are found. The generation of inspection instructions is the direct basis for the robot to perform inspection tasks, ensuring that the inspection process is carried out in an orderly manner and that abnormal conditions of the lathe can be accurately identified and recorded.

[0222] Step 210, sending an inspection instruction to the robot arm so that the robot arm collects inspection information of the target inspection area of ​​the CNC lathe.

[0223] The inspection instructions include but are not limited to the specific location that the robot needs to reach, the inspection actions to be performed, the types of tools or sensors used, and how to collect and record inspection data. After receiving the inspection instructions, the robot will autonomously move to the target inspection area and use its onboard visual system, tactile sensor, infrared thermometer and other detection tools to inspect the tool wear, clamping stability, bearing status, etc. of the CNC lathe.

[0224] During the inspection process, the robot arm will collect data in real time, such as images, temperature readings, vibration signals, etc., and feed back this inspection information to the control end. Based on this information, the control end can further analyze the operating status of the CNC lathe, determine whether there are any abnormalities, and take necessary maintenance or adjustment measures to ensure the stability and safety of the processing process.

[0225] Step 211, judging whether the CNC lathe fails according to the inspection information.

[0226] Specifically, the robot arm performs detailed inspections of the target inspection area and collects data in real time according to the inspection instructions executed. These data include visual images, vibration signals, temperature readings, etc., which are used to evaluate the status of key components of the lathe, such as tool wear, spindle balance, bearing condition, and stability of workpiece clamping. The control end compares and analyzes these inspection information with preset thresholds and normal working parameters. If the inspection data deviates significantly from the normal range, such as tool wear exceeding the safety limit, abnormal increase in bearing temperature, or loose clamping of the workpiece, it will be judged that the CNC lathe has a fault or a potential fault. This real-time monitoring and analysis process enables the system to identify problems in a timely manner, prevent potential processing errors, reduce unexpected downtime, and ensure the continuity of the processing process and product quality.

[0227] If a CNC lathe is judged to have a fault, the preset fault response procedure will be immediately initiated. First, the control end will issue a clear fault alarm to the operator, providing detailed information on the type and location of the fault. Subsequently, depending on the severity and type of the fault, the control end may automatically execute a series of countermeasures, such as pausing or stopping the processing operation, adjusting the machine tool settings, and urgently shutting down the power supply to the faulty area to prevent the fault from worsening. At the same time, the control end will generate a detailed fault report, including the time of the fault, the affected components, the recommended maintenance measures and necessary inspection items for the reference of maintenance technicians. Based on this report, maintenance personnel can quickly locate the problem and repair it. At the same time, the control end will update the maintenance log to record the fault handling process and results for future maintenance planning and control end optimization. After the fault is properly handled and it is confirmed that the status of each component of the lathe has returned to normal, the control end will instruct the lathe to restart and resume processing operations to ensure that production activities can continue safely and efficiently.

[0228] If there is no fault, continue to operate normally.

[0229] In some embodiments, the first operating status information further includes tool temperature and lathe current, and after step 203, the following steps are further included:

[0230] (1) Extract the vibration spectrum, tool temperature change, sound characteristics and lathe current waveform from the first operating state information.

[0231] In the first operating state information of the CNC lathe, it is necessary to carefully extract key operating features, which are crucial for subsequent fault diagnosis. Specifically, the vibration spectrum is obtained from the vibration sensor data through Fourier transform, revealing the frequency components of the lathe vibration; the tool temperature change is monitored in real time by the thermal sensor on the tool, reflecting the thermal state of the cutting process; the sound characteristics are captured by the sound sensor, and abnormal sound patterns can be identified from the noise; the lathe current waveform is recorded by the current sensing device, reflecting the load changes of the lathe motor. These features together constitute a comprehensive portrait of the lathe's operating status, providing a rich data basis for fault analysis.

[0232] (2) Determine a target fault prediction model from a plurality of fault prediction models according to the vibration type in the first vibration type identification result.

[0233] According to the features extracted from the first operating state information, the control end will compare multiple preset fault prediction models to determine the model that best suits the current vibration type and lathe state, and determine the model as the target fault prediction model. Each vibration type corresponds to a different fault prediction model, which is specifically manifested in different model parameters within the fault prediction model and different data samples used in the training process.

[0234] These models may include machine learning models based on different algorithms, such as random forests, support vector machines, neural networks, etc., each of which is optimized for different types of fault characteristics. By analyzing the correlation between the current vibration spectrum, temperature change and other characteristics and historical fault data, the control end will select a model that can most accurately predict the probability of failure to provide accurate predictive analysis for the operating status of the CNC lathe.

[0235] Specifically, the construction and training process of the fault prediction model is as follows:

[0236] 1. Data collection: Collect the operating data of CNC lathes under different working conditions and at different times, including vibration spectrum, tool temperature, sound characteristics, lathe current waveform, etc. At the same time, record the maintenance log and fault history of the lathe, which will be used as label data for model training.

[0237] 2. Data preprocessing: Clean the collected data to remove noise and outliers. Perform normalization to standardize data of different magnitudes and dimensions for easy model processing.

[0238] 3. Feature selection: Extract features useful for fault prediction from the raw data, such as the peak frequency of vibration, the entropy of the sound signal, the stability of the current waveform, etc. Use feature selection techniques such as principal component analysis (PCA) or recursive feature elimination (RFE) to reduce the number of features and improve model efficiency.

[0239] 4. Model selection: According to the nature of the problem and the characteristics of the data, choose an appropriate machine learning algorithm, such as random forest, support vector machine (SVM), neural network, gradient boosted tree (GBT), etc.

[0240] 5. Model training: Divide the collected data set into a training set and a test set. The training set is used to build the model, and the test set is used to evaluate the model performance. Use the training set data to train the model, adjust the model parameters, and perform multiple iterations until the optimal model configuration is found.

[0241] 6. Model validation: Use the test set data to validate the trained model and evaluate the model's predictive performance. Select appropriate evaluation indicators, such as accuracy, recall, F1 score, ROC curve, etc., to measure the model's predictive ability.

[0242] 7. Model optimization: Based on the validation results, adjust and optimize the model, which may include adjusting feature selection, changing model parameters, using different learning algorithms, etc.

[0243] 8. Model deployment: Deploy the trained and verified model to the actual CNC lathe intelligent processing system so that it can receive operating status information in real time and predict the probability of failure.

[0244] 9. Continuous learning: After the model is deployed, new operating data and fault information are continuously collected, and the model is regularly updated and retrained using this data to maintain the model’s predictive accuracy.

[0245] Through the above steps, we can build and train an intelligent model that can accurately predict CNC lathe failures, providing strong support for machine tool maintenance and fault prevention.

[0246] (3) The vibration spectrum, tool temperature change, sound characteristics, and lathe current waveform are input into the target fault prediction model to obtain the probability of failure of the CNC lathe.

[0247] Specifically, after determining the target fault prediction model, the control end will input the extracted vibration spectrum, tool temperature changes, sound characteristics, and lathe current waveform into the model. The model will conduct an in-depth analysis of these input data and use its learned rules and pattern recognition capabilities to evaluate the probability of failure of the CNC lathe. This process may involve complex data processing and pattern matching, but the final output will be a concise probability value that directly reflects the possibility of failure in the current state of the lathe. This probability value can provide decision support for operators, helping them determine whether immediate intervention is needed and what maintenance measures to take to prevent failures.

[0248] Among them, the fault prediction model calculates the fault probability based on the fault prediction formula. The mathematical expression of the fault prediction formula is:

[0249]

[0250] X(t)=[V(t),T(t),S(t),C(t)],

[0251] Where P(t) represents the probability of failure at time t; σ is the Sigmoid function, which is used to map the model output to the interval [0,1], which is expressed as the probability of failure; β0, β1, β2, β3, β4 are model parameters; V(t), T(t), S(t), C(t) represent the vibration spectrum, tool temperature change, sound characteristics and lathe current waveform vector at time t respectively; φ j is the jth characteristic function, which is a nonlinear mapping function;

[0252] α j is the characteristic function φ j The relevant weights;

[0253] X j(t) is a feature vector based on time series, including vibration, temperature, sound, and current features, representing the jth feature input at time t;

[0254] k is the number of characteristic functions;

[0255] φj(Xj(t)) represents the jth feature X in the feature vector X(t) j (t) Apply the nonlinear mapping function φ j ;

[0256] ∈(t) is the random error term at time t, which is used to capture the randomness that the model cannot explain.

[0257] The core idea of ​​the fault prediction formula is to map the multi-dimensional operating state characteristics X(t) of the CNC lathe to the fault probability P(t) through a series of linear and nonlinear transformations. In this way, the model can capture the complex relationship between the input characteristics and the fault probability. The training process of the model is to find the optimal weights β and α so that P(t) can reflect the real fault probability as accurately as possible.

[0258] The intelligent processing control method for CNC lathes of this embodiment can realize comprehensive and efficient processing process management. First, the method provides accurate data support for the intelligent analysis of the processing system by real-time monitoring of key operating parameters such as the vibration spectrum, tool temperature, sound characteristics and current waveform of the CNC lathe. Using these data, the preset vibration recognition model can accurately identify the vibration type of the lathe, including normal vibration, forced vibration, self-excited vibration and resonance, and then judge whether the current processing state is within the expected range. When abnormal vibration is identified, the processing parameters of the CNC lathe and the auxiliary parameters of the robot arm can be adjusted based on this intelligence to achieve closed-loop control of vibration, effectively reduce the instability in the processing process, and improve processing accuracy and product quality.

[0259] In addition, the method also includes an advanced fault prediction model, which predicts the probability of future failure of the lathe by analyzing historical fault data and real-time operating status information, thereby achieving early diagnosis and prevention of faults. When the system predicts a potential failure risk, it will automatically generate inspection instructions to guide the robot arm to perform necessary inspections and maintenance work, reducing manual intervention and improving the response speed and accuracy of maintenance.

[0260] Through this intelligent control method, CNC lathes can operate in the best condition, reduce downtime caused by vibration and failure, and significantly improve production efficiency. At the same time, effective monitoring and timely maintenance of the system reduce equipment damage, extend equipment life, and reduce maintenance costs. In addition, the automated maintenance and fault prediction functions of this method provide strong technical support for the realization of intelligent manufacturing, and promote the production process to progress in the direction of intelligence, automation and sustainable development.

[0261] In general, this CNC lathe intelligent machining control method achieves comprehensive optimization of the machining system process through its highly integrated monitoring, identification, prediction and maintenance functions, and provides an efficient, reliable and intelligent machining solution for the modern manufacturing industry.

[0262] In some embodiments, AR (Augmented Reality) technology can also be used to provide operators with real-time visualization of processing status to assist in making more accurate decisions. Specifically, the following steps are included:

[0263] Data integration: First, the real-time monitoring data of the CNC lathe (such as vibration frequency, sound signals, temperature, cutting force, etc.) needs to be integrated with the AR system. This usually involves developing or using an existing Industrial Internet of Things (IIoT) platform to collect and transmit data.

[0264] AR content development: Develop AR content, including 3D models, animations, charts, and other visual elements to intuitively represent the status and performance indicators of the lathe. These contents should be designed to clearly convey key information such as tool wear, machine temperature distribution, vibration patterns, etc.

[0265] User interface design: Design an intuitive user interface (UI) that allows operators to easily interact with the AR system. The UI should include controls for viewing real-time data, historical trends, alarms, and fault diagnostics.

[0266] AR equipment configuration: Operators are equipped with AR headsets or smart glasses so that they can view virtual information superimposed on the actual machine tools. The control end of the machining system can communicate with the AR device.

[0267] Real-time data overlay: Through the AR system, real-time monitoring data can be overlaid on the operator's field of view in the form of graphics, charts or animations. For example, vibration data can be displayed as a vibration pattern animation, and temperature data can be represented by a heat map with different colors.

[0268] Interactive fault diagnosis: Develop interactive features that allow operators to interact with virtual elements through an AR interface to get more information about the lathe status. For example, an operator can click on a component displayed in AR to view its detailed status and historical maintenance records.

[0269] Decision support systems: Integrate decision support tools such as fault trees, diagnostic guides, or predictive maintenance recommendations to help operators make more precise decisions based on the information displayed by AR.

[0270] Training and Simulation: Use AR technology to train operators and simulate them, allowing them to familiarize themselves with machine maintenance and troubleshooting processes without risk.

[0271] System integration: Integrate the AR system with the existing CNC lathe control system and enterprise resource planning (ERP) system to achieve seamless data flow and a wider range of application scenarios.

[0272] User feedback and iteration: Collect feedback from operators and continuously optimize AR content and functions to improve their practicality and user experience.

[0273] Through the above steps, AR technology can provide CNC lathe operators with an enhanced visual layer, converting complex data into intuitive graphic information, thereby assisting them in making more accurate decisions and improving processing efficiency and safety.

[0274] In some embodiments, the intelligent processing system can also be integrated with the cloud platform to achieve remote monitoring, data analysis and resource sharing. Specifically, the following steps are included:

[0275] Cloud service selection: Choose the right cloud service provider based on the needs of the processing system, taking into account data security, service stability, cost-effectiveness, and specific services provided (such as data storage, computing power, machine learning services, etc.).

[0276] Data interface development: Develop or integrate data interfaces so that the control end can transmit the collected data (vibration data, sound signals, temperature, cutting force, etc.) to the cloud platform in real time.

[0277] Data transmission security: Ensure the security of data during transmission and use encryption technologies such as TLS / SSL to protect data from unauthorized access and tampering.

[0278] Cloud platform configuration: Configure the required resources on the cloud platform, such as servers, databases, and storage space, to support the data storage and processing needs of the intelligent processing system.

[0279] Remote access function: Implement remote access function, allowing authorized users to access data and analysis tools on the cloud platform through a secure network connection.

[0280] Real-time monitoring: Develop a monitoring dashboard to display the real-time status and performance indicators of the CNC lathe, enabling remote users to monitor the equipment operation in real time.

[0281] Data analysis: Leverage the computing power of the cloud platform to conduct in-depth analysis of the large amount of data collected, using machine learning algorithms to identify patterns, predict failures and optimize processing.

[0282] Resource sharing: Establish a resource sharing mechanism to allow different users and systems to access authorized data and analysis results, and promote knowledge sharing and technical collaboration.

[0283] User interface and experience: Design an easy-to-use user interface to ensure that users at different levels can easily access and understand monitoring data and analysis results.

[0284] System integration: Integrate the intelligent processing system with the cloud platform's ERP, MES (manufacturing execution system) and other enterprise management systems to achieve seamless data flow and a wider range of application scenarios.

[0285] Maintenance and support: Provide continuous system maintenance and technical support to ensure the stable operation of the cloud-integrated intelligent processing system and optimize it based on user feedback.

[0286] Training and documentation: Provide users with training and detailed operation documentation to help them become familiar with the cloud-integrated system and effectively utilize remote monitoring and data analysis capabilities.

[0287] Compliance and standards: Ensure that the entire system integration complies with relevant industry standards and laws and regulations, especially those regarding data protection and privacy.

[0288] Through the above steps, the integration of intelligent processing systems and cloud platforms can provide manufacturing companies with powerful data support and services, enabling them to achieve more intelligent and automated production management, and improve production efficiency and market response speed.

[0289] The method provided in the above embodiment can be executed by the control end of the intelligent processing system, which is an electronic device. The electronic device in the embodiment of the present invention is described from the perspective of hardware processing. Figure 4 , is a schematic diagram of a physical device structure of an electronic device in an embodiment of the present application.

[0290] It should be noted that Figure 4 The structure of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0291] like Figure 4 As shown, the electronic device includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage part 408 to the random access memory (RAM) 403, such as executing the method described in the above embodiment. In RAM 403, various programs and data required for system operation are also stored. CPU 401, ROM 402 and RAM 403 are connected to each other through bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0292] The following components are connected to the I / O interface 405: an input section 406 including an audio input device, a button switch, etc.; an output section 407 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0293] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, various functions defined in the present invention are performed.

[0294] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0295] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0296] Specifically, the electronic device of this embodiment includes a processor and a memory, the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to enable the electronic device to execute the method provided by the above embodiment.

[0297] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the electronic device, the electronic device implements the method provided in the above embodiment.

[0298] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0299] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0300] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. An intelligent processing control method for a CNC lathe, characterized in that: include: Acquire first operating state information when a CNC lathe processes a workpiece, wherein the first operating state information includes at least one of lathe vibration data, sound signals, feed rate, and cutting force data; the first operating state information also includes tool temperature and lathe current; The first operating state information is identified by a preset vibration identification model to obtain a first vibration type identification result of the CNC lathe; the vibration identification model is a first abnormal vibration type calculated by a vibration identification formula, and the vibration identification formula includes: F = [f, A, S, v, P]; Z = φ (F); Where f represents the vibration frequency, A represents the vibration amplitude, S represents the characteristics of the sound signal, v represents the feed rate, P represents the tool cutting force, Z is the nonlinear mapping function of F, Zi is the nonlinear mapping function of F corresponding to the i-th training sample, y is the output result of the classifier, indicating the abnormal vibration type, α i is the Lagrange multiplier obtained through training, y i is the category label of the i-th training sample, K is the kernel function used to calculate the similarity between two data points in the original feature space, b is the bias term, and N is the number of training samples; Extracting vibration spectrum, tool temperature change, sound characteristics and lathe current waveform from the first operating state information; determining a target fault prediction model from multiple fault prediction models according to the vibration type in the first vibration type identification result; inputting the vibration spectrum, the tool temperature change, the sound characteristics and the lathe current waveform into the target fault prediction model to obtain the probability of failure of the CNC lathe, wherein the fault prediction model calculates the fault probability based on a fault prediction formula, and the mathematical expression of the fault prediction formula is: X(t)=[V(t),T(t),S(t),C(t)], Where P(t) represents the probability of failure at time t; σ is the Sigmoid function, which is used to map the model output to the interval [0,1], which is expressed as the probability of failure; β0, β1, β2, β3, β4 are model parameters; V(t), T(t), S(t), C(t) represent the vibration spectrum, tool temperature change, sound characteristics and lathe current waveform vector at time t respectively; α j is the characteristic function φ j The relevant weight; X j (t) is a time series-based feature vector, representing the jth feature input at time t; k is the number of feature functions; φ j is the jth characteristic function; ∈(t) is the random error term at time t; When the first vibration type identification result is a first abnormal vibration type, determining a processing parameter adjustment strategy according to the first abnormal vibration type; Based on the processing parameter adjustment strategy and the first operating state information, determining a first adjustment parameter of the CNC lathe and a second adjustment parameter of the robotic arm, wherein the first adjustment parameter is used for the CNC lathe to perform vibration control operation on itself, and the second adjustment parameter is used for the robotic arm to perform vibration control operation on the CNC lathe; The first adjustment parameter and the second adjustment parameter are sent to the CNC lathe and the robotic arm respectively.

2. The method according to claim 1, characterized in that After the step of sending the first adjustment parameter and the second adjustment parameter to the CNC lathe and the robotic arm respectively, the method further includes: After the CNC lathe and the robot arm perform corresponding vibration control operations, obtaining second operating state information of the CNC lathe; Identify the second operating state information by using the vibration identification model to obtain a second vibration type identification result of the CNC lathe; When the second vibration type identification result is a second abnormal vibration type, generating an inspection instruction according to the second abnormal vibration type; Sending the inspection instruction to the robot arm so that the robot arm collects inspection information of the target inspection area of ​​the CNC lathe; Determine whether the CNC lathe fails based on the inspection information.

3. The method according to claim 2, characterized in that The step of generating an inspection instruction according to the second abnormal vibration type comprises: Determining a target inspection area and inspection items for the CNC lathe according to the second abnormal vibration type; Determining a moving path of the robot arm according to the target inspection area and the inspection items; The inspection instruction is generated according to the moving path.

4. The method according to claim 1, characterized in that: Before the step of obtaining the first operating state information when the CNC lathe processes the workpiece, the method further includes: Determine expected pressure information of a workstation on the machine platform according to the processing progress of the CNC lathe, the workstation of the machine platform is used to place the workpiece, and a pressure sensor is provided on the workstation; Whether the processing flow of the workpiece is normal is determined based on the current workstation pressure information detected by the pressure sensor and the expected pressure information.

5. The method according to claim 1, characterized in that Before the step of obtaining the first operating state information when the CNC lathe processes the workpiece, the method further includes: Obtain the workpiece placement image of the machine; Recognize the workpiece placement image by computer vision technology to obtain the number and placement positions of the workpieces on the machine; According to the processing progress of the CNC lathe, and the number and placement of the workpieces on the machine platform, it is determined whether the processing flow of the workpieces is normal.

6. An electronic device, characterized in that: include: one or more processors and memory; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, wherein the computer program codes include computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the method according to any one of claims 1 to 5.

7. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on an electronic device, the electronic device is caused to execute the method as claimed in any one of claims 1 to 5.

8. A computer program product, characterized in that When the computer program product is executed on an electronic device, the electronic device is enabled to execute the method according to any one of claims 1 to 5.

Images