Long graphite tube end machining control method based on servo drive and pneumatic centering

Abstract

The present invention relates to the field of mechanical processing technology, and specifically to a method for controlling the end processing of a long graphite tube based on servo drive and pneumatic centering. The position information of the graphite tube is obtained, and the graphite tube is accurately centered and adjusted in position using pneumatic centering technology and a servo motor; a processing parameter calculation model is constructed through a deep learning algorithm, and the optimal processing parameters are output in combination with the current graphite tube information; the servo motor controls the processing according to the optimized parameters, and the pneumatic system realizes real-time regulation based on visual feedback; after the processing is completed, the surface quality and size of the graphite tube are automatically detected using a visual detection model. The present invention realizes intelligent control of the entire graphite tube processing process, improves processing accuracy, efficiency and quality stability, and has good engineering application value.

Description

Technical Field

[0001] The invention relates to the technical field of mechanical processing, in particular to a processing control method for an end portion of a long graphite tube based on servo drive and pneumatic centering. Background Art

[0002] Graphite materials are widely used in electronics, metallurgy, chemical industry and other fields due to their excellent electrical conductivity, thermal conductivity and high temperature resistance. Among them, large-sized high-end graphite products require precise machining, especially graphite tube products with large diameter and long length, which have high requirements for processing technology and equipment.

[0003] Traditional graphite tube processing methods mainly use manual clamping and empirical parameters, which have problems such as difficult centering, low dimensional accuracy, and low efficiency. At present, the graphite tube processing technology still has the following shortcomings: First, for long graphite tubes with a large aspect ratio, the existing clamping and centering methods are difficult to effectively avoid the shaking and offset of the graphite tube during processing, affecting the processing accuracy; second, the selection of processing parameters mainly relies on experience, lacks intelligent optimization means, and it is difficult to obtain the best process parameter combination; third, there is a lack of effective online detection means to monitor the processing status, and it is impossible to achieve closed-loop feedback control and adaptive adjustment.

[0004] The patent with publication number CN110000480A discloses a high-precision graphite laser cutting machine and a method for processing high-precision graphite pipes, including an L-shaped frame, a marble table, a left clamp, a right clamp, a laser head, a compressed air nozzle, an X-axis linear motor guide system, a Z-axis linear motor guide system, a Y-axis linear motor guide system, a horizontal slider, a vertical slider and a controller, which are specifically designed for the processing of brittle pipes such as graphite pipes, and are also used for the processing of high-tight pipes. The graphite laser cutting machine of the invention has the function of adaptively changing the clamping support into a flexible support according to the processing position of the laser head. At the same time, through a simple mechanical structure, the laser head can dynamically follow the displacement of the graphite pipe to be processed, further improving the accuracy of laser cutting. The cutting machine has a high degree of automation and high precision.

[0005] However, the prior art does not solve the problem of how to position and position the graphite tube during processing, and does not solve the problem of processing parameters and quality detection of the graphite tube.

[0006] In order to overcome the above-mentioned shortcomings, the present invention proposes a long graphite tube end processing control method based on servo drive and pneumatic centering. Summary of the invention

[0007] The present invention aims to overcome the above-mentioned technical difficulties in graphite tube processing by innovatively integrating machine vision-guided pneumatic centering technology, servo-driven precision control technology, deep learning processing parameter optimization technology, and online visual inspection technology, so as to achieve intelligent control of the processing process and reliable quality assurance, and promote the upgrading of graphite tube processing technology.

[0008] To achieve the above object, the present invention provides a long graphite tube end processing control method based on servo drive and pneumatic centering, and the specific technical scheme is as follows:

[0009] Obtain image data of the graphite tube, construct an edge detection algorithm to detect the graphite tube image, and obtain the position of the graphite tube;

[0010] According to the position of the graphite tube, pneumatic centering is performed by adjusting the rubber inner support of the three-jaw chuck, and the position of the graphite tube is adjusted by the servo motor;

[0011] Based on the historical processing parameters of the graphite tube, a deep learning algorithm is constructed and trained as a graphite tube processing parameter calculation model; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the graphite tube processing parameters;

[0012] Based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, the servo motor controls the graphite tube for turning processing, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure according to the visual feedback in real time to maintain the stability of the graphite tube.

[0013] After the graphite tube is processed, a visual inspection model is constructed to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified.

[0014] Preferably, an industrial camera is used to scan the graphite tube, and the size of the graphite tube and the spatial position information of the graphite tube are obtained by an edge detection algorithm;

[0015] Scan the graphite tube with an industrial camera to obtain the original image , the original image Convert to grayscale ; Use Gaussian filter Smooth the image and remove noise interference: ,in represents the convolution operation, is the image after Gaussian filtering;

[0016] Gaussian filter ,in is the standard deviation; Sobel operator is used to calculate the gradient of the graphite tube image in the x-axis and y-axis directions and ; Calculate the gradient amplitude and direction :

[0017]

[0018] Gradient Amplitude Perform non-maximum suppression to obtain refined edges ;

[0019] Detect and connect edges using a dual threshold algorithm, defining a high threshold and low threshold ,like The pixel point is a strong edge, is non-marginal, The weak edge;

[0020] The weak edges with strong edges in the pixel neighborhood of the graphite tube image are regarded as true edges, and the remaining weak edges are regarded as non-edges, and the final edge image of the graphite tube is obtained. ;extract The outline in ,in , Indicates the contour number, Indicates the number of edge contours; for graphite tube contours Fitting ellipse, the fitting formula is:

[0021]

[0022] in is the coordinate of the center of the graphite tube ellipse, is the half length of the major and minor axes, is the rotation angle;

[0023] If the major and minor axes of the ellipse ,in is the ellipse approximation threshold, and the graphite tube cross section is regarded as a circle; the coordinates of the center of the circle are ,graphite tube radius ; Output the coordinates of the center of the circle and the radius of the graphite tube ;

[0024] The graphite tube is spatially positioned, and the pixel coordinates of the center of the circle are calculated according to the camera's intrinsic parameter matrix. Convert to normalized coordinates ;

[0025] The distance from the camera to the graphite tube and angle The external parameters are: is the pitch angle, is the yaw angle, is the roll angle; construct the transformation matrix from the camera coordinate system to the world coordinate system ,in is the rotation matrix, is the transpose of the zero vector, is the translation vector;

[0026] Normalize the camera coordinates of the circle center Multiply , get the spatial coordinates of the center of the graphite tube .

[0027] Preferably, based on the graphite tube position information, the three-jaw chuck is adjusted to achieve pneumatic centering of the graphite tube to be processed;

[0028] Get the center coordinates of the three-jaw chuck ; Calculate the deviation between the center of the graphite tube and the center of the three-jaw chuck : , ;

[0029] According to the deviation between the center of the graphite tube and the center of the three-jaw chuck, calculate the cylinder stroke adjustment of the three-jaw chuck:

[0030]

[0031] in, For the The stroke adjustment of each cylinder, For the The angle between the cylinder and the x-axis, is the proportional coefficient of deviation and cylinder stroke;

[0032] Control the extension and retraction of the three-jaw chuck cylinder to adjust the position of the three-jaw chuck's claws: ;in, For the Target stroke of each cylinder, is the initial stroke of the cylinder;

[0033] Control the clamping force of the three-jaw chuck according to the radius of the graphite tube Calculate the required clamping force: ;in, is the target clamping force, is the initial clamping force, is the proportionality coefficient between radius and clamping force increment;

[0034] Set the target clamping force Convert to the air pressure value of the three-jaw chuck cylinder : ;in, is the effective pressure area of ​​the three-jaw chuck;

[0035] After adjusting the three-jaw chuck to achieve pneumatic centering, the three-dimensional spatial coordinates of the center of the graphite tube are obtained through an industrial camera. .

[0036] Preferably, a three-axis servo drive is used to control the movement of the graphite tube, and the three-axis servo drive includes an axial servo motor, a radial servo motor and a rotary servo motor;

[0037] The axial servo motor is used for the axial movement of the graphite tube in the horizontal direction; the radial servo motor is used for the radial movement of the graphite tube in the vertical direction; the rotary servo motor is used for the rotary movement of the graphite tube;

[0038] The horizontal two-dimensional coordinates of the center of the graphite tube obtained by the industrial camera , calculate the axial adjustment of the graphite tube in the horizontal direction: ;in, is the initial axial position of the graphite tube;

[0039] Convert the axial adjustment to the target position of the graphite tube end point: ; ; Control the axial servo motor to drive the graphite tube to the target position ;

[0040] After the three-jaw chuck is used to clamp the graphite tube, the three-dimensional spatial coordinates of the center of the graphite tube are obtained by the industrial camera: , calculate the radial adjustment of the graphite tube in the vertical direction: ;in, is the radial initial position of the graphite tube; controls the radial servo motor to drive the graphite tube to move to the target position : ;

[0041] The rotary servo motor is controlled to transmit the motor rotational motion to the graphite tube, so that the graphite tube rotates.

[0042] Preferably, historical processing parameters of graphite tubes of multiple specifications are obtained, wherein the historical processing parameters include spindle speed , Feed speed , cutting depth And the tool corner radius ;

[0043] Construct a graphite tube processing parameter calculation model based on the long short-term memory neural network LSTM model;

[0044] Define the LSTM model input as the historical processing parameters and graphite tube size, expressed as ;in, is the spindle speed, is the feed speed, is the cutting depth, is the tool corner radius, is the radius of the graphite tube, is the processing length of graphite tube;

[0045] Define the LSTM model output as the processing parameters of the graphite tube, expressed as ;

[0046] is the spindle speed command, Feed speed command, is the cutting depth instruction;

[0047] Constructing a network structure of an LSTM model, wherein the network structure of the LSTM model includes an input layer, a hidden layer, and a fully connected output layer, and using the Adam optimization algorithm to train the model;

[0048] The mean square error (MSE) is used as the loss function, and the trained LSTM model is used in combination with the current size parameters of the graphite tube to generate the processing parameters of the graphite tube.

[0049] Preferably, according to the trained graphite tube processing parameter calculation model, combined with the size information of the current graphite tube, the processing parameters of the graphite tube are output, including the servo motor spindle speed, feed speed and cutting depth;

[0050] According to the cutting depth instruction, the position of the tool holder servo motor is controlled to adjust the cutting depth of the turning tool; during the processing, the position and vibration of the graphite tube are monitored in real time through an industrial camera, and the visual feedback information is transmitted to the three-jaw chuck cylinder for real-time pneumatic center adjustment.

[0051] Preferably, after the graphite tube is processed, an industrial camera is used to collect images of the processed graphite tube, a graphite tube surface defect detection model and a dimension measurement model are constructed, and surface defect detection and dimension detection are performed on the processed graphite tube;

[0052] For surface defect detection, a convolutional neural network (CNN) model is used. The graphite tube surface image is taken as input. The texture and color features of the graphite tube surface are extracted through the convolution layer, and then classified through the fully connected layer to determine whether there are scratches, cracks, and pits on the graphite tube surface.

[0053] For size detection, the convolutional neural network (CNN) model extracts edges and performs ellipse fitting on the graphite tube contour image to obtain the diameter and roundness size parameters of the graphite tube, and compares them with the design size to calculate whether the size error is within the allowable tolerance range.

[0054] When the processed graphite tube has no surface defects and the size is qualified, the processed graphite tube is marked as a qualified product; when the processed graphite tube has any of the surface defects and unqualified size defects, the processed graphite tube is marked as an unqualified product.

[0055] A long graphite tube end processing control system based on servo drive and pneumatic centering, which is used to implement the long graphite tube end processing control method based on servo drive and pneumatic centering, a data acquisition module, a processing control module, a processing parameter calculation module, a processing control module and a quality detection module;

[0056] The data acquisition module acquires the image data of the graphite tube, constructs an edge detection algorithm to detect the image of the graphite tube, and acquires the position of the graphite tube;

[0057] The processing control module adjusts the position of the graphite tube by adjusting the rubber inner support of the three-jaw chuck for pneumatic centering according to the position of the graphite tube and adjusting the position of the graphite tube by the servo motor;

[0058] The processing parameter calculation module constructs and trains a deep learning algorithm as a graphite tube processing parameter calculation model based on the historical processing parameters of the graphite tube; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the processing parameters of the graphite tube;

[0059] The processing control module controls the graphite tube to perform turning processing based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure according to visual feedback in real time to maintain the stability of the graphite tube;

[0060] After the graphite tube is processed, the quality inspection module constructs a visual inspection model to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified.

[0061] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the long graphite tube end processing control method based on servo drive and pneumatic centering by calling the computer program stored in the memory.

[0062] A storage medium stores instructions. When the instructions are run on a computer, the computer is enabled to execute the long graphite tube end processing control method based on servo drive and pneumatic centering.

[0063] The beneficial effects of the present invention are as follows: the present invention improves the centering accuracy of graphite tube processing, introduces machine vision technology to scan the graphite tube, obtains accurate size, shape and position information, combines pneumatic centering technology, and adaptively adjusts the rubber inner support to achieve accurate centering of graphite tubes with different size deviations, avoiding the limitations of traditional mechanical centering methods, thereby ensuring the coaxiality and processing quality of subsequent processing.

[0064] The present invention realizes intelligent optimization of graphite tube processing parameters. Through deep learning algorithms, historical processing parameter data is learned to establish a graphite tube processing parameter calculation model. The model can adaptively generate optimal processing parameters according to the characteristics of the current graphite tube, replacing the traditional parameter setting method that relies on workers' experience, improving the scientificity and reliability of processing parameter selection, and laying the foundation for realizing process standardization and optimization.

[0065] The present invention enhances the real-time control capability of the machining process. During the machining process, the servo motor accurately controls the feed speed and cutting depth according to the optimized machining parameters. At the same time, the graphite tube state is monitored in real time through machine vision, and the visual feedback information is transmitted to the pneumatic control system to dynamically adjust the internal support pressure of the three-jaw chuck. The closed-loop control method can respond to changes in the machining state in a timely manner, ensure the stability of the graphite tube throughout the machining process, and improve the reliability and consistency of the machining.

[0066] The present invention improves the quality inspection level of graphite tube processing; after processing is completed, a visual inspection model is used to inspect the surface quality and size of the graphite tube, thereby realizing automatic identification of defects and accurate measurement of key dimensions; compared with manual sampling, the intelligent inspection method is more efficient and more accurate, and can comprehensively evaluate the processing quality, discover problems in a timely manner, and reduce the defective product rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A flow chart of a long graphite tube end processing control method based on servo drive and pneumatic centering provided by the present invention;

[0068] Figure 2 A structural diagram of a long graphite tube end processing control system based on servo drive and pneumatic centering provided by the present invention;

[0069] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0070] Figure 4 It is a schematic diagram of the storage medium structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to better understand the present invention, a more detailed description will be made of various aspects of the present invention with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0072] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The accompanying drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in the present invention, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0073] It should also be understood that expressions such as "comprises", "including", "having", "includes" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present invention, "may" is used to mean "one or more embodiments of the present invention". And, the term "exemplary" is intended to refer to an example or illustration.

[0074] Unless otherwise defined, all terms (including engineering terms and scientific and technological terms) used in this document have the same meaning as commonly understood by ordinary technicians in the field to which the present invention belongs. It should also be understood that unless otherwise clearly stated in the present invention, words defined in commonly used dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0075] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0076] Example 1

[0077] Reference Figure 1 , which is the first embodiment of the present invention, provides a long graphite tube end processing control method based on servo drive and pneumatic centering.

[0078] S1: transport the graphite tube to the processing station, scan the graphite tube, and obtain the scanned size, shape and position information of the graphite tube.

[0079] An industrial camera is used to scan the graphite tube, and the size and spatial position information of the graphite tube are obtained through an edge detection algorithm.

[0080] Scan the graphite tube with an industrial camera to obtain the original image , the original image Convert to grayscale ; Use Gaussian filter Smooth the image and remove noise interference: ,in represents the convolution operation, is the image after Gaussian filtering.

[0081] Gaussian filter ,in is the standard deviation, which is used to control the degree of smoothing; the Sobel operator is used to calculate the gradient of the graphite tube image in the x-axis and y-axis directions and :

[0082]

[0083]

[0084] Calculate the gradient magnitude and direction :

[0085]

[0086] Gradient Amplitude Perform non-maximum suppression to obtain refined edges .

[0087] Detect and connect edges using a dual threshold algorithm, defining a high threshold and low threshold ,like The pixel point is a strong edge, is non-marginal, The weak edge.

[0088] The weak edges with strong edges in the pixel neighborhood of the graphite tube image are regarded as true edges, and the remaining weak edges are regarded as non-edges, and the final edge image of the graphite tube is obtained. ;extract The outline in ,in , Indicates the contour number, Indicates the number of edge contours; for graphite tube contours Fitting ellipse, the fitting formula is:

[0089]

[0090] in is the coordinate of the center of the graphite tube ellipse, is the half length of the major and minor axes, is the rotation angle;

[0091] If the major and minor axes of the ellipse ,in is the ellipse approximation threshold, and the graphite tube cross section is regarded as a circle; the coordinates of the center of the circle are ,graphite tube radius ; Output the coordinates of the center of the circle and the radius of the graphite tube .

[0092] Position the graphite tube in space according to the camera's intrinsic parameter matrix :

[0093]

[0094] in is the focal length, is the principal point coordinate, the pixel coordinate of the center of the circle Convert to normalized coordinates : .

[0095] is the pixel coordinate of the center of the circle in the image coordinate system; is the coordinate of the center of the circle in the normalized coordinate system; by subtracting the coordinates of the principal point , translate the coordinates of the center of the circle to the coordinate system with the principal point as the origin, and then divide it by the corresponding focal length , normalize the coordinates to the focal plane of the camera.

[0096] The distance from the camera to the graphite tube and angle The external parameters are: is the pitch angle, is the yaw angle, is the roll angle; construct the transformation matrix from the camera coordinate system to the world coordinate system ,in is the rotation matrix, is the transpose of the zero vector, is the translation vector.

[0097] Normalize the camera coordinates of the circle center Multiply , get the spatial coordinates of the center of the graphite tube .

[0098] The edge detection algorithm is used to locate the graphite tube, which can significantly improve the centering accuracy of graphite tube processing. The high-precision size measurement and spatial positioning of the graphite tube are achieved through machine vision technology, and the positioning accuracy can reach 0.1mm, which is much higher than the traditional manual centering method. Accurate centering can reduce processing deviations and improve the dimensional accuracy and consistency of products.

[0099] S2: According to the position of the graphite tube, pneumatic centering is performed by adjusting the rubber inner support of the three-jaw chuck, and the position of the graphite tube is adjusted by the servo motor.

[0100] The three-jaw chuck uses three independently controlled cylinders to drive the three claws. The cylinders are evenly distributed along the circumference, and the angle between adjacent cylinders is 120°. The telescopic rod of each cylinder is hinged to the bottom of the claw, and the radial position of the claw is adjusted by changing the stroke of the cylinder.

[0101] The process of pneumatic centering is as follows: First, the deviation vector of the graphite tube center coordinates and the three-jaw chuck center coordinates obtained by visual positioning is calculated. Then, the deviation vector is decomposed into the directions of the three cylinders to obtain the stroke amount that needs to be adjusted for each cylinder. Finally, by controlling the intake and exhaust volume of the three cylinders, the cylinder stroke is precisely adjusted, so that the center of the three-jaw chuck coincides with the center of the graphite tube, and pneumatic centering is achieved.

[0102] While pneumatic centering, the clamping force of the three-jaw chuck also needs to be controlled. According to the diameter of the graphite tube, the theoretical clamping force can be calculated. The air supply pressure of the cylinder can be adjusted to a suitable value through a proportional valve or an electrically controlled air pressure valve to achieve the required clamping force. During the clamping process, the cylinder pressure and the displacement of the jaws need to be monitored in real time to prevent the surface of the graphite tube from being damaged by excessive clamping force, or the graphite tube from slipping due to insufficient clamping force.

[0103] Specifically, obtain the center coordinates of the three-jaw chuck ; Calculate the deviation between the center of the graphite tube and the center of the three-jaw chuck : , ;

[0104] According to the deviation between the center of the graphite tube and the center of the three-jaw chuck, calculate the cylinder stroke adjustment of the three-jaw chuck:

[0105]

[0106] in, For the The stroke adjustment of each cylinder, For the The angle between the cylinder and the x-axis, It is the proportional coefficient between the deviation and the cylinder stroke and is set according to the rated stroke of the cylinder and the allowable deviation range.

[0107] Control the extension and retraction of the three-jaw chuck cylinder to adjust the position of the three-jaw chuck's claws: ;in, For the Target stroke of each cylinder, is the initial stroke of the cylinder.

[0108] Control the clamping force of the three-jaw chuck according to the radius of the graphite tube Calculate the required clamping force: ;in, is the target clamping force, is the initial clamping force, It is the proportional coefficient between radius and clamping force increment, which is set according to factors such as graphite tube material and dimensional tolerance.

[0109] Set the target clamping force Convert to the air pressure value of the three-jaw chuck cylinder : ;in, It is the effective pressure area of ​​the three-jaw chuck; control the air supply pressure of the cylinder to make the chuck jaws produce the required clamping force.

[0110] After adjusting the three-jaw chuck to achieve pneumatic centering, the three-dimensional spatial coordinates of the center of the graphite tube are obtained through an industrial camera. .

[0111] A three-axis servo drive is used to control the movement of the graphite tube. The three-axis servo drive includes an axial servo motor, a radial servo motor and a rotary servo motor.

[0112] The servo motors all adopt a three-closed-loop control structure, including a current loop, a speed loop, and a position loop. The current loop is responsible for controlling the output torque of the motor, the speed loop is responsible for controlling the speed of the motor, and the position loop is responsible for controlling the position of the motor. The three control loops are nested with each other, and the output of the outer loop is used as the given value of the inner loop, thereby achieving precise control of the servo motor.

[0113] The axial servo motor is used for the axial movement of the graphite tube in the horizontal direction; the radial servo motor is used for the radial movement of the graphite tube in the vertical direction; and the rotary servo motor is used for the rotary movement of the graphite tube.

[0114] The horizontal two-dimensional coordinates of the center of the graphite tube obtained by the industrial camera , calculate the axial adjustment of the graphite tube in the horizontal direction: ;in, is the initial axial position of the graphite tube.

[0115] Convert the axial adjustment to the target position of the graphite tube end point: ; ; Control the axial servo motor to drive the graphite tube to the target position .

[0116] After the three-jaw chuck is used to clamp the graphite tube, the three-dimensional spatial coordinates of the center of the graphite tube are obtained by the industrial camera: , calculate the radial adjustment of the graphite tube in the vertical direction: ;in, is the radial initial position of the graphite tube; controls the radial servo motor to drive the graphite tube to move to the target position : .

[0117] The rotary servo motor is controlled to transmit the motor rotational motion to the graphite tube, so that the graphite tube rotates.

[0118] Through the pneumatic centering technology, the automatic centering and clamping of the graphite tube is realized, which overcomes the inefficiency and inaccuracy of manual centering and improves the processing efficiency and centering accuracy.

[0119] A combination of pneumatic and servo motors is adopted. The pneumatic three-jaw chuck is responsible for the initial centering and clamping of the graphite tube, and the servo motor is responsible for accurately adjusting the position of the graphite tube, ensuring the accuracy and reliability of centering.

[0120] S3: Based on the historical processing parameters of the graphite tube, a deep learning algorithm is constructed and trained as a graphite tube processing parameter calculation model; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the processing parameters of the graphite tube.

[0121] Obtain historical processing parameters of graphite tubes of multiple specifications, including spindle speed , Feed speed , cutting depth And the tool corner radius .

[0122] A graphite tube processing parameter calculation model based on the long short-term memory neural network LSTM model is constructed. The LSTM model training process is as follows:

[0123] Extract historical graphite tube processing data, including input features And the corresponding optimized processing parameters ;in, is the spindle speed, is the feed speed, is the cutting depth, is the tool corner radius, is the radius of the graphite tube, is the processing length of graphite tube; is the spindle speed command, Feed speed command, It is the cutting depth instruction.

[0124] The data set is randomly divided into training set and test set in a ratio of 8:2, the input features are normalized, and the data is scaled to the interval [0, 1]: ;in and are the minimum and maximum values ​​of the data in the data set respectively; is the normalized data, is the original data.

[0125] Constructing LSTM model The Keras framework is used to construct the LSTM model. The LSTM model structure includes input layer, hidden layer, Dropout layer, fully connected layer and output layer.

[0126] The input layer has input data of shape (timesteps, input_dim); the hidden layer contains 64 neurons and uses tanh activation function; the Dropout layer has a dropout rate of 0.2 to prevent overfitting; the fully connected layer contains 32 neurons and uses ReLU activation function; the output layer has output data of shape (output_dim).

[0127] The forward propagation process of the LSTM model can be expressed as:

[0128]

[0129] in, is the input of the current time step, is the hidden state at the current time step, The predicted processing parameters.

[0130] Set the training parameter loss function to mean square error MSE:

[0131]

[0132] in, and are the weights and biases of the model, respectively. is the number of training samples, and Respectively The true output of each sample and the model predicted output.

[0133] The optimization algorithm is Adam optimizer and the learning rate is 0.001.

[0134] The training model sends the training set data into the model and performs forward propagation calculation to predict the output . Calculate the loss function , perform back propagation and update model parameters and ; Evaluate model performance on the test set and calculate mean square error and coefficient of determination

[0135]

[0136] in, is the mean of the true outputs.

[0137] Monitor validation loss and value, perform early stopping and save the model; evaluate the trained model on the test set, calculate the mean square error, mean absolute error and value.

[0138] Draw a scatter plot of the actual output and the predicted output, observe the fitting effect, and perform hyperparameter tuning, such as adjusting the number of LSTM layers, the number of neurons, the Dropout rate, the learning rate, etc. Repeat the training process and select the model with the best performance.

[0139] The model application uses the trained LSTM model to predict the optimized processing parameters based on the real-time processing data of the current graphite tube.

[0140] The predicted and optimized machining parameters are passed to the CNC system and PLC control system to guide the actual machining process; new machining data is continuously collected, and the model is retrained and updated regularly to adapt to changes in process conditions.

[0141] This step introduces a graphite tube processing parameter optimization method based on deep learning, which overcomes the limitations of traditional empirical parameter adjustment, realizes the intelligent selection and optimization of processing parameters, and improves processing efficiency and quality.

[0142] The LSTM neural network model can make full use of the historical data of graphite tube processing, learn the complex mapping relationship between optimization parameters and processing status in a data-driven way, has strong nonlinear processing capabilities, and improves the accuracy of parameter optimization.

[0143] S4: Based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, the servo motor controls the graphite tube for turning processing, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure according to visual feedback in real time to maintain the stability of the graphite tube.

[0144] According to the trained graphite tube processing parameter calculation model and combined with the current graphite tube size information, the graphite tube processing parameters are output, including the servo motor spindle speed, feed speed and cutting depth.

[0145] According to the cutting depth instruction, the position of the tool holder servo motor is controlled to adjust the cutting depth of the turning tool; during the processing, the position and vibration of the graphite tube are monitored in real time through an industrial camera, and the visual feedback information is transmitted to the three-jaw chuck cylinder for real-time pneumatic center adjustment.

[0146] By changing the contact force between the rubber inner support and the graphite tube, the graphite tube is kept in the center of the chuck, reducing the impact of vibration on processing accuracy; at the same time, the three-jaw chuck works together with the servo motor. When the graphite tube deviates, the servo motor adjusts the feed speed and cutting depth in real time according to the offset direction and distance to ensure the stability of the processing process.

[0147] S5: After the graphite tube is processed, a visual inspection model is constructed to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified.

[0148] After the graphite tube is processed, an industrial camera is used to capture images of the processed graphite tube, and a graphite tube surface defect detection model and a dimension measurement model are constructed to perform surface defect detection and dimension detection on the processed graphite tube.

[0149] For surface defect detection, a convolutional neural network (CNN) model is used. The graphite tube surface image is taken as input. The texture and color features of the graphite tube surface are extracted through the convolution layer, and then classified through the fully connected layer to determine whether there are scratches, cracks, and pits on the graphite tube surface.

[0150] For size detection, the convolutional neural network (CNN) model performs edge extraction and ellipse fitting on the graphite tube contour image to obtain the size parameters of the diameter and roundness of the graphite tube, and compares them with the design dimensions to calculate whether the size error is within the allowable tolerance range.

[0151] When the processed graphite tube has no surface defects and the size is qualified, the processed graphite tube is marked as a qualified product; when the processed graphite tube has any of the surface defects and unqualified size defects, the processed graphite tube is marked as an unqualified product.

[0152] Through visual inspection technology, automated quality inspection of processed graphite tubes can be achieved, overcoming the subjectivity and inefficiency of manual inspection and improving inspection efficiency and reliability.

[0153] The deep learning CNN model is used to intelligently identify and classify defects on the graphite tube surface. The CNN model extracts multi-scale texture and defect features on the graphite tube surface through convolution operations, and establishes a direct mapping from the original image to the defect classification results through end-to-end learning. It has strong feature extraction and classification capabilities, which greatly improves the accuracy of defect detection.

[0154] Example 2

[0155] Reference Figure 2 , which is the second embodiment of the present invention, provides a long graphite tube end processing control system based on servo drive and pneumatic centering.

[0156] The data acquisition module acquires the image data of the graphite tube, constructs an edge detection algorithm to detect the graphite tube image, and acquires the position of the graphite tube.

[0157] The processing control module performs pneumatic centering by adjusting the rubber inner support of the three-jaw chuck according to the position of the graphite tube, and adjusts the position of the graphite tube through the servo motor.

[0158] The processing parameter calculation module constructs and trains a deep learning algorithm as a graphite tube processing parameter calculation model based on the historical processing parameters of the graphite tube; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the processing parameters of the graphite tube.

[0159] The processing control module controls the graphite tube to perform turning processing based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure in real time according to visual feedback to maintain the stability of the graphite tube.

[0160] After the graphite tube is processed, the quality inspection module constructs a visual inspection model to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified.

[0161] Example 3

[0162] Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, according to another aspect of the present invention, an electronic device 500 is also provided. The electronic device 500 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are run by one or more processors, the long graphite tube end processing control method based on servo drive and pneumatic centering as described above can be executed.

[0163] The method or system according to the embodiment of the present invention can also be used by Figure 3The architecture of the electronic device shown is implemented.

[0164] like Figure 3 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, and the like.

[0165] The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, can store the long graphite tube end processing control method based on servo drive and pneumatic centering provided by the present invention.

[0166] A long graphite tube end processing control method based on servo drive and pneumatic centering includes: obtaining image data of the graphite tube, constructing an edge detection algorithm to detect the graphite tube image, and obtaining the position of the graphite tube; according to the position of the graphite tube, pneumatic centering is performed by adjusting the rubber inner support of a three-jaw chuck, and the position of the graphite tube is adjusted by a servo motor; based on the historical processing parameters of the graphite tube, a deep learning algorithm is constructed and trained as a graphite tube processing parameter calculation model; the trained graphite tube processing parameter calculation model is combined with current graphite tube information to output the processing parameters of the graphite tube; based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, a servo motor controls the graphite tube to perform turning processing, and the feed speed and cutting depth are adjusted in real time. The three-jaw chuck dynamically adjusts the inner support pressure according to visual feedback in real time to maintain the stability of the graphite tube; after the graphite tube processing is completed, a visual detection model is constructed to perform surface quality and size detection on the graphite tube to determine whether the processed graphite tube is qualified.

[0167] Furthermore, the electronic device 500 may also include a user interface 508. Figure 3 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 3 One or more components of an electronic device are shown.

[0168] Example 4

[0169] Figure 4 It is a schematic diagram of the storage medium structure provided by an embodiment of the present invention.

[0170] like Figure 4 As shown is a storage medium 600 according to one embodiment of the present invention.

[0171] The storage medium 600 has computer readable instructions stored thereon.

[0172] When the computer-readable instructions are executed by the processor, the long graphite tube end processing control method based on servo drive and pneumatic centering according to the embodiments of the present invention described with reference to the above drawings can be executed.

[0173] The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. In addition, according to an embodiment of the present invention, the process described above with reference to the flowchart may be implemented as a computer software program.

[0174] For example, the present invention provides a non-temporary machine storage medium, which stores machine-readable instructions, and the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present invention, for example: obtaining image data of a graphite tube, constructing an edge detection algorithm to detect the graphite tube image, and obtaining the position of the graphite tube; according to the position of the graphite tube, pneumatic centering is performed by adjusting the rubber inner support of the three-jaw chuck, and the position of the graphite tube is adjusted by a servo motor; based on the historical processing parameters of the graphite tube, a deep learning algorithm is constructed and trained as a graphite tube processing parameter calculation model; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the processing parameters of the graphite tube; based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, the servo motor controls the graphite tube for turning processing, and the feed speed and cutting depth are adjusted in real time. The three-jaw chuck dynamically adjusts the inner support pressure according to visual feedback in real time to maintain the stability of the graphite tube; after the graphite tube processing is completed, a visual inspection model is constructed to perform surface quality and size inspection on the graphite tube to determine whether the processed graphite tube is qualified.

[0175] When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present invention are performed. The method, apparatus, and device of the present invention may be implemented in many ways. For example, the method, apparatus, and device of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware.

[0176] The above sequence for the steps of the method is for illustration only, and the steps of the method of the present invention are not limited to the sequence specifically described above unless otherwise specifically stated.

[0177] In addition, in some embodiments, the present invention can also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Therefore, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0178] In addition, the parts of the above technical solutions provided in the embodiments of the present invention that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0179] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A long graphite tube end processing control method based on servo drive and pneumatic centering, characterized in that: include: Obtain image data of the graphite tube, construct an edge detection algorithm to detect the graphite tube image, and obtain the position of the graphite tube; According to the position of the graphite tube, pneumatic centering is performed by adjusting the rubber inner support of the three-jaw chuck, and the position of the graphite tube is adjusted by the servo motor; Based on the historical processing parameters of the graphite tube, a deep learning algorithm is constructed and trained as a graphite tube processing parameter calculation model; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the graphite tube processing parameters; Based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, the servo motor controls the graphite tube for turning processing, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure according to the visual feedback in real time to maintain the stability of the graphite tube. After the graphite tube is processed, a visual inspection model is built to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified; An industrial camera is used to scan the graphite tube, and the size and spatial position information of the graphite tube are obtained through an edge detection algorithm; Scan the graphite tube with an industrial camera to obtain the original image , the original image Convert to grayscale ; Use Gaussian filter Smooth the image and remove noise interference: ,in represents the convolution operation, is the image after Gaussian filtering; Use Sobel operator to calculate the gradient of graphite tube image in x-axis and y-axis direction and ; Calculate the gradient amplitude and direction : ； Gradient Amplitude Perform non-maximum suppression to obtain refined edges ; Detect and connect edges using a dual threshold algorithm, defining a high threshold and low threshold ,like The pixel point is a strong edge. is non-marginal, The weak edge; The weak edges with strong edges in the pixel neighborhood of the graphite tube image are regarded as true edges, and the remaining weak edges are regarded as non-edges, and the final edge image of the graphite tube is obtained. ;extract The outline in ,in , Indicates the contour number, Indicates the number of edge contours; for graphite tube contours Fitting ellipse, the fitting formula is: ； in is the coordinate of the center of the graphite tube ellipse, is the half length of the major and minor axes, is the rotation angle; If the major and minor axes of the ellipse ,in is the ellipse approximation threshold, and the graphite tube cross section is regarded as a circle; the coordinates of the center of the circle are ,graphite tube radius ; Output the coordinates of the center of the circle and the radius of the graphite tube ; Position the graphite tube in space and calculate the center coordinates according to the camera's intrinsic parameter matrix. Convert to normalized coordinates ; The distance from the camera to the graphite tube and angle The external parameters are: is the pitch angle, is the yaw angle, is the roll angle; construct the transformation matrix from the camera coordinate system to the world coordinate system ,in is the rotation matrix, is the transpose of the zero vector, is the translation vector; Normalize the camera coordinates of the circle center Multiply , get the spatial coordinates of the center of the graphite tube .

2. The long graphite tube end processing control method based on servo drive and pneumatic centering according to claim 1 is characterized in that: Based on the graphite tube position information, the three-jaw chuck is adjusted to achieve pneumatic centering of the graphite tube to be processed; Get the center coordinates of the three-jaw chuck ; Calculate the deviation between the center of the graphite tube and the center of the three-jaw chuck : = , = ; According to the deviation between the center of the graphite tube and the center of the three-jaw chuck, calculate the cylinder stroke adjustment of the three-jaw chuck: ； in, For the The stroke adjustment of each cylinder, For the The angle between the cylinder and the x-axis, is the proportional coefficient of deviation and cylinder stroke; Control the extension and retraction of the three-jaw chuck cylinder to adjust the position of the three-jaw chuck's claws: ;in, For the Target stroke of each cylinder, is the initial stroke of the cylinder; Control the clamping force of the three-jaw chuck according to the radius of the graphite tube Calculate the required clamping force: ;in, is the target clamping force, is the initial clamping force, is the proportionality coefficient between radius and clamping force increment; Set the target clamping force Convert to the air pressure value of the three-jaw chuck cylinder : ;in, is the effective pressure area of ​​the three-jaw chuck; After adjusting the three-jaw chuck to achieve pneumatic centering, the three-dimensional spatial coordinates of the center of the graphite tube are obtained through an industrial camera. .

3. The long graphite tube end processing control method based on servo drive and pneumatic centering according to claim 2 is characterized in that: A three-axis servo drive is used to control the movement of the graphite tube. The three-axis servo drive includes an axial servo motor, a radial servo motor and a rotary servo motor. The axial servo motor is used for the axial movement of the graphite tube in the horizontal direction; the radial servo motor is used for the radial movement of the graphite tube in the vertical direction; the rotary servo motor is used for the rotary movement of the graphite tube; The horizontal two-dimensional coordinates of the center of the graphite tube obtained by the industrial camera , calculate the axial adjustment of the graphite tube in the horizontal direction: ;in, is the initial axial position of the graphite tube; Convert the axial adjustment to the target position of the graphite tube end point: ; ; Control the axial servo motor to drive the graphite tube to the target position ; After the three-jaw chuck is used to clamp the graphite tube, the three-dimensional spatial coordinates of the center of the graphite tube are obtained by the industrial camera: , calculate the radial adjustment of the graphite tube in the vertical direction: ;in, is the radial initial position of the graphite tube; controls the radial servo motor to drive the graphite tube to move to the target position : ; The rotary servo motor is controlled to transmit the motor rotational motion to the graphite tube, so that the graphite tube rotates.

4. The long graphite tube end processing control method based on servo drive and pneumatic centering according to claim 3 is characterized in that: Obtain historical processing parameters of graphite tubes of multiple specifications, including spindle speed , Feed speed , cutting depth And the tool corner radius ; Construct a graphite tube processing parameter calculation model based on the long short-term memory neural network LSTM model; Define the LSTM model input as the historical processing parameters and graphite tube size, expressed as ;in, is the spindle speed, is the feed speed, is the cutting depth, is the tool corner radius, is the radius of the graphite tube, is the processing length of graphite tube; Define the LSTM model output as the processing parameters of the graphite tube, expressed as ; is the spindle speed command, Feed speed command, is the cutting depth instruction; Constructing a network structure of an LSTM model, wherein the network structure of the LSTM model includes an input layer, a hidden layer, and a fully connected output layer, and using the Adam optimization algorithm to train the model; The mean square error (MSE) is used as the loss function, and the trained LSTM model is used in combination with the current size parameters of the graphite tube to generate the processing parameters of the graphite tube.

5. The long graphite tube end processing control method based on servo drive and pneumatic centering according to claim 4 is characterized in that: According to the trained graphite tube processing parameter calculation model, combined with the current graphite tube size information, the graphite tube processing parameters are output, including the servo motor spindle speed, feed speed and cutting depth; According to the cutting depth instruction, the position of the tool holder servo motor is controlled to adjust the cutting depth of the turning tool; during the processing, the position and vibration of the graphite tube are monitored in real time through an industrial camera, and the visual feedback information is transmitted to the three-jaw chuck cylinder for real-time pneumatic center adjustment.

6. The method for controlling the end of a long graphite tube based on servo drive and pneumatic centering according to claim 5, characterized in that: After the graphite tube is processed, an industrial camera is used to collect images of the processed graphite tube, and a graphite tube surface defect detection model and a dimension measurement model are constructed to perform surface defect detection and dimension detection on the processed graphite tube. For surface defect detection, a convolutional neural network (CNN) model is used. The graphite tube surface image is used as input. The texture and color features of the graphite tube surface are extracted through the convolution layer, and then classified through the fully connected layer to determine whether there are scratches, cracks, and pits on the graphite tube surface. For size detection, the convolutional neural network (CNN) model extracts edges and performs ellipse fitting on the graphite tube contour image to obtain the diameter and roundness size parameters of the graphite tube, and compares them with the design size to calculate whether the size error is within the allowable tolerance range. When the processed graphite tube has no surface defects and the size is qualified, the processed graphite tube is marked as a qualified product; When the processed graphite tube has any of surface defects and unqualified dimensions, the processed graphite tube is marked as a defective product.

7. A control system for machining the end of a long graphite tube based on servo drive and pneumatic centering, which is used to implement the control method for machining the end of a long graphite tube based on servo drive and pneumatic centering as claimed in any one of claims 1 to 6, characterized in that: include: Data acquisition module, processing control module, processing parameter calculation module, processing control module and quality inspection module; The data acquisition module acquires the image data of the graphite tube, constructs an edge detection algorithm to detect the image of the graphite tube, and acquires the position of the graphite tube; The processing control module adjusts the position of the graphite tube by adjusting the rubber inner support of the three-jaw chuck for pneumatic centering according to the position of the graphite tube and adjusting the position of the graphite tube by the servo motor; The processing parameter calculation module constructs and trains a deep learning algorithm as a graphite tube processing parameter calculation model based on the historical processing parameters of the graphite tube; the trained graphite tube processing parameter calculation model is combined with the current graphite tube information to output the processing parameters of the graphite tube; The processing control module controls the graphite tube to perform turning processing based on the graphite tube processing parameters output by the graphite tube processing parameter calculation model, and adjusts the feed speed and cutting depth in real time. The three-jaw chuck dynamically adjusts the internal support pressure according to visual feedback in real time to maintain the stability of the graphite tube. After the graphite tube is processed, the quality inspection module constructs a visual inspection model to inspect the surface quality and size of the graphite tube to determine whether the processed graphite tube is qualified.

8. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the long graphite tube end processing control method based on servo drive and pneumatic centering as described in any one of claims 1 to 6 by calling the computer program stored in the memory.

9. A storage medium, characterized in that: Instructions are stored, and when the instructions are run on a computer, the computer is caused to execute the long graphite tube end processing control method based on servo drive and pneumatic centering as described in any one of claims 1 to 6.

Images