Method, apparatus, device and medium for real-time adjusting the shape of an additive manufacturing molten pool

Abstract

This application belongs to the technical field of regulating the shape of a molten pool, and discloses a method, device, equipment and medium for real-time regulating the shape of an additive manufacturing molten pool. The method includes: obtaining a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image, constructing a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image, periodically obtaining a real-time molten pool image of the additive manufacturing molten pool, and inputting the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image containing only the additive manufacturing molten pool, calculating the real-time molten pool flow velocity through the approximate geometric center of the molten pool of adjacent frames of molten pool segmentation images, and regulating the shape of the additive manufacturing molten pool with the target molten pool shape as the target through a laser regulation formula; regulating the shape of the molten pool through the laser regulation formula, the real-time molten pool flow velocity and the target molten pool shape, thereby improving the accuracy and stability of the additive manufacturing molten pool.

Description

Technical Field

[0001] The present application relates to the technical field of regulating the shape of a molten pool, and more particularly, to a method, apparatus, device and medium for real-time regulating the shape of an additive manufacturing molten pool. Background Art

[0002] With the rapid development of additive manufacturing technology, the shape of an additive manufacturing molten pool has become an important factor affecting the quality of finished products and process stability. However, traditional control methods often cannot adapt to changes in process parameters in real time, resulting in an unsatisfactory shape of the molten pool (additive manufacturing molten pool), thereby affecting the performance and quality of products.

[0003] Currently, traditional molten pool adjustment methods mainly rely on experimental observation and analysis to obtain the width, height, area and flow rate of the molten pool, and then determine the shape of the molten pool based on the width, height, area and flow rate of the molten pool. However, this method is not only time-consuming and laborious, but also has certain subjectivity and limitations.

[0004] Therefore, in order to solve the technical problems of the existing molten pool adjustment methods being time-consuming and laborious and having certain subjectivity and limitations, there is an urgent need for a method, apparatus, device and medium for real-time regulating the shape of an additive manufacturing molten pool. Summary of the Invention

[0005] The purpose of the present application is to provide a method, apparatus, device and medium for real-time regulating the shape of an additive manufacturing molten pool. By means of a preset laser adjustment formula, according to the real-time molten pool flow rate calculated from the approximate geometric center of the molten pool in the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula, the shape of the molten pool is adjusted in real time, solving the problems of the existing molten pool adjustment methods being time-consuming and laborious and having certain subjectivity and limitations, being able to monitor the additive manufacturing molten pool in real time and accurately and adjust the shape of the additive manufacturing molten pool, and improving the adjustment efficiency of the additive manufacturing molten pool.

[0006] In a first aspect, the present application provides a method for real-time regulating the shape of an additive manufacturing molten pool, including:

[0007] Obtaining a molten pool image of an additive manufacturing molten pool and a corresponding molten pool annotation image;

[0008] Constructing a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image;

[0009] Periodically obtaining a real-time molten pool image of the additive manufacturing molten pool, and inputting the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image containing only the additive manufacturing molten pool;

[0010] Determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculate the real-time molten pool flow rate of the additive manufacturing molten pool based on the approximate geometric centers of the molten pool segmentation images of adjacent frames and in combination with a preset flow rate calculation formula;

[0011] Utilize a preset laser adjustment formula, based on the real-time molten pool flow rate, in combination with a preset PID controller, and with a preset target molten pool shape as the target, to perform real-time adjustment on the laser information for preparing the additive manufacturing molten pool, so as to real-time adjust the molten pool shape of the additive manufacturing molten pool.

[0012] The method for real-time adjusting the shape of the additive manufacturing molten pool provided by this application can achieve the adjustment of the shape of the additive manufacturing molten pool. Through the preset laser adjustment formula, based on the real-time molten pool flow rate calculated from the approximate geometric centers of the molten pool segmentation images of adjacent frames and the preset flow rate calculation formula, the molten pool shape is real-time adjusted, solving the problems of time-consuming and laborious existing molten pool adjustment methods and having certain subjectivity and limitations, being able to monitor the additive manufacturing molten pool in real time and accurately and adjust the molten pool shape of the additive manufacturing molten pool, improving the adjustment efficiency of the additive manufacturing molten pool.

[0013] Optionally, obtaining the molten pool image and the corresponding molten pool annotation image of the additive manufacturing molten pool includes:

[0014] Take a picture of the additive manufacturing molten pool to obtain the molten pool image;

[0015] Mark the contour position of the additive manufacturing molten pool from the molten pool image to obtain the molten pool annotation image.

[0016] Optionally, constructing a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image includes:

[0017] Construct the YOLOv8 model corresponding to the molten pool image to obtain a preliminary molten pool image segmentation model;

[0018] Train the preliminary molten pool image segmentation model according to the molten pool images for training and the corresponding molten pool annotation images in the molten pool image to obtain a trained preliminary molten pool image segmentation model;

[0019] Verify the trained preliminary molten pool image segmentation model based on the molten pool images not used for training and the corresponding molten pool annotation images in the molten pool image to obtain the molten pool image segmentation model.

[0020] The method for real-time adjusting the shape of the additive manufacturing molten pool provided by this application can adjust the shape of the additive manufacturing molten pool. By using the molten pool image and the corresponding molten pool annotation image, the YOLOv8 model is constructed and trained to obtain a molten pool image segmentation model, which can quickly identify the position of the additive manufacturing molten pool and extract the image features outside the molten pool position to obtain an image containing only the molten pool, which is beneficial to improving the adjustment efficiency of the additive manufacturing molten pool.

[0021] Optionally, training the preliminary molten pool image segmentation model according to the molten pool image for training and the corresponding molten pool annotation image in the molten pool image to obtain the trained preliminary molten pool image segmentation model includes:

[0022] Input the molten pool image for training in the molten pool image into the preliminary molten pool image segmentation model to obtain a corresponding image output; the image output is an image containing only the additive manufacturing molten pool;

[0023] Determine the training error according to the molten pool annotation image corresponding to the molten pool image for training in the molten pool image and the corresponding image output;

[0024] Based on the training error, adjust the parameters of the preliminary molten pool image segmentation model to obtain the optimal parameters, and use the optimal parameters to optimize the preliminary molten pool image segmentation model to obtain the trained preliminary molten pool image segmentation model.

[0025] Optionally, before inputting the molten pool image for training in the molten pool image into the preliminary molten pool image segmentation model to obtain a corresponding image output, it further includes:

[0026] Initialize the parameters of the preliminary molten pool image segmentation model.

[0027] Optionally, determining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and according to the approximate geometric centers of the molten pool segmentation images of adjacent frames, combining a preset flow rate calculation formula, calculating the real-time molten pool flow rate of the additive manufacturing molten pool includes steps executed in a loop:

[0028] Determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image according to the minimum circumscribed rectangle of the additive manufacturing molten pool corresponding to the molten pool segmentation image;

[0029] After obtaining the first frame of the molten pool segmentation image, when each time the molten pool segmentation image is obtained, through the preset flow rate calculation formula, combining the approximate geometric center of the molten pool segmentation image obtained this time with the approximate geometric center of the molten pool segmentation image obtained last time, calculate the real-time molten pool flow rate of the additive manufacturing molten pool.

[0030] The method for real-time adjusting the shape of an additive manufacturing molten pool provided by this application can achieve the adjustment of the shape of the additive manufacturing molten pool. By using a preset flow rate calculation formula, the real-time molten pool flow rate of the additive manufacturing molten pool is calculated using the approximate geometric center of the molten pool in the molten pool segmentation image obtained this time and the approximate geometric center of the molten pool in the molten pool segmentation image obtained last time. The shape of the additive manufacturing molten pool can be adjusted through the real-time molten pool flow rate, which is beneficial to improving the adjustment efficiency of the additive manufacturing molten pool.

[0031] Optionally, the preset laser adjustment formula includes a preset laser power adjustment formula, a preset laser powder feeding rate adjustment formula, a preset laser radius adjustment formula, and a preset laser scanning speed adjustment formula.

[0032] In a second aspect, this application provides a device for real-time adjusting the shape of an additive manufacturing molten pool, including:

[0033] An acquisition module, configured to acquire a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image;

[0034] A construction module, configured to construct a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image;

[0035] An extraction module, configured to periodically acquire a real-time molten pool image of the additive manufacturing molten pool, and input the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image that only contains the additive manufacturing molten pool;

[0036] A calculation module, configured to determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculate the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames, in combination with a preset flow rate calculation formula;

[0037] An adjustment module, configured to use a preset laser adjustment formula, according to the real-time molten pool flow rate, in combination with a preset PID controller, with a preset target molten pool shape as the target, to perform real-time adjustment on the laser information for preparing the additive manufacturing molten pool, so as to real-time adjust the shape of the additive manufacturing molten pool.

[0038] This device for real-time adjusting the shape of an additive manufacturing molten pool performs real-time adjustment on the molten pool shape through a preset laser adjustment formula, according to the real-time molten pool flow rate calculated from the approximate geometric centers of the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula, solves the problems of time-consuming and laborious existing molten pool adjustment methods and certain subjectivity and limitations, can monitor the additive manufacturing molten pool in real time and accurately, and adjust the shape of the additive manufacturing molten pool, thereby improving the adjustment efficiency of the additive manufacturing molten pool.

[0039] In a third aspect, the present application provides an electronic device, including a processor and a memory. The memory stores a computer program executable by the processor. When the processor executes the computer program, it runs the steps in the method for real-time adjusting the shape of an additive manufacturing molten pool as described above.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it runs the steps in the method for real-time adjusting the shape of an additive manufacturing molten pool as described above.

[0041] Beneficial effects: The method, device, equipment, and medium for real-time adjusting the shape of an additive manufacturing molten pool provided by the present application, through a preset laser adjustment formula, according to the real-time molten pool flow rate calculated from the approximate geometric center of the molten pool in the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula, perform real-time adjustment on the shape of the molten pool, solve the problems of time-consuming and laborious existing molten pool adjustment methods and certain subjectivity and limitations, and can monitor the additive manufacturing molten pool in real time and accurately and adjust the shape of the additive manufacturing molten pool, improving the adjustment efficiency of the additive manufacturing molten pool. Description of the Drawings

[0042] Figure 1 It is a flowchart of the method for real-time adjusting the shape of an additive manufacturing molten pool provided by an embodiment of the present application.

[0043] Figure 2 It is a schematic structural diagram of the device for real-time adjusting the shape of an additive manufacturing molten pool provided by an embodiment of the present application.

[0044] Figure 3 It is a schematic structural diagram of the electronic device provided by an embodiment of the present application.

[0045] Label description: 1. Acquisition module; 2. Construction module; 3. Extraction module; 4. Calculation module; 5. Adjustment module; 301. Processor; 302. Memory; 303. Communication bus. Detailed Embodiments

[0046] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0047] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0048] Please refer to Figure 1 , Figure 1 which is a method for real-time adjusting the shape of an additive manufacturing molten pool in some embodiments of this application, used to adjust the shape of the additive manufacturing molten pool, including the steps of:

[0049] Step S101, obtaining a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image;

[0050] Step S102, constructing a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image;

[0051] Step S103, periodically obtaining a real-time molten pool image of the additive manufacturing molten pool, and inputting the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image that only contains the additive manufacturing molten pool;

[0052] Step S104, determining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculating the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula;

[0053] Step S105, using a preset laser adjustment formula, according to the real-time molten pool flow rate, in combination with a preset PID controller, aiming at a preset target molten pool shape, to perform real-time adjustment on the laser information for preparing the additive manufacturing molten pool, so as to real-time adjust the shape of the additive manufacturing molten pool.

[0054] This method for real-time adjusting the shape of the additive manufacturing molten pool performs real-time adjustment on the molten pool shape through a preset laser adjustment formula according to the real-time molten pool flow rate calculated from the approximate geometric centers of the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula, solves the problems of time-consuming and laborious existing molten pool adjustment methods and certain subjectivity and limitations, can monitor the additive manufacturing molten pool in real time and accurately and adjust the shape of the additive manufacturing molten pool, and improves the adjustment efficiency of the additive manufacturing molten pool.

[0055] Specifically, in step S101, obtaining a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image includes:

[0056] Taking a picture of the additive manufacturing molten pool to obtain a molten pool image;

[0057] Mark the contour position of the additive manufacturing molten pool from the molten pool image to obtain the molten pool marked image.

[0058] In step S101, first, an additive manufacturing molten pool is prepared by laser technology, and then, using existing imaging techniques (such as cameras or video cameras), the additive manufacturing molten pool is imaged from multiple angles with the additive manufacturing molten pool as the center to obtain the molten pool image of the additive manufacturing molten pool. By manual annotation methods or mechanical annotation methods (such as using labelme software), the contour position of the additive manufacturing molten pool is marked in the molten pool image to obtain the molten pool marked image; thus, a large number of molten pool images and molten pool marked images are obtained for training the model, so that the trained model can identify the molten pool position to extract molten pool features.

[0059] Specifically, in step S102, according to the molten pool image and the corresponding molten pool marked image, a molten pool image segmentation model is constructed, including:

[0060] Construct a YOLOv8 model corresponding to the molten pool image to obtain a preliminary molten pool image segmentation model;

[0061] According to the molten pool images for training in the molten pool image and the corresponding molten pool marked images, train the preliminary molten pool image segmentation model to obtain the trained preliminary molten pool image segmentation model;

[0062] Based on the molten pool images not used for training in the molten pool image and the corresponding molten pool marked images, verify the trained preliminary molten pool image segmentation model to obtain the molten pool image segmentation model.

[0063] In step S102, use the molten pool image to construct a YOLOv8 model to obtain a preliminary molten pool image segmentation model. By inputting the molten pool image into the preliminary molten pool image segmentation model, irrelevant features such as the background or additive manufacturing equipment in the molten pool image can be removed (i.e., the location of the additive manufacturing molten pool is segmented from the molten pool image), and the target feature of the additive manufacturing molten pool is extracted to obtain an image containing only the additive manufacturing molten pool. Among them, the YOLOv8 model is a prior art and will not be elaborated here.

[0064] Specifically, in step S102, according to the molten pool images for training in the molten pool image and the corresponding molten pool marked images, train the preliminary molten pool image segmentation model to obtain the trained preliminary molten pool image segmentation model, including:

[0065] Input the molten pool images for training in the molten pool image into the preliminary molten pool image segmentation model to obtain the corresponding image output; the image output is an image containing only the additive manufacturing molten pool;

[0066] Determine the training error according to the molten pool marked images corresponding to the molten pool images for training in the molten pool image and the corresponding image output.

[0067] Based on the training error, adjust the parameters of the preliminary molten pool image segmentation model to obtain the optimal parameters, and use the optimal parameters to optimize the preliminary molten pool image segmentation model to obtain the trained preliminary molten pool image segmentation model.

[0068] In step S102, by comparing the molten pool annotation image corresponding to the molten pool image used for training with the image output obtained by inputting the molten pool image into the preliminary molten pool image segmentation model, the training error is obtained. For example, the degree of size deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output (such as the size of the corresponding molten pool in the image output is inconsistent with the molten pool contour area annotated in the molten pool annotation image), or the degree of deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output (such as the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output are not in the same area), etc. Using the training error, adjust the parameters of the preliminary molten pool image segmentation model. For example, if the training error is the degree of size deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output, modify the corresponding model parameters to make the corresponding molten pool in the image output more accurate, obtain the optimal parameters, and optimize the preliminary molten pool image segmentation model through the optimal parameters to obtain the trained preliminary molten pool image segmentation model.

[0069] In step S102, input the molten pool image that has not been used for training in the molten pool image into the trained preliminary molten pool image segmentation model to obtain the corresponding image output. Use this image output as verification data, and compare the verification data with the molten pool annotation image corresponding to the molten pool image that has not been used for training to determine whether the error is within the acceptable range (the acceptable range of error is generally 0 to 3%, which can be modified according to actual needs), verify the accuracy of the trained preliminary molten pool image segmentation model, and obtain the molten pool image segmentation model.

[0070] Specifically, in step S102, before inputting the molten pool image used for training in the molten pool image into the preliminary molten pool image segmentation model to obtain the corresponding image output, it further includes:

[0071] Initialize the parameters of the preliminary molten pool image segmentation model.

[0072] In step S102, before training the model with the input data, it is necessary to initialize the parameters of the preliminary molten pool image segmentation model to ensure that the model is in a normal use state.

[0073] Specifically, in step S103, taking a fixed position directly above the additive manufacturing molten pool at a preset shooting time interval as a cycle, the additive manufacturing molten pool is photographed, and real-time molten pool images of the additive manufacturing molten pool are obtained periodically. The real-time molten pool images are input into the molten pool image segmentation model. Through the calculation of the molten pool image segmentation model, a molten pool segmentation image containing only the additive manufacturing molten pool is extracted. Among them, the preset shooting time can be set according to actual needs.

[0074] Specifically, in step S104, the approximate geometric center of the molten pool corresponding to the molten pool segmentation image is determined, and based on the approximate geometric centers of the molten pool segmentation images of adjacent frames, combined with a preset flow velocity calculation formula, the real-time molten pool flow velocity of the additive manufacturing molten pool is calculated, including the steps executed in a loop:

[0075] According to the minimum circumscribed rectangle of the additive manufacturing molten pool corresponding to the molten pool segmentation image, the approximate geometric center of the molten pool corresponding to the molten pool segmentation image is determined;

[0076] After obtaining the first frame of the molten pool segmentation image, when each molten pool segmentation image is obtained, through a preset flow velocity calculation formula, combined with the approximate geometric center of the molten pool of the currently obtained molten pool segmentation image and the approximate geometric center of the molten pool of the previously obtained molten pool segmentation image, the real-time molten pool flow velocity of the additive manufacturing molten pool is calculated.

[0077] In step S104, when each molten pool segmentation image is obtained (except for the first time of obtaining the molten pool segmentation image), taking the additive manufacturing molten pool in the molten pool segmentation image as the center to make a minimum circumscribed rectangle, obtaining the height and width of the minimum circumscribed rectangle, determining the height and width of the minimum circumscribed rectangle as the height and width of the additive manufacturing molten pool, and determining the midpoint of the minimum circumscribed rectangle as the approximate center point of the additive manufacturing molten pool, to obtain the approximate geometric center of the molten pool corresponding to the molten pool segmentation image.

[0078] After each determination of the approximate geometric center of the molten pool (except for determining the approximate geometric center of the first frame of the molten pool segmentation image), using a preset flow velocity calculation formula, according to the coordinate values of the approximate geometric center of the molten pool of the currently obtained molten pool segmentation image and the coordinate values of the approximate geometric center of the molten pool of the previously obtained molten pool segmentation image, the real-time molten pool flow velocity of the additive manufacturing molten pool is calculated.

[0079] Among them, the preset flow velocity calculation formula is specifically:

[0080] ;

[0081] Among them, is the real-time molten pool flow velocity; is the abscissa of the approximate geometric center of the molten pool at time t (the current moment); is the abscissa of the approximate geometric center of the molten pool at time t-1 (the previous moment); is the ordinate of the approximate geometric center of the molten pool at time t; is the ordinate of the approximate geometric center of the molten pool at time t-1; is the time interval between two frames of molten pool segmentation images, and the unit is set to milliseconds; is the interference coefficient, used to represent the influence of external temperature, air flow and pressure on the flow rate of the molten pool; is the adjustment coefficient of the molten pool width and height on the flow rate of the molten pool; and are the width and height (in millimeters) of the molten pool at time t respectively; and are the width and height (in millimeters) of the molten pool at time t-1 respectively; is the influence of environmental noise at time t. Among them, the interference coefficient , the adjustment coefficient , and the influence of environmental noise can be set according to actual needs.

[0082] Specifically, in step 105, the preset laser adjustment formulas include a preset laser power adjustment formula, a preset laser powder feeding rate adjustment formula, a preset laser radius adjustment formula, and a preset laser scanning speed adjustment formula; using the preset laser adjustment formulas, according to the real-time molten pool flow rate, combined with a preset PID controller, with a preset target molten pool shape as the goal, calculate the real-time adjusted laser data, that is, calculate the real-time adjusted laser power, laser powder feeding rate, laser radius, and laser scanning speed. Among them, the preset target molten pool shape includes the target molten pool area, the target molten pool height, and the target molten pool width; the laser power is adjusted based on the molten pool area, the laser powder feeding rate is adjusted based on the molten pool area, the laser radius is adjusted based on the molten pool height, and the laser scanning speed is adjusted based on the molten pool width.

[0083] Among them, the specific preset laser power adjustment formula is:

[0084] ;

[0085] Among them, is the adjusted laser power (unit: watt); is the current laser power; , , are the laser power adjustment proportional coefficient, laser power adjustment integral coefficient, and laser power adjustment differential coefficient of the PID controller respectively, used to feedback and adjust the laser power according to the real-time molten pool area; is the target molten pool area (unit: square millimeter); is the current molten pool area (the additive manufacturing molten pool in the segmented molten pool image is segmented by pixel points, for example, the pixel points are segmented in square millimeters, and the current molten pool area is determined according to the proportion of the pixel points corresponding to the additive manufacturing molten pool in the pixel points of the corresponding minimum circumscribed rectangle. This calculation process can be set in the molten pool image segmentation model, so that the molten pool image segmentation model automatically outputs the height, width and area of the additive manufacturing molten pool when extracting the segmented molten pool image, that is, the height of the current molten pool, the width of the current molten pool and the current molten pool area); represents the adjustment coefficient of the current molten pool area and height to the laser power; represents the adjustment coefficient of the current molten pool flow rate and width to the laser power; is the width of the current molten pool (in millimeters); is the height of the current molten pool (in millimeters); is the current molten pool flow rate (i.e., the real-time molten pool flow rate, in millimeters per second); is the current molten pool temperature (in degrees Celsius); is the adjustment coefficient of the current molten pool temperature; is the adjustment coefficient of the current molten pool area; is the exponential coefficient for adjusting the influence of the current molten pool flow rate on the power. Among them, the laser power adjustment proportional coefficient the laser power adjustment integral coefficient the laser power adjustment differential coefficient the adjustment coefficient the adjustment coefficient the adjustment coefficient the adjustment coefficient the adjustment coefficient can be set according to actual needs.

[0086] The preset laser powder feeding rate adjustment formula is specifically:

[0087] ;

[0088] Among them, is the adjusted laser powder feeding rate (in grams per minute); is the current laser powder feeding rate (grams per minute); , , are the proportional, integral and differential coefficients of the PID controller respectively, used to feedback according to the real-time molten pool area to adjust the laser powder feeding rate; represents the adjustment coefficient of the current molten pool area and width to the laser powder feeding rate; represents the adjustment coefficient of the current molten pool height and flow rate to the laser powder feeding rate; Among them, the laser powder feeding rate adjustment proportional coefficient , Integral coefficient for adjusting laser powder feeding rate , Differential coefficient for adjusting laser powder feeding rate , Adjustment coefficient , Adjustment coefficient Can be set according to actual needs.

[0089] The specific formula for adjusting laser radius is:

[0090] ;

[0091] Among them, is the adjusted laser radius (unit: millimeter); is the current laser radius; is the target molten pool height; is the height of the current molten pool; , , are respectively the proportional coefficient, integral coefficient, and differential coefficient for adjusting laser radius of the PID controller, used to adjust the laser radius according to the real-time molten pool height for feedback; represents the adjustment coefficient of the current molten pool width and area on the laser radius; represents the adjustment coefficient of the current molten pool height and flow rate on the laser radius; among them, the proportional coefficient for adjusting laser radius , integral coefficient for adjusting laser radius , differential coefficient for adjusting laser radius , adjustment coefficient , adjustment coefficient Can be set according to actual needs.

[0092] The specific formula for adjusting laser scanning speed is:

[0093] ;

[0094] Among them, is the adjusted laser scanning speed (unit: millimeter per second); is the current laser scanning speed; is the target molten pool width; is the width of the current molten pool; , , are respectively the proportional coefficient, integral coefficient, and differential coefficient for adjusting laser scanning speed of the PID controller, used to adjust the laser scanning speed according to the real-time molten pool width for feedback; represents the adjustment coefficient of the current molten pool height and area on the laser radius; Represents the adjustment coefficient of the current molten pool flow rate and width on the laser radius; among them, the laser scanning speed adjustment proportional coefficient , the laser scanning speed adjustment integral coefficient , the laser scanning speed adjustment differential coefficient , the adjustment coefficient , the adjustment coefficient can be set according to actual needs. In practical applications, the laser scanning speed adjustment proportional coefficient , the laser scanning speed adjustment integral coefficient and the laser scanning speed adjustment differential coefficient can all be set to negative values. When adjusting the laser scanning speed, the negative coefficient can reversely adjust the control output of the PID controller (i.e., reversely adjust the laser scanning speed), so as to make the laser scanning speed negatively correlated with the molten pool size. Thus, according to the laser scanning speed adjustment formula, by adjusting the laser scanning speed, the control of the molten pool width can be achieved.

[0095] In summary, the laser power, laser powder feeding rate, laser radius, and laser scanning speed after real-time adjustment are calculated, and the laser information is adjusted using the laser power, laser powder feeding rate, laser radius, and laser scanning speed after real-time adjustment to adjust the molten pool shape of the additive manufacturing molten pool.

[0096] In some embodiments, through the prior art, the preset flow rate calculation formula can be set in the molten pool image segmentation model to obtain a molten pool flow rate calculation model (when constructing a numerical calculation model for molten pool stability, the molten pool flow rate calculated by the flow rate calculation formula needs to be obtained as training data and verification data). The molten pool flow rate calculation model has the ability to calculate the molten pool flow rate and can calculate the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric center of the molten pool in two frames of real-time molten pool images obtained periodically. Then, through the PID controller, using the preset laser power adjustment formula and the preset laser powder feeding rate adjustment formula, the molten pool shape of the additive manufacturing molten pool is adjusted in real time.

[0097] As described above, the method for real-time adjusting the shape of an additive manufacturing molten pool obtains the molten pool image and the corresponding molten pool annotation image of the additive manufacturing molten pool, constructs a molten pool image segmentation model based on the molten pool image and the corresponding molten pool annotation image, periodically obtains the real-time molten pool image of the additive manufacturing molten pool, inputs the real-time molten pool image into the molten pool image segmentation model, extracts the molten pool segmentation image containing only the additive manufacturing molten pool, determines the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and combines the approximate geometric centers of the molten pool segmentation images of adjacent frames with a preset flow rate calculation formula to calculate the real-time molten pool flow rate of the additive manufacturing molten pool. Using a preset laser adjustment formula, based on the real-time molten pool flow rate and in combination with a preset PID controller, with a preset target molten pool shape as the target, the laser information for preparing the additive manufacturing molten pool is adjusted in real time to adjust the shape of the additive manufacturing molten pool in real time. Therefore, through the preset laser adjustment formula, according to the real-time molten pool flow rate calculated from the approximate geometric centers of the molten pool segmentation images of adjacent frames and the preset flow rate calculation formula, the shape of the molten pool is adjusted in real time, solving the problems of time-consuming and laborious existing molten pool adjustment methods and certain subjectivity and limitations. It can monitor the additive manufacturing molten pool in real time and accurately and adjust the shape of the additive manufacturing molten pool, improving the adjustment efficiency of the additive manufacturing molten pool.

[0098] Reference Figure 2 , this application provides a device for real-time adjusting the shape of an additive manufacturing molten pool, which is used to adjust the shape of the additive manufacturing molten pool and includes:

[0099] An acquisition module 1, configured to acquire the molten pool image and the corresponding molten pool annotation image of the additive manufacturing molten pool;

[0100] A construction module 2, configured to construct a molten pool image segmentation model based on the molten pool image and the corresponding molten pool annotation image;

[0101] An extraction module 3, configured to periodically acquire the real-time molten pool image of the additive manufacturing molten pool, input the real-time molten pool image into the molten pool image segmentation model, and extract the molten pool segmentation image containing only the additive manufacturing molten pool;

[0102] A calculation module 4, configured to determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and combine the approximate geometric centers of the molten pool segmentation images of adjacent frames with a preset flow rate calculation formula to calculate the real-time molten pool flow rate of the additive manufacturing molten pool;

[0103] An adjustment module 5, configured to use a preset laser adjustment formula, based on the real-time molten pool flow rate and in combination with a preset PID controller, with a preset target molten pool shape as the target, to adjust the laser information for preparing the additive manufacturing molten pool in real time to adjust the shape of the additive manufacturing molten pool in real time.

[0104] The device for real-time adjusting the shape of the additive manufacturing molten pool adjusts the shape of the molten pool in real time according to the preset laser adjustment formula and the real-time molten pool flow rate calculated based on the approximate geometric center of the molten pool in the molten pool segmentation images of adjacent frames and the preset flow rate calculation formula, solving the problems of time-consuming and laborious existing molten pool adjustment methods and certain subjectivity and limitations, being able to monitor the additive manufacturing molten pool in real time and accurately and adjust the shape of the additive manufacturing molten pool, and improving the adjustment efficiency of the additive manufacturing molten pool.

[0105] Specifically, when the acquisition module 1 acquires the molten pool image and the corresponding molten pool annotation image of the additive manufacturing molten pool, it executes:

[0106] Take pictures of the additive manufacturing molten pool to obtain a molten pool image;

[0107] Mark the contour position of the additive manufacturing molten pool from the molten pool image to obtain a molten pool annotation image.

[0108] When the acquisition module 1 executes, first prepare the additive manufacturing molten pool through laser technology, and then use existing shooting technologies (such as cameras or video cameras) to take pictures of the additive manufacturing molten pool from multiple angles with the additive manufacturing molten pool as the center to obtain the molten pool image of the additive manufacturing molten pool. Mark the contour position of the additive manufacturing molten pool in the molten pool image through manual annotation methods or mechanical annotation methods (such as using labelme software) to obtain a molten pool annotation image; thus, a large number of molten pool images and molten pool annotation images are obtained for training the model, so that the trained model can identify the molten pool position to extract molten pool features.

[0109] Specifically, when the construction module 2 constructs a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image, it executes:

[0110] Construct a YOLOv8 model corresponding to the molten pool image to obtain a preliminary molten pool image segmentation model;

[0111] Train the preliminary molten pool image segmentation model according to the molten pool images and the corresponding molten pool annotation images used for training in the molten pool image to obtain a trained preliminary molten pool image segmentation model;

[0112] Verify the trained preliminary molten pool image segmentation model based on the molten pool images and the corresponding molten pool annotation images not used for training in the molten pool image to obtain a molten pool image segmentation model.

[0113] When the construction module 2 is executed, it utilizes the molten pool image to construct a YOLOv8 model, obtaining a preliminary molten pool image segmentation model. By inputting the molten pool image into the preliminary molten pool image segmentation model, irrelevant features such as the background or additive manufacturing equipment in the molten pool image can be removed (i.e., the location of the additive manufacturing molten pool is segmented from the molten pool image), and the target feature of the additive manufacturing molten pool is extracted to obtain an image containing only the additive manufacturing molten pool. Among them, the YOLOv8 model is a prior art and will not be elaborated here.

[0114] Specifically, when the construction module 2 trains the preliminary molten pool image segmentation model according to the molten pool image for training in the molten pool image and the corresponding molten pool annotation image, and obtains the trained preliminary molten pool image segmentation model, it executes:

[0115] Input the molten pool image for training in the molten pool image into the preliminary molten pool image segmentation model to obtain the corresponding image output; the image output is an image containing only the additive manufacturing molten pool;

[0116] Determine the training error according to the molten pool annotation image corresponding to the molten pool image for training in the molten pool image and the corresponding image output;

[0117] Based on the training error, adjust the parameters of the preliminary molten pool image segmentation model to obtain the optimal parameters, and use the optimal parameters to optimize the preliminary molten pool image segmentation model to obtain the trained preliminary molten pool image segmentation model.

[0118] When the construction module 2 is executed, by comparing the molten pool annotation image corresponding to the molten pool image for training with the image output obtained by inputting the molten pool image into the preliminary molten pool image segmentation model, the training error is obtained, such as the degree of size deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output (for example, the size of the corresponding molten pool in the image output is inconsistent with the molten pool contour area annotated in the molten pool annotation image), or the degree of deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output (for example, the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output are not in the same area), etc. Using the training error, adjust the parameters of the preliminary molten pool image segmentation model. If the training error is the degree of size deviation between the molten pool contour area annotated in the molten pool annotation image and the corresponding molten pool in the image output, then modify the corresponding model parameters to make the corresponding molten pool in the image output more accurate, obtain the optimal parameters, and optimize the preliminary molten pool image segmentation model through the optimal parameters to obtain the trained preliminary molten pool image segmentation model.

[0119] When the construction module 2 is executed, the molten pool images in the molten pool images that are not used for training are input into the trained preliminary molten pool image segmentation model to obtain corresponding image outputs. The image outputs are used as verification data, and by comparing the verification data with the molten pool annotation images corresponding to the molten pool images not used for training, it is determined whether the error is within an acceptable range (the acceptable range of error is generally 0 to 3%, which can be modified according to actual needs), and the accuracy of the trained preliminary molten pool image segmentation model is verified to obtain the molten pool image segmentation model.

[0120] Specifically, before the construction module 2 inputs the molten pool images used for training in the molten pool images into the preliminary molten pool image segmentation model to obtain corresponding image outputs, it performs:

[0121] Initialize the parameters of the preliminary molten pool image segmentation model.

[0122] When the construction module 2 is executed, before training the model with input data, it is necessary to initialize the parameters of the preliminary molten pool image segmentation model to ensure that the model is in a normal use state.

[0123] Specifically, when the extraction module 3 is executed, taking the preset shooting time interval as a cycle, the additive manufacturing molten pool is photographed at a fixed position directly above the additive manufacturing molten pool, the real-time molten pool images of the additive manufacturing molten pool are periodically obtained, and the real-time molten pool images are input into the molten pool image segmentation model. After calculation by the molten pool image segmentation model, the molten pool segmentation images containing only the additive manufacturing molten pool are extracted. Among them, the preset shooting time can be set according to actual needs.

[0124] Specifically, when the calculation module 4 determines the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculates the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames, in combination with the preset flow rate calculation formula, it executes in a loop:

[0125] Determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculate the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames, in combination with the preset flow rate calculation formula, including the steps executed in a loop:

[0126] According to the minimum bounding rectangle of the additive manufacturing molten pool corresponding to the molten pool segmentation image, determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image;

[0127] After obtaining the first frame of the molten pool segmentation image, when each molten pool segmentation image is obtained, through the preset flow rate calculation formula, in combination with the approximate geometric center of the molten pool segmentation image obtained this time and the approximate geometric center of the molten pool segmentation image obtained last time, the real-time molten pool flow rate of the additive manufacturing molten pool is calculated.

[0128] When the calculation module 4 is executing, except for the first acquisition of the molten pool segmentation image, each time the molten pool segmentation image is acquired, a minimum bounding rectangle is formed with the additive manufacturing molten pool in the molten pool segmentation image as the center, the height and width of the minimum bounding rectangle are obtained, the height and width of the minimum bounding rectangle are determined as the height and width of the additive manufacturing molten pool, and the midpoint of the minimum bounding rectangle is determined as the approximate center point of the additive manufacturing molten pool, thereby obtaining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image.

[0129] After each determination of the approximate geometric center of the molten pool (except for determining the approximate geometric center of the molten pool in the first frame of the molten pool segmentation image), using a preset flow velocity calculation formula, based on the coordinate values of the approximate geometric center of the molten pool in the molten pool segmentation image acquired this time and the coordinate values of the approximate geometric center of the molten pool in the molten pool segmentation image acquired last time, the real-time molten pool flow velocity of the additive manufacturing molten pool is calculated.

[0130] Among them, the preset flow velocity calculation formula is specifically:

[0131] ;

[0132] Among them, is the real-time molten pool flow velocity; is the abscissa of the approximate geometric center of the molten pool at time t (the current time); is the abscissa of the approximate geometric center of the molten pool at time t - 1 (the previous time); is the ordinate of the approximate geometric center of the molten pool at time t; is the ordinate of the approximate geometric center of the molten pool at time t - 1; is the time interval between two frames of molten pool segmentation images, and the unit is set to milliseconds; is the interference coefficient, which is used to represent the influence of external temperature, air flow and pressure on the molten pool flow velocity; is the adjustment coefficient of the molten pool width and height on the molten pool flow velocity; and are respectively the width and height (in millimeters) of the molten pool at time t; and are respectively the width and height (in millimeters) of the molten pool at time t - 1; is the environmental noise influence at time t. Among them, the interference coefficient , the adjustment coefficient , and the environmental noise influence can be set according to actual needs.

[0133] Specifically, when the adjustment module 5 is executing, the preset laser adjustment formulas include a preset laser power adjustment formula, a preset laser powder feeding rate adjustment formula, a preset laser radius adjustment formula, and a preset laser scanning speed adjustment formula; using the preset laser adjustment formulas, based on the real-time molten pool flow rate, in combination with a preset PID controller, with the preset target molten pool shape as the target, the real-time adjusted laser data is calculated, that is, the real-time adjusted laser power, laser powder feeding rate, laser radius, and laser scanning speed are calculated. Among them, the preset target molten pool shape includes the target molten pool area, target molten pool height, and target molten pool width; the laser power is adjusted based on the molten pool area, the laser powder feeding rate is adjusted based on the molten pool area, the laser radius is adjusted based on the molten pool height, and the laser scanning speed is adjusted based on the molten pool width.

[0134] Among them, the specific preset laser power adjustment formula is:

[0135] ;

[0136] Among them, is the adjusted laser power (in watts); is the current laser power; , , are respectively the laser power adjustment proportional coefficient, laser power adjustment integral coefficient, and laser power adjustment differential coefficient of the PID controller, which are used to feedback and adjust the laser power according to the real-time molten pool area; is the target molten pool area (in square millimeters); is the current molten pool area (the additive manufacturing molten pool in the molten pool segmentation image is segmented by pixel points, such as segmenting the pixel points in square millimeters, and determining the current molten pool area according to the pixel point proportion of the additive manufacturing molten pool corresponding pixel points in the corresponding minimum circumscribed rectangle. This calculation process can be set in the molten pool image segmentation model, so that the molten pool image segmentation model automatically outputs the height, width, and area of the additive manufacturing molten pool when extracting the molten pool segmentation image, that is, the height of the current molten pool, the width of the current molten pool, and the current molten pool area); represents the adjustment coefficient of the current molten pool area and height on the laser power; represents the adjustment coefficient of the current molten pool flow rate and width on the laser power; is the width of the current molten pool (in millimeters); is the height of the current molten pool (in millimeters); is the current molten pool flow rate (i.e., the real-time molten pool flow rate, in millimeters per second); is the current molten pool temperature (in degrees Celsius); is the adjustment coefficient of the current molten pool temperature; is the adjustment coefficient of the current molten pool area; is the exponential coefficient for adjusting the influence of the current molten pool flow rate on power. Among them, the laser power adjustment proportional coefficient , the laser power adjustment integral coefficient , the laser power adjustment differential coefficient , the adjustment coefficient , the adjustment coefficient , the adjustment coefficient , the adjustment coefficient , the adjustment coefficient can be set according to actual needs.

[0137] The preset laser powder feeding rate adjustment formula is specifically:

[0138] ;

[0139] Among them, is the adjusted laser powder feeding rate (unit: grams per minute); is the current laser powder feeding rate (grams per minute); , , are respectively the proportional, integral and differential coefficients of the PID controller, and are used to control the molten pool area feedback in real time; represents the adjustment coefficient of the current molten pool area and width on the laser powder feeding rate; represents the adjustment coefficient of the current molten pool height and flow rate on the laser powder feeding rate. Among them, the laser powder feeding rate adjustment proportional coefficient , the laser powder feeding rate adjustment integral coefficient , the laser powder feeding rate adjustment differential coefficient , the adjustment coefficient , the adjustment coefficient can be set according to actual needs.

[0140] The laser radius adjustment formula is specifically:

[0141] ;

[0142] Among them, is the adjusted laser radius (unit: millimeters); is the current laser radius; is the target molten pool height; is the height of the current molten pool; , , are respectively the laser radius adjustment proportional coefficient, laser radius adjustment integral coefficient and laser radius adjustment differential coefficient of the PID controller, and are used to adjust the laser radius according to the real-time molten pool height for feedback; Represents the adjustment coefficient of the current molten pool width and area with respect to the laser radius; Represents the adjustment coefficient of the current molten pool height and flow rate with respect to the laser radius; among them, the laser radius adjustment proportional coefficient and the laser radius adjustment integral coefficient and the laser radius adjustment differential coefficient and the adjustment coefficient and the adjustment coefficient can be set according to actual needs.

[0143] The laser scanning speed adjustment formula is specifically:

[0144] ;

[0145] Among them, is the adjusted laser scanning speed (unit: millimeter per second); is the current laser scanning speed; is the target molten pool width; is the width of the current molten pool; and and are respectively the laser scanning speed adjustment proportional coefficient, the laser scanning speed adjustment integral coefficient, and the laser scanning speed adjustment differential coefficient of the PID controller, which are used to provide feedback according to the real-time molten pool width to adjust the laser scanning speed; Represents the adjustment coefficient of the current molten pool height and area with respect to the laser radius; Represents the adjustment coefficient of the current molten pool flow rate and width with respect to the laser radius; among them, the laser scanning speed adjustment proportional coefficient and the laser scanning speed adjustment integral coefficient and the laser scanning speed adjustment differential coefficient and the adjustment coefficient and the adjustment coefficient can be set according to actual needs. In practical applications, the laser scanning speed adjustment proportional coefficient and the laser scanning speed adjustment integral coefficient and the laser scanning speed adjustment differential coefficient can all be set to negative values. When adjusting the laser scanning speed, the negative coefficients can inversely adjust the control output of the PID controller (i.e., inversely adjust the laser scanning speed), so as to make the laser scanning speed negatively correlated with the molten pool size. Thus, according to the laser scanning speed adjustment formula, by adjusting the laser scanning speed, the control of the molten pool width can be achieved.

[0146] In summary, the laser power, laser powder feeding rate, laser radius, and laser scanning speed after real-time adjustment are calculated. The laser information is adjusted using the laser power, laser powder feeding rate, laser radius, and laser scanning speed after real-time adjustment to adjust the molten pool shape of additive manufacturing.

[0147] In some embodiments, through the prior art, a preset flow rate calculation formula can be set in the molten pool image segmentation model to obtain a molten pool flow rate calculation model (when constructing a numerical calculation model for molten pool stability, the molten pool flow rate calculated by the flow rate calculation formula needs to be obtained as training data and verification data). The molten pool flow rate calculation model has the ability to calculate the molten pool flow rate and can calculate the real-time molten pool flow rate of the additive manufacturing molten pool based on the approximate geometric center of the molten pool in two frames of real-time molten pool images obtained periodically. Then, through a PID controller, using a preset laser power adjustment formula and a preset laser powder feeding rate adjustment formula, the molten pool shape of the additive manufacturing molten pool is adjusted in real time.

[0148] As can be seen from the above, the device for real-time adjustment of the additive manufacturing molten pool shape obtains the molten pool image and the corresponding molten pool annotation image of the additive manufacturing molten pool, constructs a molten pool image segmentation model based on the molten pool image and the corresponding molten pool annotation image, periodically obtains the real-time molten pool image of the additive manufacturing molten pool, inputs the real-time molten pool image into the molten pool image segmentation model, extracts the molten pool segmentation image containing only the additive manufacturing molten pool, determines the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculates the real-time molten pool flow rate of the additive manufacturing molten pool based on the approximate geometric center of the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula. Using a preset laser adjustment formula, based on the real-time molten pool flow rate, combined with a preset PID controller, with a preset target molten pool shape as the goal, the laser information for preparing the additive manufacturing molten pool is adjusted in real time to adjust the molten pool shape of the additive manufacturing molten pool in real time. Thus, through a preset laser adjustment formula, based on the real-time molten pool flow rate calculated from the approximate geometric center of the molten pool segmentation images of adjacent frames and a preset flow rate calculation formula, the molten pool shape is adjusted in real time, solving the problems of time-consuming and laborious existing molten pool adjustment methods and having certain subjectivity and limitations, being able to monitor the additive manufacturing molten pool in real time and accurately and adjust the molten pool shape of the additive manufacturing molten pool, improving the adjustment efficiency of the additive manufacturing molten pool.

[0149] Please refer to Figure 3 , Figure 3A structural schematic diagram of an electronic device provided by an embodiment of the present application. The present application provides an electronic device, including: a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanisms (not marked). The memory 302 stores a computer program executable by the processor 301. When the electronic device runs, the processor 301 executes the computer program to execute the method for real-time adjusting the shape of an additive manufacturing molten pool in any optional implementation manner of the above embodiment to implement the following functions: obtaining a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image, constructing a molten pool image segmentation model according to the molten pool image and the corresponding molten pool annotation image, periodically obtaining a real-time molten pool image of the additive manufacturing molten pool, inputting the real-time molten pool image into the molten pool image segmentation model, extracting a molten pool segmentation image containing only the additive manufacturing molten pool, determining an approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculating a real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames and in combination with a preset flow rate calculation formula. Using a preset laser adjustment formula, according to the real-time molten pool flow rate, in combination with a preset PID controller, with a preset target molten pool shape as the target, the laser information for preparing the additive manufacturing molten pool is adjusted in real time to adjust the shape of the additive manufacturing molten pool in real time.

[0150] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it executes the method for real-time adjusting the shape of an additive manufacturing molten pool in any optional implementation manner of the above embodiment to achieve the following functions: obtaining a molten pool image of the additive manufacturing molten pool and a corresponding molten pool annotation image, constructing a molten pool image segmentation model based on the molten pool image and the corresponding molten pool annotation image, periodically obtaining a real-time molten pool image of the additive manufacturing molten pool, inputting the real-time molten pool image into the molten pool image segmentation model, extracting a molten pool segmentation image that only contains the additive manufacturing molten pool, determining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculating the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric centers of the molten pool segmentation images of adjacent frames and in combination with a preset flow rate calculation formula. Using a preset laser adjustment formula, according to the real-time molten pool flow rate and in combination with a preset PID controller, with a preset target molten pool shape as the target, the laser information for preparing the additive manufacturing molten pool is adjusted in real time to adjust the shape of the additive manufacturing molten pool in real time. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0151] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical or other form.

[0152] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] Furthermore, in each of the embodiments of the present application, the functional modules may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

[0154] In this text, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0155] The above description is only for the embodiments of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for adjusting the shape of an additive manufacturing molten pool in real time, for adjusting the shape of an additive manufacturing molten pool, characterized in that: Includes steps: Obtain a melt pool image of an additive manufacturing melt pool and a corresponding melt pool annotation image; A melt pool image segmentation model is constructed according to the melt pool image and the corresponding melt pool annotated image; Periodically acquiring a real-time molten pool image of the additive manufacturing molten pool, and inputting the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image that only includes the additive manufacturing molten pool; Determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculate the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric center of the molten pool segmentation image of adjacent frames in combination with a preset flow rate calculation formula; Using a preset laser adjustment formula, according to the real-time molten pool flow rate, combined with a preset PID controller, and taking a preset target molten pool shape as a target, the laser information for preparing the additive manufacturing molten pool is adjusted in real time to adjust the molten pool shape of the additive manufacturing molten pool in real time; Determining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculating the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric center of the molten pool segmentation image of adjacent frames in combination with a preset flow rate calculation formula, including the steps of cyclic execution: Determining an approximate geometric center of the molten pool corresponding to the molten pool segmentation image according to a minimum circumscribed rectangle of the additive manufacturing molten pool corresponding to the molten pool segmentation image; After acquiring the first frame of the molten pool segmentation image, each time the molten pool segmentation image is acquired, the real-time molten pool flow rate of the additive manufacturing molten pool is calculated by combining the approximate geometric center of the molten pool of the molten pool segmentation image acquired this time with the approximate geometric center of the molten pool of the molten pool segmentation image acquired last time through the preset flow rate calculation formula; The preset flow rate calculation formula is specifically: ； in, is the real-time molten pool flow rate; is the abscissa of the approximate geometric center of the molten pool at time t; is the abscissa of the approximate geometric center of the molten pool at time t-1; is the ordinate of the approximate geometric center of the molten pool at time t; is the ordinate of the approximate geometric center of the molten pool at time t-1; The time interval between two frames of melt pool segmentation images, the unit is set to milliseconds; is the interference coefficient, Used to indicate the influence of external temperature, airflow and pressure on the flow rate of the molten pool; is the adjustment coefficient of the molten pool width and height on the molten pool flow rate; and are the width and height of the molten pool at time t respectively; and are the width and height of the molten pool at time t-1 respectively; is the environmental noise impact at time t.

2. The method for real-time adjustment of the shape of the additive manufacturing molten pool according to claim 1, characterized in that: Obtain the melt pool image of the additive manufacturing melt pool and the corresponding melt pool annotation image, including: photographing the additive manufacturing molten pool to obtain an image of the molten pool; The contour position of the additive manufacturing molten pool is marked in the molten pool image to obtain the molten pool marked image.

3. The method for real-time adjustment of the shape of the additive manufacturing molten pool according to claim 1, characterized in that: According to the melt pool image and the corresponding melt pool annotated image, a melt pool image segmentation model is constructed, including: Construct a YOLOv8 model corresponding to the molten pool image to obtain a preliminary molten pool image segmentation model; According to the melt pool image used for training and the corresponding melt pool annotated image in the melt pool image, the preliminary melt pool image segmentation model is trained to obtain a trained preliminary melt pool image segmentation model; Based on the melt pool images that are not used for training and the corresponding melt pool annotated images in the melt pool images, the trained preliminary melt pool image segmentation model is verified to obtain the melt pool image segmentation model.

4. The method for real-time adjustment of the shape of the additive manufacturing molten pool according to claim 3, characterized in that: According to the melt pool image used for training and the corresponding melt pool annotated image in the melt pool image, the preliminary melt pool image segmentation model is trained to obtain the trained preliminary melt pool image segmentation model, including: Inputting the molten pool image used for training in the molten pool image into the preliminary molten pool image segmentation model to obtain a corresponding image output; the image output is an image containing only the additive manufacturing molten pool; determining a training error according to a melt pool annotated image corresponding to a melt pool image used for training in the melt pool image and the corresponding image output; Based on the training error, the parameters of the preliminary melt pool image segmentation model are adjusted to obtain the optimal parameters, and the preliminary melt pool image segmentation model is optimized using the optimal parameters to obtain the trained preliminary melt pool image segmentation model.

5. The method for real-time adjustment of the shape of the additive manufacturing molten pool according to claim 4, characterized in that: Before inputting the molten pool image used for training in the molten pool image into the preliminary molten pool image segmentation model to obtain the corresponding image output, the method further includes: Initialize the parameters of the preliminary melt pool image segmentation model.

6. The method for real-time adjustment of the shape of the additive manufacturing molten pool according to claim 1, characterized in that: The preset laser adjustment formula includes a preset laser power adjustment formula, a preset laser powder feed rate adjustment formula, a preset laser radius adjustment formula and a preset laser scanning speed adjustment formula.

7. A device for adjusting the shape of an additive manufacturing molten pool in real time, used to adjust the shape of an additive manufacturing molten pool, characterized in that: include: An acquisition module, used to acquire a molten pool image of an additive manufacturing molten pool and a corresponding molten pool annotated image; A construction module, used to construct a melt pool image segmentation model according to the melt pool image and the corresponding melt pool annotated image; An extraction module, used for periodically acquiring a real-time molten pool image of the additive manufacturing molten pool, and inputting the real-time molten pool image into the molten pool image segmentation model to extract a molten pool segmentation image containing only the additive manufacturing molten pool; a calculation module, used to determine the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculate the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric center of the molten pool segmentation image of adjacent frames in combination with a preset flow rate calculation formula; An adjustment module, for adjusting the laser information for preparing the additive manufacturing molten pool in real time by using a preset laser adjustment formula, according to the real-time molten pool flow rate, in combination with a preset PID controller, with a preset target molten pool shape as a target, so as to adjust the molten pool shape of the additive manufacturing molten pool in real time; Determining the approximate geometric center of the molten pool corresponding to the molten pool segmentation image, and calculating the real-time molten pool flow rate of the additive manufacturing molten pool according to the approximate geometric center of the molten pool segmentation image of adjacent frames in combination with a preset flow rate calculation formula, including the steps of cyclic execution: Determining an approximate geometric center of the molten pool corresponding to the molten pool segmentation image according to a minimum circumscribed rectangle of the additive manufacturing molten pool corresponding to the molten pool segmentation image; After acquiring the first frame of the molten pool segmentation image, each time the molten pool segmentation image is acquired, the real-time molten pool flow rate of the additive manufacturing molten pool is calculated by combining the approximate geometric center of the molten pool of the molten pool segmentation image acquired this time with the approximate geometric center of the molten pool of the molten pool segmentation image acquired last time through the preset flow rate calculation formula; The preset flow rate calculation formula is specifically: ； in, is the real-time molten pool flow rate; is the abscissa of the approximate geometric center of the molten pool at time t; is the abscissa of the approximate geometric center of the molten pool at time t-1; is the ordinate of the approximate geometric center of the molten pool at time t; is the ordinate of the approximate geometric center of the molten pool at time t-1; The time interval between two frames of melt pool segmentation images, the unit is set to milliseconds; is the interference coefficient, Used to indicate the influence of external temperature, airflow and pressure on the flow rate of the molten pool; is the adjustment coefficient of the molten pool width and height on the molten pool flow rate; and are the width and height of the molten pool at time t respectively; and are the width and height of the molten pool at time t-1 respectively; is the environmental noise impact at time t.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, the method runs the steps of the method for real-time adjustment of the shape of the additive manufacturing molten pool as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for adjusting the shape of an additive manufacturing molten pool in real time as claimed in any one of claims 1 to 6 are executed.

Images