Titanium alloy uniform mixing identification method and device

Abstract

The invention discloses a titanium alloy uniform mixing identification method and device, and the method comprises the steps: obtaining an image in a mixer, marking the national standard sponge titanium particles pasted with titanium dioxide in the image in the mixer, and obtaining a marked image; training the Deeplabv3 + neural network model through the image in the mixer and the marked image to obtain a segmentation model; segmenting an image acquired in real time in the mixing process through the segmentation model to obtain a segmented image; preprocessing the segmented image, and calculating the ratio of the total pixel to the white powder pixel on the sponge titanium particles on the preprocessed image by using a threshold segmentation method; and when the proportion of the total pixels on the sponge titanium particles and the white powder pixels meets the preset uniform mixing condition, stirring is completed, and a titanium alloy mixture is obtained. According to the invention, the material mixing uniformity degree is accurately measured in real time, and the material mixing quality is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of material mixing, and in particular relates to a method and a device for identifying uniform material mixing of titanium alloys. Background Art

[0002] In the preparation process of titanium alloy, the uniformity of raw material mixing is one of the key factors that determine the performance of the product. As a high-performance material, titanium alloy is widely used in aerospace, automobile, medical equipment and other fields. Its excellent strength, corrosion resistance and high temperature performance make it an ideal choice for many high-demand occasions. However, the performance of titanium alloy is largely affected by the uniformity of its raw material mixing. Especially in applications with high precision requirements, any slight difference in uniformity may affect the quality of the final product.

[0003] At present, traditional mixing methods mainly rely on manual experience or simple physical detection methods, such as visual inspection, particle size distribution measurement, etc. These methods usually rely on the experience of operators and cannot accurately evaluate the uniformity of the mixture in real time. In addition, traditional detection methods often rely on offline detection, which leads to a lack of timeliness and continuity in monitoring during the mixing process.

[0004] At the same time, existing technologies also have difficulty effectively identifying the characteristics of fine particles during the mixing process. For example, in titanium alloy mixing, the distribution of fine particles such as sponge titanium particles and titanium dioxide is very complex, and traditional methods have difficulty capturing the slight differences between particles and local unevenness. This leads to low mixing efficiency, long mixing time and high energy consumption, and the final mixing effect cannot meet the high-standard titanium alloy production requirements.

[0005] In addition, most traditional mixing devices rely on mechanical stirring or simple stirring devices, lacking sophisticated control and intelligent management, and unable to adjust stirring speed and other parameters according to real-time data. This inefficient manual control and basic automation level make it difficult to achieve optimal particle distribution and physical and chemical properties in the mixing process, seriously affecting the production efficiency and quality of titanium alloys. Therefore, it is urgent to develop an intelligent and automated mixing uniformity identification method and device. Summary of the invention

[0006] The present invention provides a method and device for identifying uniform mixing of titanium alloys to solve the problems existing in the above-mentioned prior art.

[0007] To achieve the above object, the present invention provides a method for identifying a uniform mixture of titanium alloys, comprising the following steps:

[0008] Acquire an image inside the mixer, mark the national standard titanium sponge particles adhered with titanium dioxide in the image inside the mixer, and obtain a marked image;

[0009] The deeplabv3+ neural network model is trained by using the images in the mixer and the labeled images to obtain a segmentation model;

[0010] The image collected in real time during the mixing process is segmented by the segmentation model to obtain a segmented image;

[0011] Preprocessing the segmented image, and using a threshold segmentation method to calculate the total pixels on the titanium sponge particles and the proportion of white powder pixels on the preprocessed image;

[0012] When the ratio of total pixels on the titanium sponge particles to white powder pixels meets the preset conditions for uniform mixing, stirring is completed to obtain a titanium alloy mixture.

[0013] Preferably, real-time image acquisition during the mixing process includes:

[0014] The shooting time period is calculated according to the rotation speed of the mixer, wherein the shooting time is when the visualization window rotates to a predetermined angle; the image acquisition process is triggered according to the shooting time period, and the mixing state is recorded in real time.

[0015] Preferably, obtaining the segmentation model includes:

[0016] Divide the manually annotated label files and original image data into training sets and validation sets in proportion;

[0017] The Deeplabv3+ neural network model is trained using the training set. The training process includes: performing feature extraction training on the convolutional neural network using the training set images to automatically identify and segment the titanium sponge particles adhering to titanium dioxide; using the cross entropy loss function to optimize the model weights to improve the segmentation accuracy of the model; and tuning the hyperparameters during the training process, including the learning rate, batch size, and number of network layers;

[0018] The trained model is verified through the verification set, and the segmentation model is obtained after the verification.

[0019] Preferably, preprocessing the segmented image includes: performing grayscale processing on the segmented image to obtain a grayscale image.

[0020] Preferably, using the threshold segmentation method to calculate the total pixels and white powder pixel ratios on the titanium sponge particles on the preprocessed image includes:

[0021] According to the gray value of the image, the white titanium dioxide area on the surface of the sponge titanium particles is segmented by an adaptive threshold segmentation method to distinguish the powder area on the particle surface from the background; the total number of pixels in the sponge titanium particle area and the number of pixels in the white titanium dioxide area in the segmentation result are calculated to obtain the ratio of white powder to the total pixels on the surface of the sponge titanium particles; the ratio is used as an indicator to measure the uniformity of the mixture.

[0022] The present invention also provides a uniform mixture identification device for titanium alloy, comprising:

[0023] A stirring module is used to stir the titanium sponge particles and titanium dioxide; the stirring module comprises a main shaft, a frame, a support base, an inlet and outlet, a double-cone barrel, a stepper motor and a digital display time relay; the double-cone barrel is provided with stirring blades and a visualization window;

[0024] A mixing uniformity measurement module is used to evaluate the mixing uniformity through image analysis and deep learning to obtain an evaluation result; the mixing uniformity measurement module includes a camera, a bracket, a lighting lamp and a central control device;

[0025] A control module is used to control discharging or continuing stirring according to the evaluation result; the control module is controlled by a PLC system.

[0026] Preferably, the working process of the stirring module includes:

[0027] Add titanium sponge particles and titanium dioxide into the double-cone barrel of the mixer according to the preset ratio. The double-cone barrel is supported by the frame and the bracket base. During the mixing process, ensure that the double-cone barrel is sealed; all inlets and outlets are in a sealed state;

[0028] Start the stepper motor to drive the main shaft to rotate, drive the stirring blades to mix the materials, and control the stirring time through the digital display time relay.

[0029] Preferably, the workflow of the mixing uniformity measurement module includes:

[0030] During the mixing process, the camera supported by the bracket collects the image data in the double-cone barrel in real time through the visual window, and the image data is illuminated by the lighting lamp during the collection;

[0031] The image processing unit uses a trained deep learning model to perform semantic segmentation on the acquired images, identifying and segmenting the titanium dioxide area attached to the surface of the titanium sponge particles;

[0032] Calculate the proportion of titanium dioxide and compare it with the preset uniformity standard; if the proportion of titanium dioxide reaches the preset standard, the mixture is judged to be uniform; otherwise, continue stirring.

[0033] Preferably, the workflow of the control module includes:

[0034] When the mixture is determined to be uniform, the central control device sends a command to stop stirring through the PLC and stops the operation of the stepper motor; the discharge port is opened by controlling the solenoid valve to release and send out the mixed titanium alloy mixture.

[0035] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the method.

[0036] Compared with the prior art, the present invention has the following advantages and technical effects:

[0037] The present invention discloses a uniform mixing identification method and device for titanium alloy, including: obtaining an image in a mixer, marking the national standard titanium sponge particles with titanium dioxide in the image in the mixer, and obtaining a marked image; training a deeplabv3+ neural network model through the image in the mixer and the marked image to obtain a segmentation model; segmenting the image collected in real time during the mixing process through the segmentation model to obtain a segmented image; preprocessing the segmented image, using a threshold segmentation method to calculate the total pixel and white powder pixel ratio on the titanium sponge particles on the preprocessed image; when the total pixel and white powder pixel ratio on the titanium sponge particles meet the preset conditions for uniform mixing, stirring is completed to obtain a titanium alloy mixture. The present invention can not only monitor and accurately evaluate the uniformity of the mixture in real time, but also automatically adjust the stirring parameters such as stirring speed, time, etc. according to data feedback, thereby improving the mixing efficiency and ensuring the uniformity of the titanium alloy and the quality of the final product. The development of this intelligent system will greatly improve the production quality and production efficiency of titanium alloys, and meet the requirements of modern manufacturing industry for high-precision and high-efficiency production. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0039] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0040] Figure 2 is a structural diagram of a device according to an embodiment of the present invention;

[0041] Among them, 1. Camera; 2. Spindle; 3. Rack; 4. PLC control system; 5. Light; 6. Visualization window; 7. Bracket base; 8. Inlet and outlet; 9. Double-cone barrel; 10. Bracket; 11. Stepper motor; 12. Digital display time relay. DETAILED DESCRIPTION

[0042] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0043] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0044] The present invention is introduced as follows:

[0045] The present invention provides a method and device for identifying uniform mixing of titanium alloys, aiming to achieve real-time monitoring and precise control of the titanium alloy mixing process through automation and intelligent technology, thereby improving the uniformity of mixing and ensuring product quality and production efficiency. In order to solve the problems of inaccurate and inefficient mixing uniformity assessment in the prior art, an embodiment of the present invention provides a method for identifying uniform mixing of titanium alloys, specifically comprising: adding various raw materials of titanium alloy (such as sponge titanium particles and titanium dioxide) to a double-cone barrel (or other type of mixing equipment) in a mixer according to a preset proportion, placing the raw materials in a sealed state and starting stirring to obtain the current mixture; at the same time, ensuring the sealing and stability of the internal environment of the double-cone barrel for subsequent uniformity detection and control; during the stirring process of the current mixture, real-time image data is collected by a camera installed in the double-cone barrel, and transmitted to an image processing unit, and the collected image data is segmented and processed using a deep learning algorithm to identify the titanium dioxide area attached to the surface of the sponge titanium particles. domain; grayscale the segmented image using an image processing unit, and calculate the pixel ratio of titanium dioxide on the surface of the titanium sponge particles by the threshold segmentation method, and use the ratio as the uniformity judgment index of the current mixture; according to the real-time feedback of the image processing unit, the central control device analyzes the uniformity data of the current mixture, and if the ratio of titanium dioxide reaches the preset uniformity standard, the current mixture is judged to be uniform, and the stirring and discharging stage is entered; if the uniformity of the mixture does not meet the requirements, the stirring is continued until the standard is reached; when the uniformity meets the requirements, the central control device issues a stop command through the PLC system, stops the stirring of the mixer, and opens the sealed electromagnetic switch valve of the discharge port to complete the discharging process of the titanium alloy mixture, ensuring that the mixed material is uniform and meets the production requirements. In the embodiment of the present invention, by combining image processing and deep learning algorithms, real-time monitoring of uniformity in the mixing process is realized, and whether the mixing reaches the predetermined standard is automatically judged, which solves the problem that traditional mechanical stirring and manual experience judgment are difficult to control uniformity in real time. The present invention can accurately adjust the mixing parameters (such as stirring time, speed, etc.), effectively improve the mixing efficiency, reduce waste, reduce energy consumption, and ensure the quality of the final titanium alloy product. As a preferred solution, the image processing unit uses a deep learning algorithm (such as Deeplabv3+) to perform pixel-level segmentation on the collected image, accurately identify the distribution of titanium dioxide and sponge titanium particles, and further improve the accuracy and efficiency of uniformity analysis. The specific steps are: design multiple cameras and image processing modules to monitor the distribution of sponge titanium particles and titanium dioxide during the mixing process in real time, evaluate the uniformity of the mixture by calculating the titanium dioxide coverage, and accurately control the stirring time through a digital display time relay to ensure that the mixing time of each stage meets the set requirements. The device of the present invention avoids the problems of insufficient manual experience or inaccurate traditional detection methods through multiple intelligent means of automatically collecting image data, analyzing uniformity data in real time, and controlling the stirring process.Through real-time control and optimization of parameters such as mixing time and stirring speed, the mixing process becomes more efficient, stable and more consistent.

[0046] Embodiment 1

[0047] like Figure 1 As shown, in this embodiment, a method for identifying a uniform mixture of titanium alloys is provided, comprising the following steps:

[0048] Acquire an image inside the mixer, mark the national standard titanium sponge particles adhered with titanium dioxide in the image inside the mixer, and obtain a marked image;

[0049] The deeplabv3+ neural network model is trained using the images in the mixer and the labeled images to obtain the segmentation model;

[0050] The images collected in real time during the mixing process are segmented by a segmentation model to obtain a segmented image;

[0051] The segmented image is preprocessed, and the total pixels and the percentage of white powder pixels on the titanium sponge particles in the preprocessed image are calculated using the threshold segmentation method;

[0052] When the ratio of total pixels on the titanium sponge particles to white powder pixels meets the preset conditions for uniform mixing, stirring is completed to obtain a titanium alloy mixture.

[0053] The specific steps include:

[0054] Step 1: Prepare ingredients and initialize equipment

[0055] Add titanium sponge particles and titanium dioxide in a preset ratio (e.g. 70% titanium sponge particles, 30% titanium dioxide) into the double-cone barrel 9 of the mixer. The double-cone barrel is sealed to prevent the external environment from interfering with the mixing process and also to prevent the loss of volatile substances in the titanium dioxide. Start the mixer, ensure that the equipment is in normal working condition, and check whether the camera 1, PLC control system 4, and stepper motor 11 are in normal operating condition.

[0056] Step 2: Start the image acquisition system

[0057] After the mixing process begins, the camera 1 is started to collect image data in the double-cone barrel in real time through the visualization window 6 of the mixer. The camera transmits the image data to the image processing unit, which performs preliminary processing on the image to ensure that the image quality meets the requirements of subsequent analysis. The transmission of the image data is completed by wireless or wired transmission. After the image processing unit receives the data, it immediately starts the analysis.

[0058] Step 3: Use deep learning models for image semantic segmentation

[0059] The image processing unit uses a trained deep learning model such as Deeplabv3+ to perform semantic segmentation on the image. The model identifies different areas in the image and separates the titanium dioxide area attached to the surface of the titanium sponge particles from other areas. Specifically, the deep learning model separates the titanium sponge particles from the titanium dioxide and analyzes the distribution of the titanium dioxide. The result of the image segmentation will mark the titanium dioxide attachment area in pixels, providing basic data for subsequent uniformity calculations.

[0060] Step 4: Calculate the pixel ratio of titanium dioxide on the surface of titanium sponge particles

[0061] The segmented image is grayed and converted into a black and white image. Then, the titanium dioxide area attached to the surface of the titanium sponge particles in the image is further distinguished from the background area by using the threshold segmentation method. The threshold segmentation method separates the pixel values ​​of the titanium dioxide area from other areas by setting a suitable threshold, and then calculates the pixel ratio of titanium dioxide on the surface of the titanium sponge particles. The calculation results are fed back to the central control device through the image processing unit.

[0062] Step 5: Uniformity determination and discharge control

[0063] According to the proportion of titanium dioxide calculated in step 4, the central control device determines whether the proportion meets the preset uniformity standard. For example, if the proportion of titanium dioxide reaches the set 95% or above, it means that the uniformity of the current mixture meets the requirements and the mixture is judged to be uniform. At this time, the central control device sends a stop command through the PLC control system to stop the stirring of the mixer and open the sealed electromagnetic switch valve at the discharge port to complete the discharge operation of the titanium alloy mixture.

[0064] If the calculation results show that the proportion of titanium dioxide does not meet the preset standard, the mixture is judged to be uneven. The central control device will send an instruction to continue stirring to the PLC system, and the stirring process will continue until the uniformity reaches the preset standard.

[0065] This embodiment combines deep learning image analysis with real-time data feedback to accurately evaluate the uniformity of titanium alloy mixtures and automatically control the mixing time and stirring process based on the judgment results. Compared with traditional manual detection methods, this method has higher accuracy and efficiency.

[0066] like Figure 2 As shown, this embodiment also provides a uniform mixture identification device of titanium alloy, including:

[0067] A stirring module is used to stir titanium sponge particles and titanium dioxide; the stepper motor stirring module includes a main shaft, a frame, a support base, an inlet and outlet, a double-cone barrel, a stepper motor and a digital display time relay; the stepper motor double-cone barrel is provided with stirring blades and a visualization window;

[0068] The mixing uniformity measurement module is used to evaluate the mixing uniformity through image analysis and deep learning to obtain the evaluation results; the stepper motor mixing uniformity measurement module includes a camera, a bracket, a lighting lamp and a central control device;

[0069] The control module is used to control discharging or continuing stirring according to the evaluation result of the stepper motor; the stepper motor control module is controlled by the PLC system.

[0070] The equipment is connected as follows: the central control device is connected to the feed port, each camera, PLC, digital display time relay, spindle, stepper motor, and controls the operation of each device;

[0071] The camera 1 is installed next to the visualization window 6 and is used to collect image data in the double-cone barrel in real time; the main shaft 2 is driven by a stepper motor 11 and is used to stir the titanium alloy raw material in the double-cone barrel; the stirring blade is connected to the main shaft and can rotate with the rotation of the main shaft to promote the uniform mixing of the mixed materials; the lighting lamp 5 is used to provide sufficient lighting to ensure that the camera can clearly collect image data in the double-cone barrel; the visualization window 6 is used to view the mixing process in real time, and is equipped with a camera 1 to monitor the mixing state;

[0072] The PLC control system 4 connects various devices to adjust the speed, heating status and other parameters during the mixing process; the digital display time relay 12 is used to accurately control the mixing time to ensure that the mixing time meets the preset requirements; the bracket base 7 is used to support the entire mixing device to keep the equipment stable; the double-cone barrel 9 is used to store and mix the titanium alloy raw materials and provide sufficient mixing space; the bracket 10 supports the structural part of the entire device to ensure that the connection of each component is stable. The central control device controls the sealing electromagnetic switch valve that closes the feed port 8, puts the titanium alloy raw materials in the double-cone barrel in a sealed state, and adjusts the speed of the stepper motor through the PLC control system to drive the main shaft 2 and the mixing blades to mix;

[0073] During the mixing process, the camera 1 is used to collect the image data of the mixing material in real time, and the uniformity of the mixing material is judged by combining the image processing technology; the lighting lamp 5 is controlled by the PLC control system 4 to ensure the clarity of the image acquisition;

[0074] The digital display time relay 12 is used to control the duration of the mixing process according to the preset time; based on the real-time monitoring of the mixing state and uniformity, the central control device determines whether the current mixing meets the uniformity requirements;

[0075] When the uniformity meets the requirements, the central control device stops stirring through the PLC, opens the sealed electromagnetic switch valve of the inlet and outlet 8, and completes the titanium alloy mixing process.

[0076] Furthermore, the PLC control system 4 is connected to various devices to adjust the speed parameters during the mixing process; the digital display time relay 12 is used to accurately control the mixing time to ensure that the mixing time meets the preset requirements, specifically:

[0077] The PLC control system 4 controls the rotation speed of the main shaft 2 by adjusting the speed of the stepper motor, thereby adjusting the speed of the stirring blades and optimizing the mixing efficiency; the digital display time relay 12 controls the start and end of the stirring process by setting a precise time period to ensure that the mixing time meets the predetermined conditions; the digital display time relay 12 accurately monitors the time progress of the mixing through the timing function, and stops the stirring operation according to the set time limit to avoid excessive stirring causing changes in material properties; the PLC control system 4 and the digital display time relay 12 work together to automatically adjust the time and speed parameters of the mixing to ensure that the entire mixing process meets the set uniformity requirements; when the stirring time reaches the preset conditions and the mixing uniformity meets the standard, the PLC control system 4 sends a stop signal to stop stirring and open the sealed electromagnetic switch valve at the discharge port to complete the output of the titanium alloy mixture.

[0078] Furthermore, the digital display time relay 12 is used to control the duration of the mixing process according to a preset time; based on the real-time monitored mixing state and uniformity, the central control device determines whether the current mixing meets the uniformity requirements, specifically:

[0079] The digital display time relay 12 accurately adjusts the mixing time by controlling the central control device according to the preset mixing time cycle, and ensures the accurate execution of the mixing operation through the timing function; the central control device monitors the distribution of titanium alloy particles and the adhesion of titanium dioxide in the mixing process in real time through the camera 1 and the image processing unit; based on the real-time feedback of the mixing image data, the central control device determines the uniformity of the mixture through the deep learning processing unit, and compares the obtained uniformity data with the preset standard; if the real-time monitored mixing uniformity data meets the preset standard, the central control device determines that the current mixing has reached the ideal state; if the mixing uniformity does not meet the preset requirements, the central control device will extend the mixing time, and continue to adjust the speed of the stepper motor, increase the stirring intensity, until the mixing uniformity meets the standard; once the mixing uniformity meets the requirements, the digital display time relay 12 terminates the stirring operation according to the preset time, opens the sealed electromagnetic switch valve of the discharge port, completes the mixing process, and obtains a uniformly mixed titanium alloy mixture.

[0080] Furthermore, when the uniformity meets the requirements, the central control device stops stirring through the PLC, opens the sealed electromagnetic switch valve of the inlet and outlet 8, and completes the titanium alloy mixing process, specifically:

[0081] After receiving the signal that the uniformity meets the preset standard, the central control device sends a stop command to the stepper motor through the PLC system to stop the stirring operation; the central control device confirms that the current mixing state meets the preset uniformity requirement based on the monitored uniformity data, and simultaneously monitors the vibration signal of the mixing process to ensure that the mixing has achieved the ideal uniformity; the PLC control system 4 opens the sealed electromagnetic switch valve of the inlet and outlet 8 through the solenoid valve control signal to release the titanium alloy mixture; before opening the discharge port, the PLC system will perform delay control according to the preset discharge timing and equipment status to ensure that there is no external interference in the mixing process and the discharge is completed smoothly; after stopping stirring and completing the discharge, the central control device will reset the system status to prepare for the next mixing cycle.

[0082] Furthermore, the stirring module comprises:

[0083] Raw materials preparation:

[0084] Titanium sponge particles and titanium dioxide are added to the double-cone barrel 9 of the mixer according to a preset ratio. The double-cone barrel 9 is supported by the frame 3 and the support base 7. During the mixing process, the double-cone barrel 9 is ensured to be sealed. All inlets and outlets 8 are in a sealed state to prevent external substances from interfering with the mixing process.

[0085] Start stirring:

[0086] The stepper motor 11 is started to drive the main shaft 2 to rotate, driving the stirring blades to mix the materials. The stirring time is accurately controlled by the digital display time relay 12 to ensure that the sponge titanium particles and titanium dioxide are fully contacted and evenly distributed during the mixing process.

[0087] Furthermore, the mixing uniformity measurement module includes:

[0088] Image acquisition and processing:

[0089] During the mixing process, the camera 1 is supported by the bracket 10 and collects image data in the double-cone barrel in real time through the visualization window 6, and is illuminated by the lighting lamp 5. The image data is transmitted to the image processing unit for preliminary analysis.

[0090] Deep Learning Analysis:

[0091] The image processing unit uses the trained deep learning model Deeplabv3+ to perform semantic segmentation on the collected images, identifying and segmenting the titanium dioxide area attached to the surface of the titanium sponge particles. The segmented images are transmitted to the deep learning processing unit for further analysis of the area and proportion covered by the titanium dioxide.

[0092] Uniformity assessment:

[0093] The deep learning processing unit calculates the proportion of titanium dioxide and compares it with the preset uniformity standard. If the proportion of titanium dioxide reaches more than 95% of the preset standard, the mixture is considered uniform. Otherwise, it returns to continue stirring.

[0094] Furthermore, the control module includes:

[0095] Stop stirring and discharging: If the uniformity of the current mixture meets the preset conditions, the central control device sends a command to stop stirring through the PLC control system 4, and stops the operation of the stepper motor 11. Then, the discharge port is opened by controlling the solenoid valve to release and send out the mixed titanium alloy mixture.

[0096] This embodiment further provides an electronic device, including: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the method.

[0097] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for identifying uniform mixing of titanium alloys, characterized in that: The following steps are involved: Acquire an image inside the mixer, mark the national standard titanium sponge particles adhered with titanium dioxide in the image inside the mixer, and obtain a marked image; The deeplabv3+ neural network model is trained by using the images in the mixer and the labeled images to obtain a segmentation model; The image collected in real time during the mixing process is segmented by the segmentation model to obtain a segmented image; Preprocessing the segmented image, and using a threshold segmentation method to calculate the total pixels on the titanium sponge particles and the proportion of white powder pixels on the preprocessed image; When the ratio of total pixels on the titanium sponge particles to white powder pixels meets the preset conditions for uniform mixing, stirring is completed to obtain a titanium alloy mixture.

2. The method according to claim 1, characterized in that Real-time image acquisition during the mixing process includes: The shooting time period is calculated according to the rotation speed of the mixer, wherein the shooting time is when the visualization window rotates to a predetermined angle; the image acquisition process is triggered according to the shooting time period, and the mixing state is recorded in real time.

3. The method according to claim 1, characterized in that Obtaining the segmentation model includes: Divide the manually annotated label files and original image data into training sets and validation sets in proportion; The Deeplabv3+ neural network model is trained using the training set. The training process includes: performing feature extraction training on the convolutional neural network using the training set images to automatically identify and segment the titanium sponge particles adhering to titanium dioxide; using the cross entropy loss function to optimize the model weights to improve the segmentation accuracy of the model; and tuning the hyperparameters during the training process, including the learning rate, batch size, and number of network layers; The trained model is verified through the verification set, and the segmentation model is obtained after the verification.

4. The method according to claim 1, characterized in that The preprocessing of the segmented image includes: performing grayscale processing on the segmented image to obtain a grayscale image.

5. The method according to claim 1, characterized in that The total pixels and white powder pixel ratios on the titanium sponge particles in the preprocessed image are calculated using the threshold segmentation method: According to the gray value of the image, the white titanium dioxide area on the surface of the sponge titanium particles is segmented by an adaptive threshold segmentation method to distinguish the powder area on the particle surface from the background; the total number of pixels in the sponge titanium particle area and the number of pixels in the white titanium dioxide area in the segmentation result are calculated to obtain the ratio of white powder to the total pixels on the surface of the sponge titanium particles; the ratio is used as an indicator to measure the uniformity of the mixture.

6. A uniform mixing identification device for titanium alloy, characterized in that: include: A stirring module is used to stir the titanium sponge particles and titanium dioxide; the stirring module comprises a main shaft, a frame, a support base, an inlet and outlet, a double-cone barrel, a stepper motor and a digital display time relay; the double-cone barrel is provided with stirring blades and a visualization window; A mixing uniformity measurement module is used to evaluate the mixing uniformity through image analysis and deep learning to obtain an evaluation result; the mixing uniformity measurement module includes a camera, a bracket, a lighting lamp and a central control device; A control module is used to control discharging or continuing stirring according to the evaluation result; the control module is controlled by a PLC system.

7. The device according to claim 6, characterized in that The working process of the stirring module includes: Add titanium sponge particles and titanium dioxide into the double-cone barrel of the mixer according to the preset ratio. The double-cone barrel is supported by the frame and the bracket base. During the mixing process, ensure that the double-cone barrel is sealed; all inlets and outlets are in a sealed state; Start the stepper motor to drive the main shaft to rotate, drive the stirring blades to mix the materials, and control the stirring time through the digital display time relay.

8. The device according to claim 6, characterized in that The workflow of the mixing uniformity measurement module includes: During the mixing process, the camera supported by the bracket collects the image data in the double-cone barrel in real time through the visual window, and the image data is illuminated by the lighting lamp during the collection; The image processing unit uses a trained deep learning model to perform semantic segmentation on the acquired images, identifying and segmenting the titanium dioxide area attached to the surface of the titanium sponge particles; Calculate the proportion of titanium dioxide and compare it with the preset uniformity standard; if the proportion of titanium dioxide reaches the preset standard, the mixture is judged to be uniform; otherwise, continue stirring.

9. The device according to claim 6, characterized in that The workflow of the control module includes: When the mixture is determined to be uniform, the central control device sends a command to stop stirring through the PLC control system to stop the operation of the stepper motor; the discharge port is opened by controlling the solenoid valve to release and send out the mixed titanium alloy mixture.

10. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the method according to any one of claims 1 to 5.

Images