Titanium alloy forging process optimization method and system based on image processing

Abstract

The invention relates to the technical field of titanium alloy forging, in particular to a titanium alloy forging process optimization method and system based on image processing, and the method comprises the steps: obtaining a surface image of a target titanium alloy workpiece, and generating a discrimination result; if the defect is a crack type defect, generating an optimal repair parameter and transferring the optimal repair parameter to repair equipment; if the defect is a deformation defect, generating an optimal compensation parameter and transferring the optimal compensation parameter to forging equipment; and if no defect exists, transferring to a cooling work station. Real-time monitoring and rapid feedback of dynamic changes in the forging process are achieved through the image acquisition and analysis technology, the timeliness and accuracy of process adjustment are improved, the quality stability and production efficiency of titanium alloy workpieces are improved, and the efficient and intelligent requirements of modern industry are met.

Description

Technical Field

[0001] The present invention belongs to the field of titanium alloy forging technology, specifically a titanium alloy forging process optimization method and system based on image processing Background Art

[0002] With the widespread application of titanium alloys in high-end manufacturing fields such as aerospace, energy equipment, and medical devices, the optimization of titanium alloy forging technology has become a key research direction for improving product quality, reducing costs, and shortening development cycles. Currently, the optimization of titanium alloy forging processes mainly relies on methods such as finite element numerical simulation, experimental data fitting, and artificial intelligence-assisted modeling. For example, patent publication number CN114818437B proposes a method for optimizing the isothermal forging process of titanium alloy integral disks based on constitutive models and three-dimensional finite element numerical simulation. The method optimizes process parameters by analyzing the stress, strain, and temperature field distributions of pre-forged and final forgings under different conditions. However, this method mainly relies on offline numerical simulation and experimental data analysis, lacking a real-time monitoring and feedback mechanism for dynamic changes during the actual forging process. This may lead to untimely adjustment of process parameters, thereby affecting the microstructure, properties, and quality stability of the final product. In addition, this method is highly dependent on initial conditions and boundary definitions. If there are deviations in the input data, the accuracy of the optimization results may be significantly reduced.

[0003] On the other hand, the patent with publication number CN117352109B provides a virtual modeling method for titanium alloy forging. By constructing a model of titanium alloy materials and forging tools and performing simulated forging, the process parameters are optimized and the model is adjusted according to the simulation results to achieve the expected performance indicators. However, this technical solution focuses on virtual modeling and simulation, and cannot directly process the image data or real-time monitoring information generated during the actual forging process, making it difficult to achieve dynamic optimization of the forging process. At the same time, this method has high requirements for the accuracy of the initial model and computing resources. In large-scale industrial production scenarios, there may be problems such as low efficiency and poor economy, which limits its practical application value.

[0004] The above-mentioned existing technologies show that the current titanium alloy forging process optimization methods still have obvious deficiencies in terms of real-time monitoring, dynamic feedback, and image data processing. In particular, during the actual forging process, due to the influence of complex working conditions such as high temperature and high pressure, the morphology, temperature distribution, and surface state of the forgings may change significantly. However, existing technical means are difficult to capture and quickly feedback these dynamic changes in real time, resulting in delayed adjustment of process parameters, affecting the consistency of product quality and production efficiency. Therefore, there is an urgent need for a titanium alloy forging process optimization method and system that can combine image acquisition and analysis technology to improve the timeliness and accuracy of process parameter adjustment by real-time monitoring and rapid feedback of dynamic changes in the forging process, thereby meeting the needs of modern industry for efficient and intelligent forging technology. Summary of the Invention

[0005] The present invention relates to the technical field of titanium alloy forging, and in particular to a titanium alloy forging process optimization method and system based on image processing.

[0006] As noted in the background, existing titanium alloy forging process optimization methods still have shortcomings in real-time monitoring, dynamic feedback, and image data processing. In particular, the lack of real-time monitoring and rapid feedback mechanisms for dynamic changes during the forging process leads to delayed adjustments to process parameters, impacting the microstructure, properties, and quality stability of the final product. Therefore, how to use image acquisition and analysis technology to monitor dynamic changes during the forging process in real time and provide rapid feedback on optimization recommendations is an urgent technical challenge.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] The first aspect of the present invention discloses a titanium alloy forging process optimization method based on image processing, comprising the following steps:

[0009] Acquire a surface image of a target titanium alloy workpiece during a forging process, and perform state recognition and judgment processing on the target titanium alloy workpiece based on the surface image to generate a judgment result;

[0010] If the discrimination result is the first discrimination result, it indicates that there is a crack-type defect in the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment;

[0011] If the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment.

[0012] If the judgment result is the third judgment result, it means that there are neither crack-type defects nor deformation-type defects on the surface of the target titanium alloy workpiece. At this time, the industrial robot is directly controlled to transfer the target titanium alloy workpiece from the inspection workstation to the cooling workstation, so that the cooling equipment can cool the target titanium alloy workpiece according to the preset cooling parameters.

[0013] Preferably, a surface image of the target titanium alloy workpiece is obtained, and a state recognition and judgment process is performed on the target titanium alloy workpiece based on the surface image to generate a judgment result, specifically:

[0014] All possible defects that may occur in the forging process of titanium alloy workpieces are obtained through the big data network, and the defect feature images corresponding to each defect are obtained; the defect feature images corresponding to each defect are classified and processed to obtain defect feature images of crack defects and defect feature images of deformation defects;

[0015] Constructing a first knowledge graph, and importing defect feature images of crack defects into the first knowledge graph; constructing a second knowledge graph, and importing defect feature images of deformation defects into the second knowledge graph;

[0016] The surface image of the target titanium alloy workpiece is acquired through the industrial camera mounted on the inspection workstation;

[0017] Importing the surface image into the first knowledge graph, calculating a first structural matching index between the surface image and each defect feature image in the first knowledge graph, and obtaining a plurality of first structural matching indexes;

[0018] Comparing each first structure matching index with a preset threshold value; if at least one first structure matching index is greater than the preset threshold value, generating a first discrimination result;

[0019] If each first structural matching index is not greater than a preset threshold, importing the surface image into the second knowledge graph, calculating the second structural matching index between the surface image and each defect feature image in the second knowledge graph, and obtaining a plurality of second structural matching indexes;

[0020] Each second structure matching index is compared with a preset threshold value. If at least one second structure matching index is greater than the preset threshold value, a second discrimination result is generated; if none of the second structure matching indexes is greater than the preset threshold value, a third discrimination result is generated.

[0021] Preferably, if the discrimination result is the first discrimination result, it indicates that there is a crack-type defect in the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment, specifically:

[0022] If the discrimination result is the first discrimination result, it indicates that there is a crack defect in the surface of the target titanium alloy workpiece, and then the crack position area of ​​the crack defect in the target titanium alloy workpiece is obtained according to the surface image recognition of the target titanium alloy workpiece;

[0023] Segmenting a crack defect image of a crack location area from the surface image, performing three-dimensional reconstruction processing on the crack defect image to obtain a three-dimensional crack model image of the crack-like defect;

[0024] Determining optimal repair parameters for repairing the crack-like defects in the target titanium alloy workpiece based on a three-dimensional crack model diagram of the crack-like defects in the target titanium alloy workpiece; wherein the repair parameters include repair temperature, repair pressure, repair time, and repair direction;

[0025] Obtain the coded identification information of the target titanium alloy workpiece, transmit the coded identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect and the optimal repair parameters to the control terminal of the repair equipment based on the Internet of Things, and control the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment.

[0026] Preferably, the optimal repair parameters for repairing the crack-like defects in the target titanium alloy workpiece are determined based on the three-dimensional crack model diagram of the crack-like defects in the target titanium alloy workpiece, specifically:

[0027] Develop in advance the optimal repair parameters of the repair equipment for various crack defect conditions, and obtain the optimal repair parameters corresponding to various crack defect conditions; and use 3D modeling software to draw 3D models of crack conditions corresponding to various crack defect conditions;

[0028] Bundling optimal repair parameters corresponding to various crack defect working conditions and corresponding three-dimensional model diagrams of crack working conditions to obtain a plurality of repair parameter paired data packages; constructing a database, and importing each repair parameter paired data package into the database;

[0029] Obtaining a three-dimensional crack model diagram of a crack-like defect in a target titanium alloy workpiece, and calculating similarities between the three-dimensional crack model diagram and three-dimensional crack condition model diagrams in each repair parameter pairing data package using a geometric matching method to obtain multiple similarities;

[0030] The maximum similarity is extracted from the multiple similarities, and the three-dimensional crack model image corresponding to the maximum similarity is obtained, and the repair parameter pairing data package of the three-dimensional crack model image corresponding to the maximum similarity is extracted. According to the extracted repair parameter pairing data package, the optimal repair parameters for repairing crack defects in the target titanium alloy workpiece are obtained.

[0031] Preferably, if the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment, specifically:

[0032] If the discrimination result is the second discrimination result, it means that there is no crack defect but there is a deformation defect on the surface of the target titanium alloy workpiece, and the deformation position area of ​​the deformation defect in the target titanium alloy workpiece is obtained according to the surface image recognition of the target titanium alloy workpiece;

[0033] Segmenting a deformation defect image of the deformation position area in the surface image, constructing a deformation three-dimensional model diagram of the deformation defect in the target titanium alloy workpiece based on the deformation defect image; and calculating the deformation amount of the deformation three-dimensional model diagram;

[0034] Obtaining standard shape parameters of the target titanium alloy workpiece under an ideal state, performing ratio processing on the deformation variables of the deformed three-dimensional model diagram and the standard shape parameters of the target titanium alloy workpiece under the ideal state to obtain a deformation ratio; and comparing the deformation ratio with a preset ratio;

[0035] If the deformation ratio is greater than a preset ratio, the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the scrapping workstation to scrap the target titanium alloy workpiece.

[0036] Preferably, the method further comprises the following steps:

[0037] If the deformation ratio is not greater than the preset ratio, the deformation position area of ​​the deformation defect in the target titanium alloy workpiece is marked as an area requiring compensation and repair;

[0038] Obtain the deformation amount of the deformation defect in the area that needs to be compensated and repaired;

[0039] Determining the compensation amount required to repair the area to be compensated and repaired according to the deformation amount;

[0040] Obtaining preset forging parameters of a forging device, and extracting a preset compensation amount for forging the area to be compensated and repaired in the target titanium alloy workpiece according to the preset forging parameters; determining an optimal compensation amount for the area to be compensated and repaired according to the compensation amount and the preset compensation amount;

[0041] Obtaining a forging time node at which the forging equipment performs forging on the area requiring compensation repair, replacing a preset compensation amount at the forging time node with an optimal compensation amount, and after the replacement is completed, updating the preset forging parameters to obtain the optimal compensation parameters for forging the target titanium alloy workpiece;

[0042] Obtain the coded identification information of the target titanium alloy workpiece, transmit the coded identification information and the optimal compensation parameters of the target titanium alloy workpiece to the control terminal of the forging equipment based on the Internet of Things, and control the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment.

[0043] A second aspect of the present invention discloses a titanium alloy forging process optimization system based on image processing. The titanium alloy forging process optimization system includes a memory and a processor. The memory stores a titanium alloy forging process optimization method program. When the titanium alloy forging process optimization method program is executed by the processor, the following steps are implemented:

[0044] Acquire a surface image of a target titanium alloy workpiece, and perform state recognition and judgment processing on the target titanium alloy workpiece based on the surface image to generate a judgment result;

[0045] If the discrimination result is the first discrimination result, it indicates that there is a crack-type defect in the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment;

[0046] If the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment.

[0047] If the judgment result is the third judgment result, it means that there are neither crack-type defects nor deformation-type defects on the surface of the target titanium alloy workpiece. At this time, the industrial robot is directly controlled to transfer the target titanium alloy workpiece from the inspection workstation to the cooling workstation, so that the cooling equipment can cool the target titanium alloy workpiece according to the preset cooling parameters.

[0048] The present invention solves the technical defects existing in the background technology and has the following beneficial effects: through image acquisition and analysis technology, real-time monitoring of dynamic changes in the titanium alloy forging process is realized, and optimization suggestions can be quickly fed back, thereby improving the timeliness and accuracy of process parameter adjustment; through the precise identification and processing of crack defects and deformation defects, the quality stability and production efficiency of titanium alloy workpieces are effectively improved, meeting the needs of modern industry for efficient and intelligent forging technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 The figure is a schematic diagram of the overall process of a titanium alloy forging process optimization method based on image processing;

[0050] Figure 2 A detailed flow chart of state recognition and parameter generation in a titanium alloy forging process optimization method based on image processing;

[0051] Figure 3 This is a system structure diagram of a titanium alloy forging process optimization system based on image processing. DETAILED DESCRIPTION

[0052] The present invention provides a titanium alloy forging process optimization method and system based on image processing, combined with the attached Figure 1 To the attached Figure 3 , describe in detail the specific implementation of the method and system. Figure 1 As shown in the figure, the overall process includes acquiring an image of the target titanium alloy workpiece surface, identifying and generating a judgment result, and then adjusting process parameters or performing equipment operations based on the judgment result. In practical application, this method is implemented through a complete hardware and software collaborative system, covering the entire process from image acquisition to defect identification, parameter optimization, and equipment control.

[0053] First, an industrial camera, serving as the core image acquisition device, is installed on the inspection workstation to capture real-time surface images of the target titanium alloy workpiece during the forging process. This image data is transmitted via a high-speed network to a processor, which stores a program for optimizing the titanium alloy forging process. This program, when executed, analyzes the image data and generates a judgment result. To ensure image quality, the industrial camera uses a high-resolution sensor and is equipped with an appropriate light source to minimize ambient light interference. Furthermore, the camera's field of view is precisely calibrated to cover the entire workpiece surface and avoid distortion. After image acquisition, the processor imports the image into a pre-built first and second knowledge graphs, corresponding to feature image libraries for crack-related defects and deformation-related defects, respectively. These two knowledge graphs are constructed by acquiring and classifying feature images of various defects that may occur during the forging process through a big data network. The first knowledge graph stores feature images of crack-related defects, while the second knowledge graph stores feature images of deformation-related defects. Through the geometric matching algorithm, the processor calculates the structural matching index between the target titanium alloy workpiece surface image and each defect feature image in the knowledge graph. The specific formula is as follows: For crack defects, the calculation formula of the first structural matching index I1 is:

[0054]

[0055] Among them S i represents the similarity score between the target image and the i-th crack feature image. For deformation defects, the calculation formula for the second structural matching index I2 is similar, but is calculated for the deformation feature image library. If at least one first structural matching index is greater than a preset threshold, a first discrimination result is generated. If none of the first structural matching indices is greater than the preset threshold, the second structural matching index is further calculated. If at least one second structural matching index is greater than the preset threshold, a second discrimination result is generated; otherwise, a third discrimination result is generated. This judgment logic ensures accurate identification of workpiece surface defects.

[0056] When the discrimination result is the first discrimination result, it indicates that there is a crack defect on the surface of the target titanium alloy workpiece. At this time, the processor will automatically identify the crack location area and segment the crack defect image of the area. Subsequently, the crack defect image is processed by 3D reconstruction technology to generate a 3D crack model of the crack defect. In the 3D reconstruction process, a multi-view image fusion algorithm is used, and its core formula is Where M(x,y,z) represents the value of the final three-dimensional model at the spatial point (x,y,z), w k is the weight coefficient of the k-th perspective, P k(x, y, z) is the pixel value at the kth viewing angle. The generated three-dimensional model diagram of the crack is then used to determine the optimal repair parameters. Specifically, the system formulates the optimal repair parameters corresponding to various crack defect conditions in advance, and draws the corresponding three-dimensional model diagrams of the crack conditions through three-dimensional modeling software. These parameters and model diagrams are bundled into several repair parameter pairing data packets and stored in the database. The processor calculates the similarity between the target crack three-dimensional model diagram and the crack condition three-dimensional model diagrams in each repair parameter pairing data packet in the database through a geometric matching method, extracts the maximum similarity and its corresponding repair parameter pairing data packet, and thus obtains the optimal repair parameters. The repair parameters include repair temperature, repair pressure, repair time and repair direction. These parameters are transmitted to the control terminal of the repair equipment through the Internet of Things, and at the same time control the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment.

[0057] If the second discrimination result is obtained, it indicates that the target titanium alloy workpiece surface does not contain crack defects but does contain deformation defects. At this point, the processor identifies the deformation location area and segments the deformation defect image of that area. Based on this deformation defect image, a deformed 3D model of the deformation defect in the target titanium alloy workpiece is constructed, and its deformation variable is calculated. The calculation formula for the deformation variable D is:

[0058]

[0059] Where V actual and V ideal Represent the actual shape volume and the ideal shape volume respectively. Subsequently, the processor performs ratio processing on the deformation variable and the standard shape parameter of the target titanium alloy workpiece under the ideal state to obtain the deformation ratio. If the deformation ratio is greater than the preset ratio, it is determined that the workpiece cannot be repaired, and the industrial robot is controlled to transfer it to the scrap workstation for scrapping; if the deformation ratio is not greater than the preset ratio, the area is marked as an area requiring compensation and repair. The processor determines the compensation amount based on the deformation variable, and calculates the optimal compensation amount in combination with the preset forging parameters of the forging equipment. The calculation formula for the optimal compensation amount is C optimal =C preset +ΔC, where C preset is the preset compensation amount, and ΔC is the compensation increment adjusted according to the deformation. Ultimately, the updated optimal compensation parameters are transmitted to the control terminal of the forging equipment via the Internet of Things, which in turn controls the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment.

[0060] If the third discrimination result is obtained, it indicates that neither crack-related defects nor deformation-related defects exist on the target titanium alloy workpiece's surface. At this point, the processor directly controls the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the cooling workstation, where the cooling equipment cools the workpiece according to preset cooling parameters. These parameters include cooling rate, cooling time, and cooling medium type. These parameters are pre-set based on the thermophysical properties of the target titanium alloy material to ensure that the workpiece maintains good structural properties during the cooling process.

[0061] like Figure 2 As shown in the figure, the detailed process for state identification and parameter generation further refines the above steps. For example, when identifying crack defects and generating repair parameters, the processor not only needs to perform image segmentation and 3D reconstruction but also quickly identify the optimal repair parameters through database query and matching. When identifying deformation defects and generating compensation parameters, the processor combines geometric modeling and mechanical analysis techniques to ensure the accuracy of the compensation. The efficient execution of these steps is due to the powerful computing power of the processor and the real-time data transmission support of IoT technology.

[0062] like Figure 3 As shown in the figure, the titanium alloy forging process optimization system includes memory and a processor. The memory stores a titanium alloy forging process optimization method program. When running this program, the processor can access resources such as image data, knowledge graphs, and databases in the memory, and interact with peripheral equipment such as industrial robots, repair equipment, forging equipment, and cooling equipment through a control terminal. The system structure block diagram clearly shows the connection between each module and the data flow path, ensuring the automation and intelligence of the entire process.

[0063] In summary, the specific embodiments of the present invention utilize image acquisition and analysis technology to achieve real-time monitoring of dynamic changes during the titanium alloy forging process, and can quickly provide feedback on optimization suggestions, thereby improving the timeliness and accuracy of process parameter adjustments. By accurately identifying and treating crack and deformation defects, the quality stability and production efficiency of titanium alloy workpieces are effectively improved, meeting the modern industry's demand for efficient and intelligent forging technologies.

Claims

1. A titanium alloy forging process optimization method based on image processing, characterized in that: The following steps are involved: Acquire a surface image of a target titanium alloy workpiece during a forging process, and perform state recognition and judgment processing on the target titanium alloy workpiece based on the surface image to generate a judgment result; If the discrimination result is the first discrimination result, it indicates that there is a crack-type defect in the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment; If the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment. If the judgment result is the third judgment result, it means that there are neither crack-type defects nor deformation-type defects on the surface of the target titanium alloy workpiece. At this time, the industrial robot is directly controlled to transfer the target titanium alloy workpiece from the inspection workstation to the cooling workstation, so that the cooling equipment cools the target titanium alloy workpiece according to the preset cooling parameters.

2. The titanium alloy forging process optimization method based on image processing according to claim 1, characterized in that: Acquire a surface image of the target titanium alloy workpiece, and perform state recognition and judgment processing on the target titanium alloy workpiece based on the surface image to generate a judgment result, specifically: All possible defects that may occur in the forging process of titanium alloy workpieces are obtained through the big data network, and the defect feature images corresponding to each defect are obtained; the defect feature images corresponding to each defect are classified and processed to obtain defect feature images of crack defects and defect feature images of deformation defects; Constructing a first knowledge graph, and importing defect feature images of crack defects into the first knowledge graph; constructing a second knowledge graph, and importing defect feature images of deformation defects into the second knowledge graph; The surface image of the target titanium alloy workpiece is acquired through the industrial camera mounted on the inspection workstation; Importing the surface image into the first knowledge graph, calculating a first structural matching index between the surface image and each defect feature image in the first knowledge graph, and obtaining a plurality of first structural matching indexes; Comparing each first structure matching index with a preset threshold value respectively; If there is at least one first structure matching index greater than a preset threshold, generating a first discrimination result; If each first structural matching index is not greater than a preset threshold, importing the surface image into the second knowledge graph, calculating the second structural matching index between the surface image and each defect feature image in the second knowledge graph, and obtaining a plurality of second structural matching indexes; Comparing each second structure matching index with a preset threshold value respectively, and generating a second discrimination result if at least one second structure matching index is greater than the preset threshold value; If the second structure matching indexes are all not greater than the preset threshold, a third judgment result is generated.

3. The titanium alloy forging process optimization method based on image processing according to claim 1, characterized in that: If the discrimination result is the first discrimination result, it means that there is a crack defect on the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment. Specifically: If the discrimination result is the first discrimination result, it indicates that there is a crack defect in the surface of the target titanium alloy workpiece, and then the crack position area of ​​the crack defect in the target titanium alloy workpiece is obtained according to the surface image recognition of the target titanium alloy workpiece; Segmenting a crack defect image of a crack location area from the surface image, performing three-dimensional reconstruction processing on the crack defect image to obtain a three-dimensional crack model image of the crack-like defect; Determining optimal repair parameters for repairing the crack-like defects in the target titanium alloy workpiece based on a three-dimensional crack model diagram of the crack-like defects in the target titanium alloy workpiece; wherein the repair parameters include repair temperature, repair pressure, repair time, and repair direction; Obtain the coded identification information of the target titanium alloy workpiece, transmit the coded identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect and the optimal repair parameters to the control terminal of the repair equipment based on the Internet of Things, and control the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment.

4. The titanium alloy forging process optimization method based on image processing according to claim 3 is characterized in that: According to the three-dimensional crack model diagram of the crack defect in the target titanium alloy workpiece, the optimal repair parameters for repairing the crack defect in the target titanium alloy workpiece are determined, specifically: Develop in advance the optimal repair parameters of the repair equipment for various crack defect conditions, and obtain the optimal repair parameters corresponding to various crack defect conditions; and use 3D modeling software to draw 3D models of crack conditions corresponding to various crack defect conditions; Bundling optimal repair parameters corresponding to various crack defect working conditions and corresponding three-dimensional model diagrams of crack working conditions to obtain a plurality of repair parameter paired data packages; constructing a database, and importing each repair parameter paired data package into the database; Obtaining a three-dimensional crack model diagram of a crack-like defect in a target titanium alloy workpiece, and calculating similarities between the three-dimensional crack model diagram and three-dimensional crack condition model diagrams in each repair parameter pairing data package using a geometric matching method to obtain multiple similarities; The maximum similarity is extracted from the multiple similarities, and the three-dimensional crack model image corresponding to the maximum similarity is obtained, and the repair parameter pairing data package of the three-dimensional crack model image corresponding to the maximum similarity is extracted. According to the extracted repair parameter pairing data package, the optimal repair parameters for repairing crack defects in the target titanium alloy workpiece are obtained.

5. The titanium alloy forging process optimization method based on image processing according to claim 1, characterized in that: If the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment. Specifically: If the discrimination result is the second discrimination result, it means that there is no crack defect but there is a deformation defect on the surface of the target titanium alloy workpiece, and the deformation position area of ​​the deformation defect in the target titanium alloy workpiece is obtained according to the surface image recognition of the target titanium alloy workpiece; Segmenting a deformation defect image of the deformation position area in the surface image, and constructing a deformation three-dimensional model diagram of the deformation defect in the target titanium alloy workpiece according to the deformation defect image; Calculating the deformation amount of the deformed three-dimensional model image; Obtaining the standard shape parameters of the target titanium alloy workpiece under an ideal state, performing ratio processing on the deformation variable of the deformed three-dimensional model diagram and the standard shape parameters of the target titanium alloy workpiece under the ideal state to obtain the deformation ratio; Comparing the deformation ratio with a preset ratio; If the deformation ratio is greater than a preset ratio, the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the scrapping workstation to scrap the target titanium alloy workpiece.

6. The titanium alloy forging process optimization method based on image processing according to claim 5, characterized in that: The following steps are also included: If the deformation ratio is not greater than the preset ratio, the deformation position area of ​​the deformation defect in the target titanium alloy workpiece is marked as an area requiring compensation and repair; Obtain the deformation amount of the deformation defect in the area that needs to be compensated and repaired; Determining the compensation amount required to repair the area to be compensated and repaired according to the deformation amount; Obtaining preset forging parameters of a forging device, and extracting a preset compensation amount for forging the area requiring compensation repair in the target titanium alloy workpiece according to the preset forging parameters; Determining the optimal compensation amount for the area to be compensated and repaired based on the compensation amount and a preset compensation amount; Obtaining a forging time node at which the forging equipment performs forging on the area requiring compensation repair, replacing a preset compensation amount at the forging time node with an optimal compensation amount, and after the replacement is completed, updating the preset forging parameters to obtain the optimal compensation parameters for forging the target titanium alloy workpiece; Obtain the coded identification information of the target titanium alloy workpiece, transmit the coded identification information and the optimal compensation parameters of the target titanium alloy workpiece to the control terminal of the forging equipment based on the Internet of Things, and control the industrial robot to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment.

7. A titanium alloy forging process optimization system based on image processing, used to implement the titanium alloy forging process optimization method based on image processing according to claims 1-6, characterized in that: The titanium alloy forging process optimization system includes a memory and a processor. The memory stores a titanium alloy forging process optimization method program. When the titanium alloy forging process optimization method program is executed by the processor, the following steps are implemented: Acquire a surface image of a target titanium alloy workpiece, and perform state recognition and judgment processing on the target titanium alloy workpiece based on the surface image to generate a judgment result; If the discrimination result is the first discrimination result, it indicates that there is a crack-type defect in the surface of the target titanium alloy workpiece. At this time, the optimal repair parameters are generated, the coding identification information of the target titanium alloy workpiece, the crack location area of ​​the crack-type defect, and the optimal repair parameters are transmitted to the repair equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the repair equipment; If the discrimination result is the second discrimination result, it means that there are no crack defects but deformation defects on the surface of the target titanium alloy workpiece. At this time, the optimal compensation parameters are generated, the coded identification information of the target titanium alloy workpiece and the optimal compensation parameters are transmitted to the forging equipment, and the industrial robot is controlled to transfer the target titanium alloy workpiece from the inspection workstation to the forging equipment. If the judgment result is the third judgment result, it means that there are neither crack-type defects nor deformation-type defects on the surface of the target titanium alloy workpiece. At this time, the industrial robot is directly controlled to transfer the target titanium alloy workpiece from the inspection workstation to the cooling workstation, so that the cooling equipment cools the target titanium alloy workpiece according to the preset cooling parameters.

8. The titanium alloy forging process optimization system based on image processing according to claim 7 is characterized in that: The system also includes an industrial camera for acquiring a surface image of a target titanium alloy workpiece; the industrial camera is installed on the inspection workstation.

9. The titanium alloy forging process optimization system based on image processing according to claim 7, characterized in that: The system also includes a repair device, a forging device, a cooling device and an industrial manipulator. The repair device, forging device and cooling device are respectively connected to the processor for communication. The industrial manipulator is used to transfer the target titanium alloy workpiece between different workstations.

10. The titanium alloy forging process optimization system based on image processing according to claim 7, characterized in that: The memory also stores a first knowledge graph and a second knowledge graph, wherein the first knowledge graph is used to store defect feature images of crack-type defects, and the second knowledge graph is used to store defect feature images of deformation-type defects.

Images