A method and device for intelligently detecting the impact performance of titanium alloy forging blanks

Abstract

The present disclosure provides a method and device for intelligently detecting the impact performance of titanium alloy forging blanks, which relates to the technical field of industrial production, specifically to the technical fields of deep learning, image processing, etc., and can be applied to the scenario of manufacturing aviation ring forgings. The specific implementation scheme includes: impacting the blank to be detected by a testing machine to obtain the blank to be detected after the impact; photographing the cross section of the blank to be detected after the impact by an image acquisition module to obtain a cross-sectional image; inputting the cross-sectional image into a pre-trained image classification model, and outputting the classification result of the cross-sectional image through the image classification model, and the classification result includes whether the cross-sectional image is qualified or unqualified; according to the classification result, determining the impact performance test result of the blank to be detected. The present disclosure can improve the efficiency of detecting the impact performance of titanium alloys.

Description

Technical Field

[0001] The present disclosure relates to the field of industrial production technology, specifically to technical fields such as deep learning and image processing, and can be applied to the scenario of manufacturing aviation ring forgings, and in particular to a method and device for intelligently detecting the impact performance of titanium alloy forging blanks. Background Art

[0002] Titanium alloy forgings are parts made of titanium alloy materials through forging technology. Titanium alloy is a metal material widely used in aerospace, medical equipment, automobiles and other high-performance engineering fields. The impact performance quality of titanium alloy under dynamic load is related to the quality of titanium alloy forgings during use. Therefore, before processing titanium alloy, it is necessary to conduct an impact test on the titanium alloy billet to detect the impact performance of the titanium alloy material under dynamic load, so as to evaluate the impact performance quality of the titanium alloy billet.

[0003] At present, impact testing of titanium alloy billets mainly relies on manual control, through manual operation of the impact testing machine and subsequent testing to determine the impact properties of titanium alloy materials.

[0004] However, the above-mentioned method of manually testing the impact performance of titanium alloy is too inefficient. Summary of the invention

[0005] The present invention provides a method and device for intelligently detecting the impact performance of a titanium alloy forging blank, which can improve the efficiency of detecting the impact performance of the titanium alloy.

[0006] According to a first aspect of the present disclosure, a method for intelligently detecting the impact performance of a titanium alloy forging blank is provided, the method comprising: impacting a blank to be detected by a testing machine to obtain the blank to be detected after the impact; photographing a cross section of the blank to be detected after the impact by an image acquisition module to obtain a cross-sectional image; inputting the cross-sectional image into a pre-trained image classification model, and outputting a classification result of the cross-sectional image by the image classification model, the classification result including whether the cross-sectional image is qualified or unqualified, the pre-trained image classification model uses the cross-sectional image of a qualified blank and the cross-sectional image of an unqualified blank as input, marks the cross-sectional image of the qualified blank as 1, and marks the cross-sectional image of the unqualified blank as 0 as output, and trains a neural network; according to the classification result, determines the impact performance test result of the blank to be detected.

[0007] According to a second aspect of the present disclosure, a device for intelligently detecting the impact performance of a titanium alloy forging blank is provided. The device includes: an acquisition unit and a determination unit.

[0008] The acquisition unit is used to impact the blank to be tested through a testing machine to obtain the blank to be tested after the impact; to shoot the cross section of the blank to be tested after the impact through an image acquisition module to obtain a cross-sectional image; to input the cross-sectional image into a pre-trained image classification model, and to output a classification result of the cross-sectional image through the image classification model, the classification result including whether the cross-sectional image is qualified or unqualified. The pre-trained image classification model uses the cross-sectional image of a qualified blank and the cross-sectional image of an unqualified blank as input, marks the cross-sectional image of the qualified blank as 1, and marks the cross-sectional image of the unqualified blank as 0 as output, and trains the neural network to obtain the obtained result.

[0009] The determination unit is used to determine the impact performance test result of the blank to be tested according to the classification result.

[0010] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method of the first aspect.

[0011] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause a computer to execute the method according to the first aspect.

[0012] According to a fifth aspect of the present disclosure, there is provided a computer program product, comprising a computer program, which implements the method according to the first aspect when executed by a processor.

[0013] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0015] Figure 1 A schematic flow chart of a method for intelligently detecting the impact performance of a titanium alloy forging blank provided in an embodiment of the present disclosure;

[0016] Figure 2 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure;

[0017] Figure 3 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure;

[0018] Figure 4 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure;

[0019] Figure 5 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure;

[0020] Figure 6 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure;

[0021] Figure 7 A schematic diagram of the composition of a device for intelligently detecting the impact performance of a titanium alloy forging blank provided in an embodiment of the present disclosure;

[0022] Figure 8 An exemplary block diagram of an exemplary electronic device 800 that can be used to implement an embodiment of the present disclosure is provided for the present disclosure. DETAILED DESCRIPTION

[0023] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0024] It should be understood that in the embodiments of the present disclosure, the character " / " generally indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", etc. are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0025] Titanium alloy forgings are parts made of titanium alloy materials through forging technology. Titanium alloy is a metal material widely used in aerospace, medical equipment, automobiles and other high-performance engineering fields. The impact performance quality of titanium alloy under dynamic load is related to the quality of titanium alloy forgings during use. Therefore, before processing titanium alloy, it is necessary to conduct an impact test on the titanium alloy billet to detect the impact performance of the titanium alloy material under dynamic load, so as to evaluate the impact performance quality of the titanium alloy billet.

[0026] For example, the impact performance of metal materials is a very important part of mechanical properties. The performance of metal materials under dynamic loads and static loads is different. Materials that show good plasticity under static loads may show brittleness under dynamic loads. Therefore, it is necessary to carry out impact tests on metal materials subjected to dynamic loads to determine their dynamic load mechanical properties. In the determination, impact bending tests are often used to inspect the quality of metal materials, reveal the influence of internal defects on the mechanical properties that cannot be revealed in static load tests of metal materials, and the possibility that metal materials have brittle tendencies under certain conditions (such as low temperature, etc.). The above-mentioned detection of metal materials can be used to evaluate the metallurgical quality of raw materials, control the quality of hot working processes, evaluate the brittle transformation trend of metal materials at constant temperatures, and determine the strain aging sensitivity of metal materials, as a resistance index when metal materials are subjected to large energy impacts or as a performance index for evaluating the life and reliability of certain components.

[0027] At present, impact testing of titanium alloy billets mainly relies on manual control, through manual operation of the impact testing machine and subsequent testing to determine the impact properties of titanium alloy materials.

[0028] However, the above-mentioned method of manually testing the impact performance of titanium alloy is too inefficient.

[0029] Against this background, the present disclosure provides a method for intelligently detecting the impact properties of a titanium alloy forging blank, which can improve the efficiency of detecting the impact properties of the titanium alloy.

[0030] Exemplarily, the execution subject of the method for intelligently detecting the impact performance of titanium alloy forging blanks may be a computer or a server, or may also be other equipment with data processing capabilities. The execution subject of the method is not limited here.

[0031] In some embodiments, the server may be a single server, or a server cluster composed of multiple servers. In some implementations, the server cluster may also be a distributed cluster. The present disclosure does not limit the specific implementation of the server.

[0032] Figure 1 The present invention provides a flow chart of a method for intelligently detecting the impact performance of a titanium alloy forging blank. Figure 1 As shown, the method may include:

[0033] S101, impacting the blank to be tested by a testing machine to obtain the blank to be tested after impact.

[0034] Exemplarily, the testing machine may include a base, a sample support, a protective cover, and a pendulum. The blank to be tested may be fixed on the sample support, the pendulum is raised to a height of h1, and then dropped, and the pendulum is used to impact the blank to be tested, and the blank to be tested after the impact is obtained. The blank to be tested may be broken by the pendulum, and a cross-sectional image of the blank to be tested may be more conveniently collected.

[0035] For example, corresponding pendulum impact points may be marked on the blank to be inspected, and the pendulum impact points on the blank to be inspected may be impacted by the pendulum to obtain the blank to be inspected after the impact.

[0036] S102 , photographing the cross section of the blank to be inspected after impact by an image acquisition module to obtain a cross-sectional image, wherein the cross section is the cross section of the blank to be inspected after impact.

[0037] Exemplarily, the image acquisition module may be a camera, or other device or module with an image acquisition function. There is no specific limitation on the implementation of the image acquisition module, as long as it has the image acquisition function.

[0038] For example, a cross-section of the blank to be inspected after impact may be photographed by a camera to obtain a cross-section image.

[0039] S103. Input the cross-sectional image into a pre-trained image classification model, and output the classification result of the cross-sectional image through the image classification model. The classification result includes whether the cross-sectional image is qualified or unqualified. The pre-trained image classification model uses the cross-sectional image of qualified billets and the cross-sectional image of unqualified billets as input, marks the cross-sectional image of qualified billets as 1, and marks the cross-sectional image of unqualified billets as 0 as output, and trains the neural network to obtain the result.

[0040] Exemplarily, the pre-trained image classification model can be pre-trained, and can be obtained by training a neural network by taking a large number of cross-sectional images of qualified blanks and cross-sectional images of unqualified blanks as input, and manually marking the qualified cross-sectional images as 1 and the cross-sectional images of unqualified blanks as 0 as output. Through the image classification model, the classification result of the cross-sectional image can be output.

[0041] For example, the pre-trained image classification model can be pre-trained, and can be obtained by training a neural network using 10,000 cross-sectional images of qualified blanks and 10,000 cross-sectional images of unqualified blanks as input, and manually marking the qualified cross-sectional images as 1 and the unqualified cross-sectional images as 0 as output. Through the image classification model, the classification result of the cross-sectional image can be output as "qualified".

[0042] S104: Determine the impact performance test result of the blank to be tested according to the classification result.

[0043] Exemplarily, according to the classification result of the cross-sectional image being “qualified”, it can be determined that the impact performance test result of the blank to be tested is “qualified”.

[0044] In this embodiment, the pre-trained image classification model may include a convolution layer, a pooling layer, a fully connected layer, and a Softmax layer. Before step S103, the cross-sectional image may be preprocessed, for example: the cross-sectional image is processed by pixel value processing to obtain a processed cross-sectional image; or the cross-sectional image is preprocessed by cropping, rotating, etc. Further, the preprocessed cross-sectional image may be input into the convolution layer, the convolution layer convolves the processed cross-sectional image, and outputs the convolved cross-sectional image. The convolved cross-sectional image is input into the pooling layer, and the pooling layer outputs the pooled feature map of the cross-sectional image. The pooled feature map of the cross-sectional image is input into the fully connected layer, and the image feature tensor of the cross-sectional image is output through the fully connected layer. The image feature tensor is input into the Softmax layer, and the Softmax layer outputs the qualified probability of the cross-sectional image. According to the qualified probability, the classification result of the cross-sectional image is determined.

[0045] The present invention uses a testing machine to impact a blank to be tested, obtains the blank to be tested after the impact, uses an image acquisition module to shoot a cross-section of the blank to be tested after the impact, obtains a cross-section image, inputs the cross-section image into a pre-trained image classification model, outputs a classification result of the cross-section image through the image classification model, and determines a test result of the impact performance of the blank to be tested based on the classification result, thereby improving the efficiency of testing the impact performance of the blank to be tested.

[0046] Figure 2 Another schematic diagram of a method for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure. Figure 2 As shown, before the cross-sectional image is input into a pre-trained image classification model and the classification result of the cross-sectional image is output through the image classification model, the method for intelligently detecting the impact properties of titanium alloy forging blanks may further include steps S201-S202.

[0047] S201, obtaining the type of the blank to be detected.

[0048] Exemplarily, the blank to be tested may include at least one of the following: plate-shaped, rod-shaped, tube-shaped, and wire-shaped. Different types of blanks to be tested may have different impact properties.

[0049] S202. According to the type of the blank to be detected, determine a target image classification model corresponding to the type of the blank to be detected from a plurality of preset image classification models.

[0050] Exemplarily, different multiple image classification models can be preset for different types of blanks to be detected. According to the type of the blank to be detected, the image classification model corresponding to the type of the blank to be detected can be determined from the preset multiple different image classification models. After the target image classification model is determined, the cross-sectional image of the blank to be detected can be input into the target image classification model, and the classification result of the cross-sectional image can be output through the target image classification model.

[0051] For example, the preset multiple image classification models are: Model 1, Model 2, and the corresponding types of the blanks to be detected are: plate-shaped, rod-shaped. When the type of the blank to be detected is plate-shaped, the target image classification model corresponding to the plate-shaped can be determined from the preset multiple image classification models, that is, the determined target image classification model is Model 1.

[0052] This embodiment obtains the type of the blank to be detected, and determines the target image classification model corresponding to the type of the blank to be detected from multiple preset image classification models according to the type of the blank to be detected, thereby improving the accuracy of classification of the blank to be detected, adapting to different types of blanks to be detected, reducing the misjudgment rate, and further improving the efficiency of detecting the impact performance of the blank to be detected.

[0053] Figure 3 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure. Figure 3 As shown, the method for intelligently detecting the impact performance of titanium alloy forging blanks may also include steps S301-S302. Figure 1 Step S104 includes: generating an impact performance test result of the blank to be tested according to the classification result and the first impact toughness score.

[0054] S301. Output an image score of a cross-sectional image through an image classification model, where the image score is used to indicate the probability that the cross-sectional image is a qualified cross-sectional image.

[0055] For example, the image classification model can also output the image score of the cross-sectional image. For example, if the image score of the cross-sectional image of the blank to be inspected is 80 points, it means that the probability that the cross-sectional image is a qualified cross-sectional image is 80%.

[0056] S302 . Determine a first impact toughness score of the blank to be tested according to the image score, where the first impact toughness score is used to indicate the impact performance of the blank to be tested.

[0057] Exemplarily, the first impact toughness score can be directly determined by the image score, or can be obtained by converting the image score. After determining the first impact toughness score, the impact performance test result of the blank to be tested can be generated according to the classification result and the first impact toughness score.

[0058] For example, the classification result of the blank to be tested can be "qualified", the image score of the cross-sectional image of the blank to be tested can be 80 points, and the first impact toughness score of the blank to be tested can be 80 points, indicating that the impact performance of the blank to be tested is 80, and the generated impact performance test result of the blank to be tested is "qualified, 80".

[0059] For another example, the classification result of the blank to be tested may be "unqualified", the image score of the cross-sectional image of the blank to be tested is 50 points, and the first impact toughness score of the blank to be tested obtained by converting the image score may be 0.5, indicating that the impact performance of the blank to be tested is 0.5, and the generated impact performance test result of the blank to be tested is "unqualified, 0.5".

[0060] This embodiment outputs the image score of the cross-sectional image through an image classification model, determines the first impact toughness score of the blank to be tested based on the image score, and generates the impact performance test result of the blank to be tested based on the classification result and the first impact toughness score. The impact performance of the blank to be tested can be comprehensively evaluated and the test results can be quantified, thereby improving the accuracy and efficiency of the impact performance test of the blank to be tested.

[0061] Figure 4 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure. Figure 4 As shown, the method for intelligently detecting the impact performance of titanium alloy forging blanks may also include steps S401-S404. Figure 1 Step S104 includes: generating an impact performance test result of the blank to be tested according to the classification result and the second impact toughness score.

[0062] S401, determining the energy value lost by impacting the blank to be tested by a testing machine.

[0063] For example, the energy value lost by impacting the blank to be tested can be determined by the testing machine in the following manner: a pendulum with a weight of G is released from a height of h1, and the pendulum breaks the blank to be tested and then rises to a height of h2. Then, the energy value K lost by impacting the blank to be tested can be calculated as:

[0064] K=G*(h1-h2) Formula (1)

[0065] In formula (1), K represents the energy value lost by impacting the blank to be tested, h1 represents the height of the pendulum before it is released, and h2 represents the height of the pendulum after it impacts the blank to be tested.

[0066] For example, a corresponding V-shaped notch may be provided on the blank to be inspected so that the pendulum can better break the blank to be inspected.

[0067] S402, determining the cross-sectional area of ​​the blank to be inspected through the cross-sectional image.

[0068] Exemplarily, the cross-sectional image can be detected by an image detection model, and the cross-sectional area of ​​the blank to be detected can be output by the image detection model. The image detection model can be obtained by training a neural network using the cross-sectional image of the blank to be detected as input and the cross-sectional area corresponding to each cross-sectional image as output, and the implementation of the image detection model is not limited here.

[0069] For example, after the pendulum breaks the blank to be tested, the crack source will first be formed near the notch of the blank to be tested, followed by the fiber zone, the radial zone and the shear lip zone. The shear lip zone is distributed along the other three sides of the edge without the notch, and the boundary connecting the fiber zone with the radial zone or the shear lip zone is often arc-shaped. The better the plasticity of the blank to be tested, the smaller the radial zone. If the plasticity of the blank to be tested is good enough, the radial zone disappears completely, and there are only two areas, the fiber zone and the shear lip, on the cross section of the entire blank to be tested. Under the impact of the pendulum, one side of the V-shaped notch is subjected to tensile stress, and the other side without the notch is subjected to tensile stress. Therefore, the force directions on the entire cross section of the blank to be tested are different. Therefore, the radial zone that appears under tensile stress may disappear when entering the compressive zone, and the fiber zone will reappear. The main reason for the appearance of fiber zones at two places on the notch is that when the crack enters the compressive stress zone, the compression deformation plays a blocking role in the expansion of the crack, which significantly reduces the expansion speed.

[0070] Optionally, the cross-sectional area of ​​the blank to be inspected determined by the cross-sectional image is A.

[0071] S403, determining the impact toughness of the blank to be tested according to the energy value and the cross-sectional area.

[0072] For example, the calculation method for determining the impact toughness B of the blank to be tested according to the energy value and the cross-sectional area may be:

[0073] B=K / A Formula (2)

[0074] In formula (2), B represents the impact toughness of the blank to be tested, K represents the energy value lost by impacting the blank to be tested, and A represents the cross-sectional area of ​​the blank to be tested.

[0075] For example, according to the energy value and the cross-sectional area, the impact toughness of the blank to be tested is calculated as 70 by formula (2).

[0076] S404. Determine a second impact toughness score of the blank to be tested according to the impact toughness, where the second impact toughness score is used to indicate the impact performance of the blank to be tested.

[0077] For example, the second impact toughness score can be directly determined by the impact toughness, or can be obtained by converting the impact toughness. After determining the second impact toughness score, the impact performance test result of the blank to be tested can be generated according to the classification result and the second impact toughness score.

[0078] For example, the classification result of the blank to be tested may be "qualified", the impact toughness of the blank to be tested may be 70, and the second impact toughness score of the blank to be tested may be 70 points, indicating that the impact performance of the blank to be tested is 70, and the generated impact performance test result of the blank to be tested is "qualified, 70".

[0079] For another example, the classification result of the blank to be tested may be "unqualified", the impact toughness of the blank to be tested is 50 points, and the second impact toughness score of the blank to be tested obtained by converting the impact toughness may be 0.5, indicating that the impact performance of the blank to be tested is 0.5, and the generated impact performance test result of the blank to be tested is "unqualified, 0.5".

[0080] In this embodiment, the energy value lost by impacting the blank to be tested is determined by a testing machine, the cross-sectional area of ​​the blank to be tested is determined by a cross-sectional image, the impact toughness of the blank to be tested is determined according to the energy value and the cross-sectional area, the second impact toughness score of the blank to be tested is determined according to the impact toughness, and the impact performance test result of the blank to be tested is generated according to the classification result and the second impact toughness score. The impact performance of the blank to be tested can be further comprehensively evaluated, and the test results can be quantified, thereby further improving the accuracy and efficiency of the impact performance test of the blank to be tested.

[0081] Figure 5 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure. Figure 5 As shown, the method for intelligently detecting the impact performance of titanium alloy forging blanks may also include steps S501-S502. Figure 1 Step S104 includes: generating an impact performance test result of the blank to be tested according to the classification result and the third impact toughness score.

[0082] S501, determining the shear cross-sectional ratio of the blank to be tested by using a planimeter.

[0083] Exemplarily, a cross section of the blank to be inspected may be magnified using a suitable magnification, and a cross-sectional image may be acquired using an image acquisition module, and a planimeter may be used to measure the shear cross section rate of the blank to be inspected by cleavage.

[0084] For example, the shear cross-sectional ratio of the blank to be tested is determined to be 80% by means of a planimeter.

[0085] S502. Determine a third impact toughness score of the blank to be tested according to the shear section ratio, where the third impact toughness score is used to indicate the impact performance of the blank to be tested.

[0086] Exemplarily, the shear section rate of the blank to be tested is determined to be 80% by the planimeter. The third impact toughness score can be directly determined by the shear section rate, or it can be obtained by converting the shear section rate. After determining the third impact toughness score, the impact performance test result of the blank to be tested can be generated according to the classification result and the third impact toughness score. After determining the third impact toughness score, the impact performance test result of the blank to be tested can be generated according to the classification result and the third impact toughness score.

[0087] For example, the classification result of the blank to be tested can be "qualified", the shear section rate of the blank to be tested can be 80%, and the third impact toughness score of the blank to be tested can be determined to be 80 points, indicating that the impact performance of the blank to be tested is 80, and the generated impact performance test result of the blank to be tested is "qualified, 80".

[0088] For another example, the classification result of the blank to be tested may be "unqualified", the shear section rate of the blank to be tested is 50 points, and the third impact toughness score of the blank to be tested obtained by converting the shear section rate may be 0.5, indicating that the impact performance of the blank to be tested is 0.5, and the generated impact performance test result of the blank to be tested is "unqualified, 0.5".

[0089] In this embodiment, the shear section ratio of the blank to be tested is determined by a planimeter, and the third impact toughness score of the blank to be tested is determined based on the shear section ratio. According to the classification result and the third impact toughness score, the impact performance test result of the blank to be tested is generated, and the impact performance of the blank to be tested can be further comprehensively evaluated, and the test results can be quantified, thereby further improving the accuracy and efficiency of the impact performance test of the blank to be tested.

[0090] Figure 6 Another schematic diagram of a process for intelligently detecting the impact performance of a titanium alloy forging blank provided by an embodiment of the present disclosure. Figure 6 As shown, the method for intelligently detecting the impact performance of titanium alloy forging blanks may also include steps S601-S602. Figure 1 Step S104 includes: generating an impact performance test result of the blank to be tested according to the classification result and the fourth impact toughness score.

[0091] S601. Obtain at least two target impact toughness scores among a first impact toughness score, a second impact toughness score, and a third impact toughness score, wherein the first impact toughness score is determined according to the image score output by the image classification model, the second impact toughness score is determined according to the impact toughness of the blank to be tested determined according to the energy value and the cross-sectional area, and the third impact toughness score is determined according to the shear section rate of the blank to be tested determined by the integrator.

[0092] Exemplarily, the first impact toughness score can be determined based on the image score output by the image classification model, the second impact toughness score can be determined based on the impact toughness of the blank to be tested determined by the energy value and the cross-sectional area, and the third impact toughness score can be determined based on the shear section rate of the blank to be tested determined by the integrator.

[0093] For example, the first impact toughness score determined according to the image score output by the image classification model can be "85", the second impact toughness score determined according to the impact toughness of the blank to be tested determined by the energy value and the cross-sectional area can be "90", and the third impact toughness score determined according to the shear section rate of the blank to be tested determined by the integrator can be "95".

[0094] Optionally, the target impact toughness score obtained may be "the first impact toughness score, the second impact toughness score", or "the first impact toughness score, the third impact toughness score", or "the second impact toughness score, the third impact toughness score", or "the first impact toughness score, the second impact toughness score, the third impact toughness score".

[0095] S602: Perform weighted averaging on at least two target impact toughness scores according to preset weights to determine a fourth impact toughness score of the blank to be tested.

[0096] Exemplarily, the preset weight may be determined according to the number of target impact toughness scores obtained, or may be determined according to a weight set by a user. After determining the fourth impact toughness score, an impact performance test result of the blank to be tested may be generated according to the classification result and the fourth impact toughness score.

[0097] For example, the preset weights may be "1 / 2, 1 / 2", and the preset weights may also be "1 / 3, 1 / 3, 1 / 3". The classification result of the blank to be tested may be "qualified", and the target impact toughness scores obtained may be: the first impact toughness score "85", the second impact toughness score "90", and the third impact toughness score "95". The preset weights may be "1 / 3, 1 / 3, 1 / 3", and the fourth impact toughness score of the blank to be tested may be determined to be "90". The generated impact performance test result of the blank to be tested is "qualified, 0.5".

[0098] Optionally, the classification result of the blank to be tested may be "unqualified", the target impact toughness scores obtained may be: the first impact toughness score "40", the second impact toughness score "30", the preset weights may be "1 / 2, 1 / 2", and the fourth impact toughness score of the blank to be tested may be "35". The generated impact performance test result of the blank to be tested is "unqualified, 35".

[0099] This embodiment obtains the first impact toughness score, the second impact toughness score, and at least two target impact toughness scores of the first impact toughness score, and weightedly averages the at least two target impact toughness scores according to preset weights to determine the fourth impact toughness score of the blank to be tested, and generates an impact performance test result of the blank to be tested based on the classification result and the fourth impact toughness score. The impact performance of the blank to be tested can be further comprehensively evaluated through multiple indicators, and the test results can be quantified, thereby further improving the accuracy and efficiency of the impact performance test of the blank to be tested.

[0100] In an exemplary embodiment, the present disclosure also provides a device for intelligently detecting the impact properties of a titanium alloy forging blank, which can be used to implement the method for intelligently detecting the impact properties of a titanium alloy forging blank as described in the aforementioned embodiment. Figure 7 The present invention provides a schematic diagram of the composition of a device for intelligently detecting the impact performance of a titanium alloy forging blank. Figure 7 As shown, the device may include: an acquisition unit 701 and a determination unit 702.

[0101] The acquisition unit 701 is used to impact the blank to be tested through a testing machine to obtain the blank to be tested after the impact; shoot the cross-section of the blank to be tested after the impact through an image acquisition module to obtain a cross-sectional image, wherein the cross-section is the cross-section of the blank to be tested after the impact; input the cross-sectional image into a pre-trained image classification model, and output a classification result of the cross-sectional image through the image classification model, wherein the classification result includes whether the cross-sectional image is qualified or unqualified. The pre-trained image classification model uses the cross-sectional image of a qualified blank and the cross-sectional image of an unqualified blank as input, marks the cross-sectional image of the qualified blank as 1, and marks the cross-sectional image of the unqualified blank as 0 as output, and trains the neural network to obtain the obtained result.

[0102] The determination unit 702 is used to determine the impact performance test result of the blank to be tested according to the classification result.

[0103] Optionally, before the cross-sectional image is input into a pre-trained image classification model and the classification result of the cross-sectional image is output through the image classification model, the acquisition unit 701 is also used to: obtain the type of the blank to be detected; according to the type of the blank to be detected, determine a target image classification model corresponding to the type of the blank to be detected from multiple preset image classification models; the acquisition unit 701 is specifically used to: input the cross-sectional image into the target image classification model, and output the classification result of the cross-sectional image through the target image classification model.

[0104] Optionally, the determination unit 702 is also used to: output an image score of the cross-sectional image through an image classification model, the image score is used to indicate the probability that the cross-sectional image is a qualified cross-sectional image; determine a first impact toughness score of the blank to be tested based on the image score, the first impact toughness score is used to indicate the impact performance of the blank to be tested; the determination unit 702 is specifically used to: generate an impact performance test result of the blank to be tested based on the classification result and the first impact toughness score.

[0105] Optionally, the determination unit 702 is also used to: determine the energy value lost by impacting the blank to be tested through a testing machine; determine the cross-sectional area of ​​the blank to be tested through a cross-sectional image; determine the impact toughness of the blank to be tested based on the energy value and the cross-sectional area; determine a second impact toughness score of the blank to be tested based on the impact toughness, the second impact toughness score being used to indicate the impact performance of the blank to be tested; the determination unit 702 is specifically used to: generate an impact performance test result of the blank to be tested based on the classification result and the second impact toughness score.

[0106] Optionally, the determination unit 702 is also used to: determine the shear section ratio of the blank to be tested by a planimeter; determine the third impact toughness score of the blank to be tested based on the shear section ratio, and the impact toughness score is used to indicate the impact performance of the blank to be tested; the determination unit 702 is specifically used to: generate an impact performance test result of the blank to be tested based on the classification result and the third impact toughness score.

[0107] Optionally, the determination unit 702 is also used to: obtain at least two impact toughness scores among a first impact toughness score, a second impact toughness score, and a third impact toughness score, wherein the first impact toughness score is the image score of the cross-sectional image output by the image classification model, the second impact toughness score is the impact toughness of the blank to be tested determined according to the energy value and the cross-sectional area, and the third impact toughness score is the shear section rate of the blank to be tested determined by a planimeter; weightedly average at least two target indicators according to preset weights to determine a fourth impact toughness score of the blank to be tested; the determination unit 702 is specifically used to: generate an impact performance test result of the blank to be tested based on the classification result and the fourth impact toughness score.

[0108] Optionally, the type of the blank to be inspected includes at least one of the following: plate-shaped, rod-shaped, tube-shaped, and wire-shaped.

[0109] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0110] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0111] In an exemplary embodiment, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in the above embodiments.

[0112] In an exemplary embodiment, the readable storage medium may be a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to the above embodiments.

[0113] In an exemplary embodiment, a computer program product includes a computer program, and when the computer program is executed by a processor, the method according to the above embodiments is implemented.

[0114] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0115] like Figure 8 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0116] Multiple components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0117] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as a method for intelligently detecting the impact performance of a titanium alloy forging blank. For example, in some embodiments, the method for intelligently detecting the impact performance of a titanium alloy forging blank may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the method for intelligently detecting the impact performance of a titanium alloy forging blank described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the method for intelligently detecting the impact properties of titanium alloy forging blanks in any other appropriate manner (for example, by means of firmware).

[0118] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0120] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0123] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0124] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0125] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for intelligently detecting the impact performance of a titanium alloy forging blank, the method comprising: Impacting the blank to be tested by a testing machine to obtain the blank to be tested after impact; The cross section of the blank to be inspected after the impact is photographed by an image acquisition module to obtain a cross-sectional image, wherein the cross section is a cross section of the blank to be inspected after the impact; Inputting the cross-sectional image into a pre-trained image classification model, and outputting a classification result of the cross-sectional image through the image classification model, wherein the classification result includes whether the cross-sectional image is qualified or unqualified, wherein the pre-trained image classification model uses the cross-sectional image of a qualified blank and the cross-sectional image of an unqualified blank as input, marks the cross-sectional image of the qualified blank as 1, and marks the cross-sectional image of the unqualified blank as 0 as output, and trains a neural network; Obtain at least two target impact toughness scores among a first impact toughness score, a second impact toughness score, and a third impact toughness score, wherein the first impact toughness score is the image score of the cross-sectional image output by the image classification model, the second impact toughness score is the impact toughness of the blank to be tested determined according to the energy value and the cross-sectional area, and the third impact toughness score is the shear section rate of the blank to be tested determined by a planimeter; weighted average the at least two target impact toughness scores according to preset weights to determine a fourth impact toughness score of the blank to be tested; and generate an impact performance test result of the blank to be tested according to the classification result and the fourth impact toughness score.

2. The method according to claim 1, before inputting the cross-sectional image into a pre-trained image classification model and outputting the classification result of the cross-sectional image through the image classification model, the method further comprises: Obtaining the type of the blank to be detected; According to the type of the blank to be detected, a target image classification model corresponding to the type of the blank to be detected is determined from a plurality of preset image classification models; the cross-sectional image is input into a pre-trained image classification model, and the classification result of the cross-sectional image is output through the image classification model, including: the cross-sectional image is input into the target image classification model, and the classification result of the cross-sectional image is output through the target image classification model.

3. The method according to any one of claims 1 to 2, wherein the type of the blank to be inspected comprises at least one of the following: plate-shaped, rod-shaped, tube-shaped or wire-shaped.

4. A device for intelligently detecting the impact performance of titanium alloy forging blanks, the device comprising: An acquisition unit is used to impact the blank to be tested through a testing machine and acquire the blank to be tested after the impact; The cross section of the blank to be inspected after the impact is photographed by an image acquisition module to obtain a cross-sectional image, wherein the cross section is a cross section of the blank to be inspected after the impact; Inputting the cross-sectional image into a pre-trained image classification model, and outputting a classification result of the cross-sectional image through the image classification model, wherein the classification result includes whether the cross-sectional image is qualified or unqualified, wherein the pre-trained image classification model uses the cross-sectional image of a qualified blank and the cross-sectional image of an unqualified blank as input, marks the cross-sectional image of the qualified blank as 1, and marks the cross-sectional image of the unqualified blank as 0 as output, and trains a neural network; A determination unit is used to obtain at least two impact toughness scores among a first impact toughness score, a second impact toughness score, and a third impact toughness score, wherein the first impact toughness score is the image score of the cross-sectional image output by the image classification model, the second impact toughness score is the impact toughness of the blank to be tested determined according to the energy value and the cross-sectional area, and the third impact toughness score is the shear section rate of the blank to be tested determined by a planimeter; at least two target indicators are weighted and averaged according to preset weights to determine a fourth impact toughness score of the blank to be tested; the determination unit is specifically used to: generate an impact performance test result of the blank to be tested according to the classification result and the fourth impact toughness score.

5. According to the device of claim 4, before inputting the cross-sectional image into a pre-trained image classification model and outputting the classification result of the cross-sectional image through the image classification model, the acquisition unit is also used to: obtain the type of the blank to be detected; and determine, according to the type of the blank to be detected, a target image classification model corresponding to the type of the blank to be detected from a plurality of preset image classification models; the acquisition unit is specifically used to: input the cross-sectional image into the target image classification model, and output the classification result of the cross-sectional image through the target image classification model. 6 . The device according to claim 4 , wherein the type of the blank to be inspected comprises at least one of the following: plate-shaped, rod-shaped, tube-shaped or wire-shaped.

7. An electronic device comprising: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1 to 3.

8. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method according to any one of claims 1 to 3.

9. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 3.

Images