A method, system, device and medium for detecting defects in a swage

Abstract

The application discloses a special-shaped forging defect detection method, system, device and medium, relates to the forging deformation detection technical field, and includes the following steps: obtaining the tapping temperature of the target forging and the real-time temperature at the current moment to obtain the temperature change rate of the target forging; obtaining the tapping image of the target forging and the measured image at the current moment, comparing the tapping image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point; obtaining the finite element simulation model of the target forging to obtain the internal residual stress of the target region of the target forging at the current moment; inputting the temperature change rate, the displacement change rate and the internal residual stress into a preset deformation rate prediction model to obtain a deformation rate prediction value; judging whether the deformation rate prediction value is greater than a preset deformation threshold value, if yes, identifying the target forging as a defective product, and the application has the advantages that the deformation rate can be accurately predicted during the cooling process of the forging.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of forging deformation detection, in particular to a special-shaped forging defect detection method, system, device and medium. BACKGROUND

[0002] During the stage of cooling from high temperature to low temperature after the forging is discharged, the forging will be deformed to different degrees due to the influence of temperature change and uneven material shrinkage stress. If the deformation is too large, the forging needs to enter the high-temperature forging process again to improve the uniformity of material shrinkage stress. At present, the deformation detection of the forging after being discharged, especially for special-shaped forgings with complex structures, is affected by the complex structure. The deformation cannot be accurately detected until the forging is completely cooled. It is difficult to accurately predict the deformation rate of the special-shaped forging during the cooling process, so it is difficult to enter the secondary high-temperature forging process in advance, which affects the production efficiency of the forging. SUMMARY

[0003] The main purpose of the present application is to provide a special-shaped forging defect detection method, system, device and medium, which aims to solve the technical problem that the existing forging deformation detection method is difficult to accurately predict the deformation rate during the forging cooling process.

[0004] To achieve the above purpose, the present application provides a special-shaped forging defect detection method, comprising the following steps:

[0005] Obtain the discharge temperature of the target forging and the real-time temperature at the current time to obtain the temperature change rate of the target forging;

[0006] Obtain the discharge image and the measured image of the target forging at the current time, and compare the discharge image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point;

[0007] Obtain the finite element simulation model of the target forging, and obtain the internal residual stress of the target region of the target forging at the current time according to the finite element simulation model; wherein the target region is the region where the target feature point is located in the target forging;

[0008] Input the temperature change rate, the displacement change rate and the internal residual stress into a preset deformation rate prediction model to obtain a deformation rate prediction value;

[0009] Determine whether the deformation rate prediction value is greater than a preset deformation threshold value. If yes, the target forging is identified as a defective product, and if no, the target forging is identified as a qualified product.

[0010] Optionally, the expression of the deformation rate prediction model is:

[0011] Q = a K1 DT / Dt + b K2 DD / Dt + g K3 s;

[0012] In the formula, Q is a deformation rate prediction value, ΔT is a temperature change amount of the target forging, ΔT = discharge temperature - real-time temperature, Δt is a time difference between the current time and the discharge time of the target forging, Δd is a displacement change amount of the target feature point at the current time, σ is an internal residual stress of the target region of the target forging at the current time, α is a first weight coefficient, β is a second weight coefficient, γ is a third weight coefficient, K1 is a first adjustment coefficient, K2 is a second adjustment coefficient, and K3 is a third adjustment coefficient.

[0013] Optionally, the first weight coefficient α, the second weight coefficient β, and the third weight coefficient γ satisfy the following conditions:

[0014] If the real-time temperature of the target forging at the current time is in the first temperature interval, then α > β > γ;

[0015] If the real-time temperature of the target forging at the current time is in the second temperature interval, then β > α ≥ γ or β > γ ≥ α;

[0016] If the real-time temperature of the target forging at the current time is in the third temperature interval, then γ > β > α;

[0017] The first temperature interval > the second temperature interval > the third temperature interval.

[0018] Optionally, the target forging includes a main plate, and the main plate has a thin-walled region.

[0019] The discharge image and the real-time image of the target forging at the current time are obtained, and the discharge image and the real-time image are compared based on the same target feature point to obtain a displacement change rate of the target feature point, including:

[0020] The discharge depth image of the target forging is obtained, and an initial depth value h0 of the target feature point is obtained according to the discharge depth image; wherein the target feature point is located in the thin-walled region.

[0021] The real-time depth image of the target forging at the current time is obtained, and a current depth value h of the target feature point is obtained according to the real-time depth image.

[0022] The initial depth value h0 and the current depth value h are used to obtain a displacement change amount Δd of the target feature point to obtain the displacement change rate; wherein Δd = |h - h0|, and the displacement change rate = Δd / Δt.

[0023] Optionally, if the thickness of the thin-walled region is uniform, at least one target feature point is selected at the center position of the thin-walled region, and if the thickness of the thin-walled region is not uniform, at least one target feature point is selected at the position with the smallest thickness of the thin-walled region.

[0024] Optionally, the target forging includes a main plate, and two support arms are connected to the side wall of the main plate.

[0025] The out-of-furnace image and the measured image at the current time of the target forging are obtained, and the out-of-furnace image and the measured image are compared based on the same target feature point to obtain a displacement change rate of the target feature point, including:

[0026] An out-of-furnace point cloud image of the target forging is obtained, and an initial interval d0 between two target feature points is obtained according to the out-of-furnace point cloud image; wherein the two target feature points are located at adjacent corner points of the two arms;

[0027] A measured point cloud image of the target forging at the current time is obtained, and a current interval d between the two target feature points is obtained according to the measured point cloud image;

[0028] The displacement change amount Δd of the target feature point is obtained according to the initial interval d0 and the current interval d to obtain the displacement change rate; wherein Δd = |d-d0|, and the displacement change rate = Δd / Δt.

[0029] To achieve the above object, the application further provides a special-shaped forging defect detection system, comprising:

[0030] A temperature detection module is configured to obtain an out-of-furnace temperature of the target forging and a real-time temperature at the current time to obtain a temperature change rate of the target forging;

[0031] A feature detection module is configured to obtain an out-of-furnace image of the target forging and a measured image at the current time, and compare the out-of-furnace image and the measured image based on the same target feature point to obtain a displacement change rate of the target feature point;

[0032] A stress detection module is configured to obtain a finite element simulation model of the target forging, and obtain an internal residual stress of a target region of the target forging at the current time according to the finite element simulation model; wherein the target region is a region where the target feature point is located in the target forging;

[0033] A prediction module is configured to input the temperature change rate, the displacement change rate and the internal residual stress into a preset deformation rate prediction model to obtain a deformation rate prediction value;

[0034] An identification module is configured to determine whether the deformation rate prediction value is greater than a preset deformation threshold value, if yes, the target forging is identified as a defective product, and if not, the target forging is identified as a qualified product.

[0035] Optionally, the expression of the deformation rate prediction model is:

[0036] Q = a K1 DT / DT + b K2 DD / DT + g K3 s;

[0037] In the formula, Q is a deformation rate prediction value, AT is a temperature change amount of the target forging, AT = discharge temperature - real-time temperature, At is a time difference between a current time and a discharge time of the target forging, Ad is a displacement change amount of a target feature point at the current time, s is an internal residual stress of a target region of the target forging at the current time, a is a first weight coefficient, b is a second weight coefficient, g is a third weight coefficient, K1 is a first adjustment coefficient, K2 is a second adjustment coefficient, and K3 is a third adjustment coefficient.

[0038] Optionally, the first weight coefficient a, the second weight coefficient b, and the third weight coefficient g satisfy the following conditions:

[0039] If the real-time temperature of the target forging at the current time is in the first temperature interval, then a > b > g;

[0040] If the real-time temperature of the target forging at the current time is in the second temperature interval, then b > a > g or b > g > a;

[0041] If the real-time temperature of the target forging at the current time is in the third temperature interval, then g > b > a;

[0042] The first temperature interval > the second temperature interval > the third temperature interval.

[0043] Optionally, the target forging includes a main plate, and the main plate has a thin-walled region; and the feature detection module is specifically configured to:

[0044] obtain a discharge depth image of the target forging, obtain an initial depth value h0 of a target feature point according to the discharge depth image, wherein the target feature point is located in the thin-walled region; obtain a real-time depth image of the target forging at the current time, obtain a current depth value h of the target feature point according to the real-time depth image, and obtain a displacement change amount Ad of the target feature point according to the initial depth value h0 and the current depth value h, so as to obtain a displacement change rate; wherein Ad = |h - h0|, and the displacement change rate = Ad / At.

[0045] Optionally, the target forging includes a main plate, and two arms are connected to a side wall of the main plate; and the feature detection module is specifically configured to:

[0046] obtain a discharge point cloud image of the target forging, obtain an initial interval d0 between two target feature points according to the discharge point cloud image, wherein the two target feature points are located at adjacent corner points of the two arms; obtain a real-time point cloud image of the target forging at the current time, obtain a current interval d between the two target feature points according to the real-time point cloud image, and obtain a displacement change amount Ad of the target feature point according to the initial interval d0 and the current interval d, so as to obtain a displacement change rate; wherein Ad = |d - d0|, and the displacement change rate = Ad / At.

[0047] To achieve the above object, the application further provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the method.

[0048] To achieve the above object, the application further provides a computer readable storage medium, which stores a computer program, and the processor executes the computer program to realize the method.

[0049] The application can achieve the following beneficial effects:

[0050] The application can obtain the temperature change rate of the target forging by monitoring the out-of-furnace temperature and the real-time temperature at the current moment, the temperature change rate can represent the thermal shrinkage effect of the forging, and at the same time, the machine vision recognition technology is combined, the out-of-furnace image and the measured image at the current moment are compared based on the same target feature point, so as to obtain the displacement change rate of the target feature point, the displacement change rate can directly reflect the geometric deformation of the forging, and the internal residual stress of the target region of the target forging at the current moment can be obtained based on the finite element simulation model of the target forging, so as to describe the secondary deformation caused by the release of the internal residual stress, and finally the temperature change rate, the displacement change rate and the internal residual stress are input into the preset deformation rate prediction model to obtain the deformation rate prediction value, so that the application considers the influence of the temperature change rate, the displacement change rate and the internal residual stress on the deformation of the forging, and any influence factor exists obvious abnormality or the three influence factors exist abnormality to a certain extent, which will cause the deformation rate prediction value to be obviously increased, so that the deformation rate prediction value of the forging can be comprehensively represented and quantitatively calculated by the above three comparative influencing factors, the accuracy of the deformation trend prediction of the forging is improved, then it is judged whether the deformation rate prediction value is greater than the preset deformation threshold value, if yes, the target forging is identified as a defective product, if not, the target forging is identified as a qualified product, so that the deformation rate of the forging can be accurately predicted during the cooling process of the forging, and detection can be performed after the forging is completely cooled, if the forging is detected as a defective product during the cooling stage, the forging can immediately enter the secondary high-temperature forging process, at this time, the forging itself still exists a certain temperature, the heating efficiency is improved, and the manufacturing and production efficiency of the forging is improved. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the specific embodiments of the application or the technical solutions in the prior art, the drawings needed to be used in the specific embodiments or the prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0052] Fig. 1 FIG. 1 is a flowchart of a method for detecting defects in a special-shaped forging according to an embodiment of the present application;

[0053] Fig. 2 FIG. 2 is a schematic diagram of a target forging according to an embodiment of the present application;

[0054] Fig. 3 FIG. 3 is a schematic diagram of a system for detecting defects in a special-shaped forging according to an embodiment of the present application.

[0055] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0057] It should be noted that if the present application has a description of "first", "second", etc., the description of "first", "second", etc. is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the fact that a person skilled in the art can realize it. When the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, and is not within the scope of protection claimed by the present application.

[0058] Embodiment 1

[0059] Referring to Figs. 1-2 The present embodiment provides a method for detecting defects in a special-shaped forging, comprising the following steps:

[0060] Obtaining the temperature change rate of the target forging by obtaining the temperature at which the target forging is discharged from the furnace and the real-time temperature of the target forging at the current time;

[0061] Obtaining the temperature change rate of the target forging by obtaining the temperature at which the target forging is discharged from the furnace and the real-time temperature of the target forging at the current time;

[0062] Obtaining the temperature change rate of the target forging by obtaining the temperature at which the target forging is discharged from the furnace and the real-time temperature of the target forging at the current time;

[0063] inputting the temperature change rate, the displacement change rate and the internal residual stress into a preset deformation rate prediction model to obtain a deformation rate prediction value;

[0064] judging whether the deformation rate prediction value is greater than a preset deformation threshold, if yes, identifying the target forging as a defective product, and if no, identifying the target forging as a qualified product.

[0065] In the embodiment, by monitoring the out-of-furnace temperature and the real-time temperature at the current moment of the target forging, the temperature change rate of the target forging can be obtained, which can represent the thermal shrinkage effect of the forging. Meanwhile, combined with the machine vision recognition technology, the out-of-furnace image and the measured image at the current moment are compared based on the same target feature point based on the obtained out-of-furnace image and measured image of the target forging, so that the displacement change rate of the target feature point can be obtained, which can directly reflect the geometric deformation of the forging. The internal residual stress of the target region of the target forging at the current moment can be obtained based on the finite element simulation model of the target forging, so that the secondary deformation caused by the release of the internal residual stress can be described. Finally, the temperature change rate, the displacement change rate and the internal residual stress are input into a preset deformation rate prediction model to obtain a deformation rate prediction value. Therefore, the embodiment takes into account the influence of the temperature change rate, the displacement change rate and the internal residual stress on the deformation of the forging. If any of the influence factors is obviously abnormal or all of the three influence factors are synchronously abnormal to a certain extent, the deformation rate prediction value will be obviously increased. Therefore, the deformation rate prediction value of the forging can be comprehensively represented and quantitatively calculated by the above three comparative influence factors, thereby improving the accuracy of the prediction of the deformation trend of the forging. Then, it is judged whether the deformation rate prediction value is greater than a preset deformation threshold, if yes, the target forging is identified as a defective product, and if no, the target forging is identified as a qualified product. Thus, the deformation rate of the forging can be accurately predicted during the cooling process of the forging, without the need to detect the forging after it is completely cooled. If the forging is detected as a defective product during the cooling stage, the forging can immediately enter the secondary high-temperature forging process. At this time, the forging itself still has a certain temperature, thereby improving the heating efficiency and the manufacturing and production efficiency of the forging.

[0066] As an optional implementation, the expression of the deformation rate prediction model is:

[0067] Q = a K1 AT / At + b K2 Ad / At + g K3 s;

[0068] In the formula, Q is a deformation rate prediction value, AT is a temperature change amount of the target forging, AT = discharge temperature - real-time temperature, At is a time difference between the current time and the discharge time of the target forging, Ad is a displacement change amount of the target feature point at the current time, s is an internal residual stress of the target region of the target forging at the current time, a is a first weight coefficient, b is a second weight coefficient, g is a third weight coefficient, K1 is a first adjustment coefficient, K2 is a second adjustment coefficient, and K3 is a third adjustment coefficient.

[0069] In the present embodiment, AT / At in the above formula is a temperature change rate, and Ad / At is a displacement change rate. Since the three influence factors have different parameter properties, the three influence factors are balanced and adjusted by the three adjustment coefficients K1, K2 and K3, so that the three influence factors can be superimposed. Since the three influence factors have different influences on the prediction value, the corresponding weight coefficients a, b and g are assigned. Therefore, based on the above formula, if any of the influence factors has an abnormally increased parameter, or if two or three of the influence factors have a small increase, the final deformation rate prediction value Q will increase, thereby characterizing the influence of a single influence factor on the deformation of the forging, and also characterizing the synergistic effect of multiple influence factors on the deformation of the forging. The calculation result is accurate and reliable, and has good reference value and guidance.

[0070] As an optional embodiment, the first weight coefficient a, the second weight coefficient b and the third weight coefficient g satisfy the following conditions:

[0071] If the real-time temperature of the target forging at the current time is in the first temperature interval, then a > b > g;

[0072] If the real-time temperature of the target forging at the current time is in the second temperature interval, then b > a > g or b > g > a;

[0073] If the real-time temperature of the target forging at the current time is in the third temperature interval, then g > b > a;

[0074] Wherein, the first temperature interval > the second temperature interval > the third temperature interval.

[0075] In the embodiment, since the forging cools down, the three factors above dominate the degree of deformation of the forging at different temperature stages. At a high temperature stage, i.e., a first temperature interval (for example, > 800℃), the temperature term (i.e., a·K1·ΔT / Δt) has a relatively obvious effect on the thermal contraction of the forging, i.e., dominates the deformation of the forging. At this time, the relationship between the weight coefficients is set as a> b> g (for example, a = 0.5, b = 0.3, g = 0.2), so as to highlight the influence of the temperature term on the deformation rate of the forging. At an intermediate temperature stage, i.e., a second temperature interval (for example, 200℃-800℃), the geometric deformation of the forging is more obvious at this time, so the displacement term (i.e., b·K2·Δd / Δt) dominates at this time. Therefore, the relationship between the weight coefficients is set as b> a≥ g or b> g≥ a (for example, b = 0.5, a = 0.3, g = 0.2 or b = 0.5, g = 0.3, a = 0.2 or b = 0.5, g = 0.25, a = 0.25), so as to highlight the influence of the displacement term on the deformation rate of the forging. At a low temperature stage, i.e., a third temperature interval (for example, 100℃-200℃), when the temperature drops below 200℃, the forging material enters an elastic dominant stage, and local plastic deformation caused by the release of residual stress is more likely to cause macro secondary deformation. At this time, the stress term (i.e., g·K3·σ) has a more obvious effect on the deformation of the forging. Therefore, the relationship between the weight coefficients is set as g> b> a (for example, g = 0.5, b = 0.3, a = 0.2), so as to highlight the influence of the stress term on the deformation rate of the forging. In summary, the embodiment can flexibly adjust the weight coefficients of the parameter terms when the forging cools down to different temperature stages, thereby further improving the accuracy of the calculation results.

[0076] It should be noted that if the forging cools down to a preset temperature (for example, 100℃) and the deformation rate prediction value Q is still not greater than the deformation threshold, the detection is stopped and the current forging is directly identified as a qualified product.

[0077] As an optional implementation, the target forging includes a main plate, and the main plate has a thin-walled area;

[0078] An out-of-furnace image and a measured image of the target forging at a current time are obtained, and the out-of-furnace image and the measured image are compared based on the same target feature point to obtain a displacement change rate of the target feature point, including:

[0079] An out-of-furnace depth image of the target forging is obtained, and an initial depth value h0 of the target feature point is obtained according to the out-of-furnace depth image; wherein the target feature point is located in the thin-walled area;

[0080] A measured depth image of the target forging at a current time is obtained, and a current depth value h of the target feature point is obtained according to the measured depth image;

[0081] According to the initial depth value h0 and the current depth value h, the displacement change amount Δd of the target feature point is obtained to obtain the displacement change rate; wherein Δd = |h-h0|, and the displacement change rate = Δd / Δt.

[0082] In the embodiment, for some complex profiled forgings, the thin-wall region of the main part is prone to deformation, so the target feature point can be collected from the thin-wall region, and the deformation trend of the forging can be detected. Because the thin-wall region is surrounded by regions with larger thickness, the deformation direction of the thin-wall region tends to be in the thickness direction. Therefore, by collecting the depth image of the target forging when it is discharged from the furnace and the measured depth image at the current time, and using the existing feature point tracking algorithm, the initial depth value h0 of the target feature point corresponding to the depth image when it is discharged from the furnace and the current depth value h corresponding to the measured depth image can be obtained, so that the displacement change amount Δd of the target feature point can be calculated, and finally the displacement change rate = Δd / Δt can be calculated. Therefore, by selecting the target feature point in the thin-wall region and identifying the depth change of the target feature point through the depth image, the displacement change rate of the target feature point can be effectively represented, and accurate data basis is provided for the final calculation result.

[0083] It should be noted that a plurality of target feature points can be selected in the thin-wall region for simultaneous calculation. The maximum value of the plurality of displacement change rates calculated is output as the final calculation value, so as to avoid data distortion caused by the contingency of a single calculation value and improve the calculation accuracy.

[0084] As an optional implementation, if the thickness of the thin-wall region is uniform, at least one target feature point is selected at the center position of the thin-wall region. If the thickness of the thin-wall region is not uniform, at least one target feature point is selected at the position with the smallest thickness in the thin-wall region.

[0085] In the embodiment, when selecting the target feature point, if the thickness of the thin-wall region is uniform, the center position is the most easily deformed position, so at least one target feature point is selected at the center position of the thin-wall region. Meanwhile, a plurality of target feature points can also be selected around the center position. If the thickness of the thin-wall region is not uniform, the position with the smallest thickness is the most easily deformed position, so at least one target feature point is selected at the position with the smallest thickness. Meanwhile, a plurality of target feature points can also be selected around the position with the smallest thickness, so as to improve the diversity of sample data collection and improve the calculation accuracy.

[0086] As an optional implementation, the target forging includes a main plate, and two branch arms are connected to the side wall of the main plate.

[0087] The discharge image of the target forging and the measured image at the current time are obtained, and the discharge image and the measured image are compared based on the same target feature point to obtain the displacement change rate of the target feature point, including:

[0088] acquire an initial interval d0 between the two target feature points according to the tapping point cloud image; wherein the two target feature points are located at adjacent corner points of the two arms;

[0089] acquire a current interval d between the two target feature points according to the measured point cloud image at the current time;

[0090] acquire a displacement change amount Ad of the target feature point according to the initial interval d0 and the current interval d, so as to obtain a displacement change rate; wherein Ad = |d-d0|, and the displacement change rate = Ad / At.

[0091] In the embodiment, in some special-shaped forgings, the side end of the main body part has two arms, the connection position of the arms with the main body part is relatively weak, and is prone to obvious deformation under stress, and the interval between the two arm ends will change obviously when deformation occurs. Therefore, the adjacent corner points of the two arms are simultaneously taken as target feature points, and then based on the tapping point cloud image of the target forging, the initial interval d0 between the two target feature points can be acquired, and then based on the measured point cloud image of the target forging at the current time, the current interval d between the two target feature points can be acquired, so that the displacement change amount Ad of the target feature point can be calculated, and finally the displacement change rate = Ad / At can be calculated. Therefore, by taking the adjacent corner points of the two arms as target feature points, the actual intervals of the two target feature points at different times can be identified by using the point cloud image, so that the displacement change rate of the target feature point can be effectively represented, and accurate data basis is provided for the final calculation result.

[0092] Embodiment 2

[0093] Based on the same inventive idea as the foregoing embodiments, referring to Figs. 2-3 , the embodiment also provides a special-shaped forging defect detection system, comprising:

[0094] a temperature detection module, configured to acquire a tapping temperature of the target forging and a real-time temperature at the current time, so as to obtain a temperature change rate of the target forging;

[0095] a feature detection module, configured to acquire a tapping image of the target forging and a measured image at the current time, and compare the tapping image and the measured image based on the same target feature point, so as to obtain a displacement change rate of the target feature point;

[0096] a stress detection module, configured to acquire a finite element simulation model of the target forging, and acquire an internal residual stress of a target region of the target forging at the current time according to the finite element simulation model; wherein the target region is a region where the target feature point is located in the target forging;

[0097] a prediction module configured to input the temperature change rate, the displacement change rate and the internal residual stress into a preset deformation rate prediction model to obtain a deformation rate prediction value;

[0098] a recognition module configured to determine whether the deformation rate prediction value is greater than a preset deformation threshold value, and if yes, recognize the target forging as a defective product, and if not, recognize the target forging as a qualified product.

[0099] As an optional implementation, the expression of the deformation rate prediction model is:

[0100] Q = a K1 AT / At + b K2 Ad / At + g K3 s;

[0101] wherein, Q is the deformation rate prediction value, AT is the temperature change amount of the target forging, AT = discharge temperature - real-time temperature, At is the time difference between the current time and the discharge time of the target forging, Ad is the displacement change amount of the target feature point at the current time, s is the internal residual stress of the target region of the target forging at the current time, a is the first weight coefficient, b is the second weight coefficient, g is the third weight coefficient, K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient.

[0102] As an optional implementation, the first weight coefficient a, the second weight coefficient b and the third weight coefficient g satisfy the following conditions:

[0103] If the real-time temperature of the target forging at the current time is in the first temperature interval, then a > b > g;

[0104] If the real-time temperature of the target forging at the current time is in the second temperature interval, then b > a > g or b > g > a;

[0105] If the real-time temperature of the target forging at the current time is in the third temperature interval, then g > b > a;

[0106] Wherein, the first temperature interval > the second temperature interval > the third temperature interval.

[0107] As an optional implementation, the target forging includes a main plate, and the main plate has a thin-walled region; the feature detection module is specifically configured to:

[0108] obtain a discharge depth image of the target forging, and obtain an initial depth value h0 of the target feature point according to the discharge depth image; wherein the target feature point is located in the thin-walled region; obtain a real-time depth image of the target forging at the current time, and obtain a current depth value h of the target feature point according to the real-time depth image; and obtain a displacement change amount Ad of the target feature point according to the initial depth value h0 and the current depth value h, to obtain the displacement change rate; wherein Ad = |h - h0|, and the displacement change rate = Ad / At.

[0109] As an optional implementation, the target forging includes a main plate, two branch arms are connected to the side wall of the main plate; and the feature detection module is specifically used for:

[0110] An out-of-furnace point cloud image of the target forging is acquired, and an initial interval d0 between two target feature points is acquired according to the out-of-furnace point cloud image; the two target feature points are located at adjacent corner points of the two branch arms; a real-time point cloud image of the target forging at a current time is acquired, and a current interval d between the two target feature points is acquired according to the real-time point cloud image; a displacement change amount Δd of the target feature points is acquired according to the initial interval d0 and the current interval d, so as to obtain a displacement change rate; wherein Δd = |d-d0|, and the displacement change rate = Δd / Δt.

[0111] The related explanations and examples of the modules in the system of the embodiment can refer to the method of the foregoing embodiment, and will not be described here.

[0112] Embodiment 3

[0113] Based on the same inventive idea as the foregoing embodiments, the embodiment provides a computer device, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the method.

[0114] Embodiment 4

[0115] Based on the same inventive idea as the foregoing embodiments, the embodiment provides a computer readable storage medium, which stores a computer program, and a processor executes the computer program to realize the method.

[0116] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation according to the content of the specification and the drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for detecting defects in irregularly shaped forgings, characterized in that, Includes the following steps: Obtain the furnace exit temperature and the real-time temperature of the target forging at the current moment to obtain the temperature change rate of the target forging; Acquire the furnace exit image and the measured image of the target forging at the current moment, and compare the furnace exit image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point; Obtain the finite element simulation model of the target forging, and obtain the internal residual stress of the target region of the target forging at the current moment based on the finite element simulation model; where the target region is the region where the target feature point is located in the target forging; The temperature change rate, displacement change rate, and internal residual stress are input into a preset deformation rate prediction model to obtain the predicted deformation rate value; the expression of the deformation rate prediction model is: Q=α·K1·ΔT / Δt+β·K2·Δd / Δt+γ·K3·σ; In the formula, Q is the predicted deformation rate, ΔT is the temperature change of the target forging, ΔT = furnace exit temperature - real-time temperature, Δt is the time difference between the current moment and the furnace exit moment of the target forging, Δd is the displacement change of the target feature point at the current moment, σ is the internal residual stress of the target area of ​​the target forging at the current moment, α is the first weighting coefficient, β is the second weighting coefficient, γ is the third weighting coefficient, K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient. Determine whether the predicted deformation rate is greater than the preset deformation threshold. If so, identify the target forging as a defective product; otherwise, identify the target forging as a qualified product.

2. The method for detecting defects in irregularly shaped forgings as described in claim 1, characterized in that, The first weighting coefficient α, the second weighting coefficient β, and the third weighting coefficient γ satisfy the following condition: If the real-time temperature of the target forging is in the first temperature range at the current moment, then α > β > γ. If the real-time temperature of the target forging is in the second temperature range at the current moment, then β>α≥γ or β>γ≥α; If the real-time temperature of the target forging is in the third temperature range at the current moment, then γ > β > α. Among them, the first temperature range > the second temperature range > the third temperature range.

3. The method for detecting defects in irregularly shaped forgings as described in claim 1, characterized in that, The target forging includes a main sheet with thin-walled areas. Acquire the furnace exit image and the measured image of the target forging at the current moment. Compare the furnace exit image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point, including: Obtain the furnace exit depth image of the target forging, and obtain the initial depth value h0 of the target feature point based on the furnace exit depth image; wherein the target feature point is located in the thin-walled region; Obtain the measured depth image of the target forging at the current moment, and obtain the current depth value h of the target feature point based on the measured depth image; Based on the initial depth value h0 and the current depth value h, the displacement change Δd of the target feature point is obtained to obtain the displacement change rate; where Δd=|h-h0|, and the displacement change rate=Δd / Δt.

4. The method for detecting defects in irregularly shaped forgings as described in claim 3, characterized in that, If the thickness of the thin-walled region is uniform, at least one target feature point should be selected at the center of the thin-walled region. If the thickness of the thin-walled region is not uniform, at least one target feature point should be selected at the position where the thickness of the thin-walled region is the minimum.

5. The method for detecting defects in irregularly shaped forgings as described in claim 1, characterized in that, The target forging includes a main board, with two support arms connected to the side wall of the main board; Acquire the furnace exit image and the measured image of the target forging at the current moment. Compare the furnace exit image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point, including: Obtain the point cloud image of the target forging after it exits the furnace. Based on the point cloud image, obtain the initial distance d0 between two target feature points. The two target feature points are located at adjacent corner points of two support arms. Obtain the measured point cloud image of the target forging at the current moment, and based on the measured point cloud image, obtain the current distance d between two target feature points; Based on the initial spacing d0 and the current spacing d, the displacement change Δd of the target feature point is obtained to obtain the displacement change rate; where Δd=|d-d0|, and the displacement change rate=Δd / Δt.

6. A defect detection system for irregularly shaped forgings, characterized in that, include: The temperature detection module is used to obtain the furnace exit temperature and the real-time temperature of the target forging at the current moment, so as to obtain the temperature change rate of the target forging. The feature detection module is used to acquire the furnace exit image and the measured image of the target forging at the current moment, and compare the furnace exit image and the measured image based on the same target feature point to obtain the displacement change rate of the target feature point. The stress detection module is used to obtain the finite element simulation model of the target forging and to obtain the internal residual stress of the target area of ​​the target forging at the current moment based on the finite element simulation model; where the target area is the area where the target feature point is located in the target forging. The prediction module is used to input the temperature change rate, displacement change rate, and internal residual stress into a preset deformation rate prediction model to obtain the predicted deformation rate value; the expression of the deformation rate prediction model is: Q=α·K1·ΔT / Δt+β·K2·Δd / Δt+γ·K3·σ; In the formula, Q is the predicted deformation rate, ΔT is the temperature change of the target forging, ΔT = furnace exit temperature - real-time temperature, Δt is the time difference between the current moment and the furnace exit moment of the target forging, Δd is the displacement change of the target feature point at the current moment, σ is the internal residual stress of the target area of ​​the target forging at the current moment, α is the first weighting coefficient, β is the second weighting coefficient, γ is the third weighting coefficient, K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient. The identification module is used to determine whether the predicted deformation rate is greater than the preset deformation threshold. If it is, the target forging is identified as a defective product; otherwise, the target forging is identified as a qualified product.

7. The defect detection system for irregularly shaped forgings as described in claim 6, characterized in that, The first weighting coefficient α, the second weighting coefficient β, and the third weighting coefficient γ satisfy the following condition: If the real-time temperature of the target forging is in the first temperature range at the current moment, then α > β > γ. If the real-time temperature of the target forging is in the second temperature range at the current moment, then β>α≥γ or β>γ≥α; If the real-time temperature of the target forging is in the third temperature range at the current moment, then γ > β > α. Among them, the first temperature range > the second temperature range > the third temperature range.

8. The defect detection system for irregularly shaped forgings as described in claim 6, characterized in that, The target forging includes a main sheet with thin-walled areas. The feature detection module is specifically used for: Obtain the furnace exit depth image of the target forging, and based on the furnace exit depth image, obtain the initial depth value h0 of the target feature point; wherein the target feature point is located in the thin-walled region; obtain the measured depth image of the target forging at the current moment, and based on the measured depth image, obtain the current depth value h of the target feature point; based on the initial depth value h0 and the current depth value h, obtain the displacement change Δd of the target feature point to obtain the displacement change rate; wherein Δd=|h-h0|, and the displacement change rate=Δd / Δt.

9. The defect detection system for irregularly shaped forgings as described in claim 6, characterized in that, The target forging includes a main board, with two support arms connected to the side wall of the main board; The feature detection module is specifically used for: Acquire the point cloud image of the target forging after it exits the furnace. Based on the point cloud image, obtain the initial distance d0 between two target feature points. The two target feature points are located at adjacent corner points of two supports. Acquire the measured point cloud image of the target forging at the current moment. Based on the measured point cloud image, obtain the current distance d between the two target feature points. Based on the initial distance d0 and the current distance d, obtain the displacement change Δd of the target feature points to obtain the displacement change rate. Wherein, Δd=|d-d0|, and the displacement change rate=Δd / Δt.

10. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores a computer program and the processor executes the computer program to implement the method as described in any one of claims 1-5.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the method as described in any one of claims 1-5.

Images