Shrinkage defect prediction method and system for side plate casting

Abstract

The present invention relates to the technical field of shrinkage defect prediction, and more specifically to a shrinkage defect prediction method and system for side plate casting. The method comprises: obtaining a temperature drop stability value and a color distribution unevenness value by using the temperature of the thin-walled area in an infrared image, thereby obtaining a shrinkage defect performance evaluation value and determining the shrinkage defect area; obtaining the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, obtaining a bubble-causing index value based on the temperature-time curve and determining the bubble shrinkage defect area; obtaining the coordinates of the bubbles to be corrected at the next moment based on the bubble movement trend curve; obtaining the metal liquid flow velocity of the bubble shrinkage defect area in each infrared image, thereby obtaining a metal liquid flow evaluation value; and correcting the bubble coordinates to be corrected based on the metal liquid flow evaluation value to obtain the corrected bubble coordinates. The present invention enables accurate prediction of the bubble position at the next moment.

Description

Technical Field

[0001] The present invention relates to the technical field of shrinkage defect prediction, and in particular to a shrinkage defect prediction method and system for side plate casting. Background Art

[0002] Side panel castings, as a common and critical casting component, are widely used in various fields such as automotive manufacturing, shipbuilding, and engineering machinery, resulting in extremely stringent quality requirements. Shrinkage defects, a frequent quality issue in the casting process, are characterized by the formation of pores and holes within or on the surface of the casting. This phenomenon typically manifests during the solidification phase of the molten metal. Failure of the molten metal to fully fill the mold cavity or excessive temperature gradients during cooling lead to uneven shrinkage of the metal. This is particularly prevalent in thin-walled areas, where significant differences in heat conduction and slow temperature recovery can easily induce shrinkage defects. The complex geometry of castings often leads to uneven wall thickness. Thin-walled areas are particularly prone to shrinkage defects due to delayed molten metal replenishment and more pronounced shrinkage. Furthermore, during the casting process, gas release or the escape of dissolved gases from the molten metal can form bubbles. These bubbles can not only alter the fluidity of the liquid metal, especially in thin-walled areas of the casting, but can also interfere with the metal filling process, causing localized metal underfilling. This further promotes the formation of thin-wall defects or other types of defects, and may exacerbate the severity of defects in these areas. The challenge faced by current technology is that it is difficult to accurately distinguish shrinkage defects caused by the inherent characteristics of the thin-walled area of the casting from similar defects caused by bubbles, which in turn affects the accuracy of shrinkage defect prediction during the casting process of side plate castings. Summary of the Invention

[0003] In order to solve the technical problem of inaccurate prediction of shrinkage defects during the casting process of side plate castings, the purpose of the present invention is to provide a method for predicting shrinkage defects in side plate castings. The technical solution adopted is as follows:

[0004] Acquire an infrared image of a preset first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared image;

[0005] Obtaining a temperature drop stability value and a color distribution unevenness value by using the temperature of the thin-walled area, obtaining a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value, and determining the thin-walled area as a shrinkage defect area when the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to each pixel in the thin-walled area;

[0006] Acquiring the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, and obtaining a bubble-causing index value of the shrinkage defect area based on the temperature-time curve, and determining that the shrinkage defect area is a bubble shrinkage defect area when the bubble-causing index value is not less than a preset bubble-causing threshold value;

[0007] A rectangular coordinate system is established with the center point of the infrared image as the coordinate origin, the center point of the bubble shrinkage defect area is used as the bubble coordinate, a bubble movement trend curve is constructed, and the bubble coordinates to be corrected at the next moment are obtained according to the bubble movement trend curve;

[0008] Obtaining a metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtaining a metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity;

[0009] The bubble coordinates to be corrected are corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates.

[0010] Furthermore, the distinguishing between thick-walled areas and thin-walled areas in the infrared image includes:

[0011] Dividing the infrared image into a second number of regions, obtaining a first mean brightness of the i-th region and a second mean brightness of regions adjacent to the i-th region, wherein the value of i ranges from 1 to the second number;

[0012] Adding the first mean and the second mean to obtain a sum of the means, obtaining an absolute value of a difference between the first mean and the second mean, and dividing the difference by the sum of the means to obtain a spectral absorption evaluation value of the i-th region;

[0013] When the spectral absorption evaluation value is not less than a preset spectral absorption threshold, the i-th region is the thin-wall region; otherwise, the i-th region is the thick-wall region.

[0014] Furthermore, the process of obtaining the temperature drop stable value includes:

[0015] In each of the infrared images, obtaining the temperature of the thin-walled area;

[0016] Obtaining the absolute value of the temperature difference between the thin-walled area in two infrared images adjacent at a time, and obtaining an average value of the absolute values as an absolute average value;

[0017] An average value of the absolute value obtained by subtracting the absolute average value from the absolute value is calculated, and the reciprocal of the average value is taken to obtain the temperature drop stable value.

[0018] Furthermore, the process of obtaining the color distribution unevenness value includes:

[0019] Calculating the variance of the temperature corresponding to the pixel points in the thin-walled area as the variance of the thin-walled area;

[0020] The color distribution unevenness value is obtained by dividing the thin-wall region variance by the temperature of the thin-wall region.

[0021] Furthermore, the process of obtaining the shrinkage defect performance evaluation value includes:

[0022] The temperature drop stability value is multiplied by the normalized value of the color distribution unevenness value to obtain the shrinkage defect performance evaluation value.

[0023] Furthermore, the process of obtaining the bubble-induced index value includes:

[0024] Obtaining the temperature of the mth shrinkage defect region and the average of the temperatures of the shrinkage defect regions when the metal liquid begins to fill the mold, and dividing the temperature of the mth shrinkage defect region by the average of the temperatures of the shrinkage defect regions to obtain a temperature ratio;

[0025] Obtaining the curvature of each moment in the temperature-time curve corresponding to the m-th shrinkage defect region, and calculating an average value of the curvature;

[0026] Calculating an average of the absolute values of the differences between the curvature at each moment and the average value of the curvature as a curvature difference value;

[0027] The temperature ratio is divided by the curvature difference value and then normalized to obtain the bubble-inducing index value.

[0028] Furthermore, the process of obtaining the coordinates of the bubble to be corrected includes:

[0029] Obtaining the current bubble coordinates, differentiating the bubble coordinates to obtain a velocity vector of the bubble, and further differentiating the velocity vector to obtain an acceleration vector of the bubble;

[0030] After obtaining the velocity vector and the acceleration vector, the Euler method is used to predict the bubble coordinates to be corrected at the next moment.

[0031] Furthermore, the process of obtaining the metal liquid flow evaluation value includes:

[0032] Obtaining the metal liquid flow velocity of the hth bubble shrinkage defect area in each of the infrared images, and obtaining an average of the metal liquid flow velocities as the metal liquid flow velocity average;

[0033] Calculating an average of the absolute values of the differences between the metal liquid flow velocities and the metal liquid flow velocities mean as a metal liquid flow velocity difference value;

[0034] The metal liquid flow evaluation value is obtained by dividing the metal liquid flow velocity difference value by the metal liquid flow velocity average value.

[0035] Furthermore, the process of obtaining the corrected bubble coordinates includes:

[0036] The metal liquid flow evaluation value is multiplied by the bubble coordinate to be corrected to obtain the corrected bubble coordinate.

[0037] An embodiment of the present invention further provides a shrinkage defect prediction system for side plate casting, the system comprising:

[0038] an acquisition module, configured to acquire infrared images of a preset first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared images;

[0039] a shrinkage defect performance evaluation value module, configured to obtain a temperature drop stability value and a color distribution unevenness value based on the temperature of the thin-walled area, and obtain a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value; and when the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, the thin-walled area is designated as a shrinkage defect area, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to the pixels in the thin-walled area;

[0040] a bubble-induced index value module, configured to obtain the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, and obtain a bubble-induced index value of the shrinkage defect area based on the temperature-time curve; when the bubble-induced index value is not less than a preset bubble-induced threshold value, the shrinkage defect area is a bubble-induced shrinkage defect area;

[0041] A coordinate module to be corrected is used to establish a rectangular coordinate system with the center point of the infrared image as the coordinate origin, use the center point of the bubble shrinkage defect area as the bubble coordinate, construct a bubble movement trend curve, and obtain the bubble coordinates to be corrected at the next moment according to the bubble movement trend curve;

[0042] a metal liquid flow evaluation value module, configured to obtain a metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtain a metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity;

[0043] The correction module is used to correct the bubble coordinates to be corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates.

[0044] The present invention has the following beneficial effects:

[0045] First, infrared images of the first number of side plate castings during the casting process are acquired to distinguish between thick-walled and thin-walled areas. The infrared images are the basis for defect detection; only after distinguishing between thick and thin-walled areas can further inspection of thin-walled areas be performed.

[0046] Next, the temperature drop stability value and color distribution unevenness value are obtained from the temperature of the thin-walled area. A shrinkage defect performance evaluation value is obtained based on the temperature drop stability value and the color distribution unevenness value. When the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, the thin-walled area is identified as a shrinkage defect area. The temperature of the thin-walled area is the average of the corresponding temperatures of each pixel in the thin-walled area. This is to select the shrinkage defect area from the thin-walled area.

[0047] The temperature of the shrinkage defect region in the infrared image obtained at each moment is obtained to form a temperature-time curve. A bubble-induced index value of the shrinkage defect region is obtained based on the temperature-time curve. When the bubble-induced index value is not less than a preset bubble-induced threshold, the shrinkage defect region is identified as a bubble shrinkage defect region. This is to select the bubble shrinkage defect region from the shrinkage defect regions.

[0048] A rectangular coordinate system is established with the center point of the infrared image as the origin. The center point of the bubble shrinkage defect area is used as the bubble coordinate. A bubble movement trend curve is constructed. The bubble coordinates to be corrected at the next moment are obtained based on the bubble movement trend curve. The corrected bubble coordinates are the initial position of the bubble at the next moment and require correction.

[0049] The metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images is obtained, and a metal liquid flow evaluation value of the bubble shrinkage defect area is obtained based on the metal liquid flow velocity. The metal liquid flow evaluation value can be used as a correction weight to correct the bubble coordinates to be corrected.

[0050] The bubble coordinates to be corrected are corrected based on the metal liquid flow evaluation value to obtain corrected bubble coordinates. The corrected bubble coordinates accurately predict the bubble's position at the next moment, effectively controlling the location of defect formation, thereby optimizing the casting process, significantly reducing or even eliminating shrinkage areas, and preventing bubble accumulation in key areas, thereby significantly improving the overall quality of the casting. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 This is a flow chart of a method for predicting shrinkage defects in side plate castings provided by the first embodiment of the present invention;

[0053] Figure 2 A flow chart for distinguishing thick-walled areas and thin-walled areas in the infrared image provided by the second embodiment of the present invention;

[0054] Figure 3 A flow chart of a process for obtaining a temperature drop stability value provided by the third embodiment of the present invention;

[0055] Figure 4 A flowchart of a process for obtaining a color distribution unevenness value provided by a fourth embodiment of the present invention;

[0056] Figure 5 A flowchart of a process for obtaining an indicator value caused by bubbles provided in a fifth embodiment of the present invention;

[0057] Figure 6 A flowchart of a process for obtaining the coordinates of a bubble to be corrected provided in a sixth embodiment of the present invention;

[0058] Figure 7 A flowchart of a process for obtaining a metal liquid flow evaluation value provided by a seventh embodiment of the present invention;

[0059] Figure 8 This is a schematic diagram of a shrinkage defect prediction system for side plate casting provided by the eighth embodiment of the present invention. DETAILED DESCRIPTION

[0060] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of the method and system for predicting shrinkage defects in side plate castings according to the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0061] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0062] It should be noted that in order to ensure that the calculation results are meaningful, when performing fractional operations in the embodiments of the present invention, when encountering a situation where the denominator is 0, it is necessary to add a parameter adjustment factor greater than 0 to the denominator to prevent the denominator from being 0. The value of the parameter adjustment factor is set by the implementer according to actual conditions, and this application does not impose any special restrictions.

[0063] The specific scheme of the method for predicting shrinkage defects in side plate casting provided by the present invention is described in detail below with reference to the accompanying drawings.

[0064] See also Figure 1 , which shows a flow chart of a method for predicting shrinkage defects in side plate castings provided by a first embodiment of the present invention, the method comprising:

[0065] S101. Obtain infrared images of a preset first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared images.

[0066] Using an infrared thermal imager, infrared images of the side plate casting are captured every preset time, such as 1 second, to obtain the first several infrared images of the casting captured from the time when the metal is in liquid state to the current time.

[0067] Infrared images can capture critical moments in a casting's temperature evolution, such as during solidification, areas of uneven cooling, or moments of temperature differentials. Infrared images typically represent temperature in color (or grayscale), with hot areas appearing red or white and cold areas appearing blue or black.

[0068] The process of distinguishing the thick-walled area and the thin-walled area in the infrared image will be described in detail in the second embodiment and will not be repeated here.

[0069] S102. Obtain a temperature drop stability value and a color distribution unevenness value through the temperature of the thin-walled area, and obtain a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value. When the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, the thin-walled area is regarded as a shrinkage defect area, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to the pixels in the thin-walled area.

[0070] Shrinkage defects are common defects in side panel castings, where the molten metal shrinks during the solidification process and fails to completely fill the mold cavity, resulting in cavities or pores within or on the surface of the casting. The geometric structure of the casting results in inconsistent wall thickness. Thinner-walled areas, due to their lower mass and higher surface area, dissipate heat quickly, leading to relatively rapid cooling. This causes the molten metal in these areas to solidify earlier, increasing the temperature gradient. In contrast, thicker-walled areas, due to their higher mass and lower surface area, cool more slowly, dissipate less heat, and allow the molten metal to remain liquid until a later stage. This discrepancy can cause the thinner-walled areas to solidify prematurely, before the thicker ones have fully solidified. This asynchronous solidification process prevents the molten metal from replenishing the volume lost in the thinner-walled areas, resulting in cavities or pores. These cavities, commonly referred to as shrinkage defects, not only affect the mechanical properties of the casting but can also adversely affect the product's appearance and service life.

[0071] The process of obtaining the temperature drop stable value will be described in detail in the third embodiment and will not be repeated here.

[0072] The process of obtaining the color distribution unevenness value will be described in detail in the fourth embodiment and will not be repeated here.

[0073] The preset shrinkage defect performance evaluation threshold can be set independently, and is preferably 0.5.

[0074] The more unstable the temperature drop of the thin-walled area during the casting process is, and the more uneven the color distribution of the hot spots in the thin-walled area in the current infrared image is, the more likely the thin-walled area is to have shrinkage defects.

[0075] The process of obtaining the shrinkage defect performance evaluation value includes:

[0076] The temperature drop stability value is multiplied by the normalized value of the color distribution unevenness value to obtain the shrinkage defect performance evaluation value.

[0077] The shrinkage defect performance evaluation value can be expressed as:

[0078]

[0079] Among them, the Indicates the The shrinkage defect performance evaluation of the thin-walled area is Indicates the The temperature drop stability value corresponding to the thin-walled area, Indicates the The color distribution unevenness value corresponding to the thin-walled area, is a normalization function, preferably a range normalization function.

[0080] described The larger the value, the The thinner the wall area, the more likely it is that shrinkage defects will occur.

[0081] S103. Obtain the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, and obtain the bubble-causing index value of the shrinkage defect area according to the temperature-time curve. When the bubble-causing index value is not less than a preset bubble-causing threshold, the shrinkage defect area is a bubble shrinkage defect area.

[0082] During the casting process, bubbles are formed due to the release of gas or the escape of dissolved gases in the molten metal. The presence of bubbles may cause changes in the fluidity of the liquid metal, especially in thinner parts of the casting. Bubbles may affect the metal filling process, resulting in incomplete metal filling in some areas and exacerbating shrinkage defects in thin-walled areas.

[0083] In the liquid metal stage, bubbles form and the temperature in the bubble area is low, appearing as a darker cold area on the infrared image. During the metal liquid filling process, bubbles will flow with the liquid metal and gradually distribute themselves inside the casting. Heat conduction in the bubble area is blocked, and the temperature is low, appearing as a more obvious low-temperature area. As the metal gradually solidifies, the bubbles gradually disappear during the cooling process of the metal liquid, the temperature gradually rises, and the brightness of the area gradually increases, but this process is relatively slow. In contrast, the cooling process of normal shrinkage defect areas is more uniform and the temperature is lower, but since there are no bubbles to interfere, the heat conduction is smoother, appearing as a brighter cold area on the infrared image, the temperature rises faster, and the heat conduction is more uniform.

[0084] In general, shrinkage defect areas containing bubbles show slower temperature recovery and lower temperatures during the cooling process, while normal shrinkage defect areas have more uniform temperature recovery and faster cooling.

[0085] Select shrinkage defect area, obtain the first The average temperature of all pixels in the shrinkage defect area is the The temperature of the shrinkage defect area is taken as the horizontal axis, and the temperature of the shrinkage defect area is taken as the vertical axis. The temperature-time curve of each shrinkage defect area is obtained, wherein the value of m ranges from 1 to the number of shrinkage defect areas.

[0086] The process of obtaining the bubble-induced index value will be described in detail in the fifth embodiment and will not be repeated here.

[0087] The preset bubble triggering threshold can be set independently, preferably 0.5.

[0088] When When the bubble-induced index value of the first shrinkage defect area is greater than or equal to 0.5, the The first shrinkage defect area is marked as the bubble shrinkage defect area. When the bubble-induced index value of the first shrinkage defect area is less than 0.5, The shrinkage defect area is marked as a normal shrinkage defect area.

[0089] S104. Establish a rectangular coordinate system with the center point of the infrared image as the coordinate origin, take the center point of the bubble shrinkage defect area as the bubble coordinate, construct a bubble movement trend curve, and obtain the bubble coordinates to be corrected at the next moment based on the bubble movement trend curve.

[0090] After the bubble shrinkage defect area is determined, the temperature field changes during the subsequent solidification process of the metal liquid. Since the density of the metal increases as the temperature decreases, the unsolidified part of the metal liquid is more likely to flow. Therefore, when the metal liquid flows, the bubbles will be carried and move in the direction of flow. In order to better control the defects that may occur in the casting process and ensure the quality of the casting, it is necessary to predict the movement direction of the bubbles.

[0091] Since the acquisition interval is short, the bubbles in a bubble shrinkage defect area in each infrared image can also appear in the next infrared image. The center point of each bubble shrinkage defect area is obtained as the bubble center point. The center point of each infrared image is used as the coordinate origin to establish a rectangular coordinate system. The coordinates of each bubble center point, i.e., the bubble coordinates, are obtained. The bubble coordinates of each bubble shrinkage defect area are marked in the established rectangular coordinate system according to the time series, and a smooth curve is used to connect all the bubbles to obtain a bubble movement trend curve, which is recorded as Bubble movement trend curve. The value of is 1 to the number of the bubble shrinkage defect areas.

[0092] The process of obtaining the coordinates of the bubble to be corrected will be described in detail in the fifth embodiment and will not be repeated here.

[0093] S105. Obtain the metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtain the metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity.

[0094] The process of obtaining the flow velocity of the metal liquid is an existing technology.

[0095] The process of obtaining the metal liquid flow evaluation value will be described in detail in the sixth embodiment and will not be repeated here.

[0096] S106. Correct the bubble coordinates to be corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates.

[0097] The process of obtaining the corrected bubble coordinates includes:

[0098] The metal liquid flow evaluation value Multiply the coordinates of the bubble to be corrected to obtain the corrected coordinates of the bubble.

[0099] The corrected bubble coordinates can be expressed as:

[0100]

[0101] Among them, the represents the metal liquid flow evaluation value, the Indicates the coordinates of the bubble to be corrected.

[0102] In infrared images, regions of varying thickness exhibit different thermal radiation characteristics due to differences in material distribution and heat conduction, resulting in different infrared spectral absorption characteristics. Thick-walled areas, with their greater mass and slower heat conduction, may exhibit more pronounced infrared absorption, appearing as bright or highly absorbing areas in the image. Thin-walled areas, on the other hand, with their less mass and faster heat conduction, exhibit relatively weaker infrared spectral absorption, appearing as dark or low-absorbing areas in the image.

[0103] Threshold segmentation is used to divide the infrared image into regions, and then the thick-walled area and thin-walled area in the infrared image are distinguished.

[0104] Figure 2 A flowchart of distinguishing thick-walled areas and thin-walled areas in the infrared image provided by the second embodiment of the present invention, wherein the distinguishing thick-walled areas and thin-walled areas in the infrared image includes:

[0105] S201. Divide the infrared image into a preset second number of areas, obtain a first mean value of the brightness of the i-th area and a second mean value of the brightness of areas adjacent to the i-th area, wherein the value range of i is 1 to the second number.

[0106] The first average value of the brightness of the i-th region refers to the average value of the brightness corresponding to each pixel in the i-th region.

[0107] S202. Add the first mean and the second mean to obtain a sum of the means, obtain the absolute value of the difference between the first mean and the second mean, and then divide the difference by the sum of the means to obtain the spectral absorption evaluation value of the i-th region.

[0108] The spectral absorption evaluation value of the i-th region can be expressed as:

[0109]

[0110] Among them, the represents the spectral absorption evaluation value of the i-th region, represents the first mean, the represents the second mean, the represents the absolute value function.

[0111] described The larger the value, the The dark area in the infrared image shows a region with relatively weak absorption of the spectrum, which is more likely to be a thin-walled region.

[0112] S203. When the spectral absorption evaluation value is not less than a preset spectral absorption threshold, the i-th region is the thin-walled region; otherwise, the i-th region is the thick-walled region.

[0113] The preset spectral absorption threshold can be set independently, and is preferably 0.3.

[0114] During the casting process, normal thin-walled areas typically have a relatively uniform temperature distribution. This is because the heat conduction in these areas is relatively fast, resulting in a uniform cooling process and a steady temperature drop. Thin-walled areas with shrinkage defects, where the liquid metal fails to completely fill the mold cavity, can cause uneven solidification, faster or slower cooling, and an uneven temperature drop.

[0115] Figure 3 This is a flow chart of a process for obtaining a temperature drop stability value provided by a third embodiment of the present invention. The process for obtaining a temperature drop stability value includes:

[0116] S301. In each of the infrared images, obtain the temperature of the thin-walled area.

[0117] The temperature of the thin-walled area is the average of the temperatures corresponding to the pixels in the thin-walled area.

[0118] S302. Obtain the absolute value of the temperature difference between the thin-walled area in the two infrared images adjacent at a time, and obtain the average value of the absolute values as the absolute average value.

[0119] Can be used To express the Infrared image No. The temperature of the thin-walled area is subtracted from the temperature of the The infrared image obtained at the last moment of the infrared image The absolute value of the temperature of the thin-walled area.

[0120] The absolute value is indivual.

[0121] S303. Calculate the average of the absolute values obtained by subtracting the absolute average value from the absolute value, and take the reciprocal of the average value to obtain the temperature drop stable value.

[0122] The temperature drop stability value can be expressed as:

[0123]

[0124] Among them, the Indicates the The temperature drop stability value corresponding to the thin-walled area, Indicates the The absolute average value corresponding to the thin-walled area, represents the average value.

[0125] The larger the average value is, the The greater the temperature fluctuation of the thin-walled area in all infrared images, the worse the temperature drop stability.

[0126] Infrared images of normal areas typically do not show significant differences in temperature, indicating that hot spots exhibit a relatively consistent color distribution, indicating that the temperature distribution in that area is uniform. Shrinkage defect areas exhibit uneven temperature distribution, potentially showing localized overheating or underheating, which manifests as bright (high temperature) or dark (low temperature) areas on the infrared image, with hot spots exhibiting an uneven color distribution.

[0127] Figure 4 This is a flowchart of a process for obtaining a color distribution unevenness value provided by a fourth embodiment of the present invention. The process for obtaining a color distribution unevenness value includes:

[0128] S401. Calculate the variance of the temperature corresponding to the pixel points in the thin-walled area as the variance of the thin-walled area.

[0129] No. The variance of the thin-walled area corresponding to the thin-walled area can be To express.

[0130] S402. The thin-walled area variance The color distribution unevenness value is obtained by dividing by the temperature of the thin-walled area.

[0131] The color distribution unevenness value can be expressed as:

[0132]

[0133] Among them, the Indicates the The color distribution unevenness value corresponding to the thin-walled area, Indicates the The temperature of the thin-walled area.

[0134] described The larger the value, the more uneven the temperature distribution in the area, and the hot spots show uneven color distribution.

[0135] Figure 5 This is a flowchart of a process for obtaining an indicator value caused by bubbles provided in a fifth embodiment of the present invention. The process for obtaining an indicator value caused by bubbles includes:

[0136] S501. Obtain the temperature of the mth shrinkage defect area and the average of the temperatures of the shrinkage defect areas when the metal liquid begins to fill the mold, and divide the temperature of the mth shrinkage defect area by the average of the temperatures of the shrinkage defect areas to obtain a temperature ratio.

[0137] The temperature ratio can be expressed as: , wherein the represents the average temperature of the shrinkage defect area, represents the temperature of the mth shrinkage defect area.

[0138] described The larger the value, the The lower the temperature of a shrinkage defect region is compared with the temperatures of the other shrinkage defect regions, the more likely it is that bubbles exist in that region.

[0139] S502. Obtain the curvature of each moment in the temperature-time curve corresponding to the m-th shrinkage defect region, and calculate the average value of the curvature.

[0140] The average value of the curvature corresponding to the mth shrinkage defect area can be expressed as: .

[0141] S503. Calculate the average of the absolute values of the differences between the curvature at each moment and the average value of the curvature as the curvature difference value.

[0142] The curvature difference value can be expressed as:

[0143]

[0144] Among them, the represents the time number in the temperature-time curve corresponding to the mth shrinkage defect area, represents the curvature of the nth moment in the temperature-time curve corresponding to the mth shrinkage defect area, represents the absolute value function.

[0145] described The smaller the value, the The flatter the temperature-time curve of a shrinkage defect area and the slower the temperature rise, the more likely it is that bubbles exist in this area.

[0146] S504: The temperature ratio is divided by the curvature difference value and then normalized to obtain the bubble-inducing index value.

[0147] The bubble-induced index value can be expressed as:

[0148]

[0149] Among them, the Indicates the bubble-induced index value, is a normalization function, preferably a range normalization function.

[0150] Figure 6 This is a flow chart of a process for obtaining the coordinates of a bubble to be corrected provided in a sixth embodiment of the present invention. The process for obtaining the coordinates of a bubble to be corrected includes:

[0151] S601. Obtain the bubble coordinates at the current moment, differentiate the bubble coordinates to obtain the velocity vector of the bubble, and differentiate the velocity vector again to obtain the acceleration vector of the bubble.

[0152] The bubble coordinates are two-dimensional coordinates, assuming (x, y). Differentiating x with respect to time t gives dx / dt, which is the component of velocity in the x-coordinate direction. , differentiating y with respect to time t gives dy / dt, which is the component of velocity in the y-coordinate direction .

[0153] The value of the velocity vector can be expressed as:

[0154]

[0155] Similarly, the velocity vector is differentiated again to obtain Differentiating with respect to time t yields d / dt to get the acceleration component in the x-coordinate direction ,right Differentiating with respect to time t yields d / dt to get the acceleration component in the y-coordinate direction .

[0156] The value of the acceleration vector can be expressed as:

[0157]

[0158] S602. After obtaining the velocity vector and the acceleration vector, the Euler method is used to predict the bubble coordinates to be corrected at the next moment.

[0159] Use the Euler method to predict the position coordinates and velocity of the bubble at the next moment: x(t+Δt), y(t+Δt), v x (t+Δt), v y (t+Δt), where x(t+Δt) is the abscissa of the bubble coordinate to be corrected, y(t+Δt) is the ordinate of the bubble coordinate to be corrected, t represents the current moment, t+Δt represents the next moment, and Δt represents the time interval between the current moment and the next moment. The direction of bubble movement can be determined by the unit vector of the velocity vector.

[0160] The flow velocity of the metal liquid affects the movement of the bubbles. The metal liquid flow evaluation value of each bubble shrinkage defect area is used as the weight of the bubble center position to correct the predicted bubble center position.

[0161] Figure 7 This is a flow chart of a process for obtaining a metal liquid flow evaluation value provided by a seventh embodiment of the present invention. The process for obtaining a metal liquid flow evaluation value includes:

[0162] S701. Obtain the metal liquid flow velocity of the hth bubble shrinkage defect area in each of the infrared images, and obtain the average of the metal liquid flow velocities as the metal liquid flow velocity average.

[0163] The process of obtaining the flow velocity of the metal liquid is an existing technology, which briefly includes:

[0164] 1. Extracting the bubble shrinkage defect area in the infrared image.

[0165] 2. Use CFD simulation results (such as temperature field and flow velocity field) to match image data.

[0166] 3. Inferring flow velocity through the relationship between temperature gradient and flow velocity: By using the temperature field changes in infrared images and combining them with the temperature field simulated by CFD, the flow velocity can be estimated by calculating the temperature gradient (rate of temperature change).

[0167] S702. Calculate the average of the absolute values of the differences between the flow velocities of the metal liquid and the mean value of the flow velocities of the metal liquid as the metal liquid flow velocity difference value.

[0168] The metal liquid flow velocity difference can be expressed as:

[0169]

[0170] Among them, the represents the number of infrared images, Indicates the The bubble shrinkage defect area is in the The metal liquid flow velocity in the infrared image is It represents the mean value of the metal liquid flow velocity.

[0171] S703. Divide the metal liquid flow velocity difference value by the metal liquid flow velocity average value to obtain the metal liquid flow evaluation value.

[0172] The metal liquid flow evaluation value can be expressed as:

[0173]

[0174] Among them, the Indicates the The metal liquid flow evaluation value corresponding to the bubble shrinkage defect area.

[0175] Figure 8 This is a schematic diagram of a shrinkage defect prediction system for side plate casting provided by an eighth embodiment of the present invention. This embodiment of the present invention further provides a shrinkage defect prediction system for side plate casting, the system comprising:

[0176] An acquisition module 801 is configured to acquire infrared images of a first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared images;

[0177] Shrinkage defect performance evaluation value module 802 is used to obtain a temperature drop stability value and a color distribution unevenness value based on the temperature of the thin-walled area, and obtain a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value. When the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, the thin-walled area is designated as a shrinkage defect area, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to each pixel in the thin-walled area;

[0178] The bubble-induced index value module 803 is configured to obtain the temperature of the shrinkage defect region in the infrared image obtained at each moment to form a temperature-time curve, and obtain a bubble-induced index value of the shrinkage defect region based on the temperature-time curve. When the bubble-induced index value is not less than a preset bubble-induced threshold, the shrinkage defect region is a bubble-induced shrinkage defect region.

[0179] The coordinates to be corrected module 804 is used to establish a rectangular coordinate system with the center point of the infrared image as the coordinate origin, use the center point of the bubble shrinkage defect area as the bubble coordinate, construct a bubble movement trend curve, and obtain the bubble coordinates to be corrected at the next moment based on the bubble movement trend curve;

[0180] The metal liquid flow evaluation value module 805 is used to obtain the metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtain the metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity;

[0181] The correction module 806 is configured to correct the bubble coordinates to be corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates.

[0182] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0183] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for predicting shrinkage defects in side plate castings, characterized in that: The method comprises: Acquire an infrared image of a preset first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared image; Obtaining a temperature drop stability value and a color distribution unevenness value by using the temperature of the thin-walled area, obtaining a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value, and determining the thin-walled area as a shrinkage defect area when the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to each pixel in the thin-walled area; Acquiring the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, and obtaining a bubble-causing index value of the shrinkage defect area based on the temperature-time curve, and determining that the shrinkage defect area is a bubble shrinkage defect area when the bubble-causing index value is not less than a preset bubble-causing threshold value; A rectangular coordinate system is established with the center point of the infrared image as the coordinate origin, the center point of the bubble shrinkage defect area is used as the bubble coordinate, a bubble movement trend curve is constructed, and the bubble coordinates to be corrected at the next moment are obtained according to the bubble movement trend curve; Obtaining a metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtaining a metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity; Correcting the bubble coordinates to be corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates; The process of obtaining the temperature drop stable value includes: In each of the infrared images, obtaining the temperature of the thin-walled area; Obtaining the absolute value of the temperature difference between the thin-walled area in two infrared images adjacent at a time, and obtaining an average value of the absolute values as an absolute average value; Calculating an average of the absolute values of the absolute values minus the absolute average value, and taking the reciprocal of the average value to obtain the temperature drop stable value; The process of obtaining the color distribution unevenness value includes: Calculating the variance of the temperature corresponding to the pixel points in the thin-walled area as the variance of the thin-walled area; Dividing the variance of the thin-walled area by the temperature of the thin-walled area to obtain the color distribution unevenness value; The process of obtaining the shrinkage defect performance evaluation value includes: Multiplying the temperature drop stability value by the normalized value of the color distribution unevenness value as the shrinkage defect performance evaluation value; The process of obtaining the bubble-induced index value includes: Obtaining the temperature of the mth shrinkage defect region and the average of the temperatures of the shrinkage defect regions when the metal liquid begins to fill the mold, and dividing the temperature of the mth shrinkage defect region by the average of the temperatures of the shrinkage defect regions to obtain a temperature ratio; Obtaining the curvature of each moment in the temperature-time curve corresponding to the m-th shrinkage defect region, and calculating an average value of the curvature; Calculating an average of the absolute values of the differences between the curvature at each moment and the average value of the curvature as a curvature difference value; The temperature ratio is divided by the curvature difference value and then normalized to obtain the bubble-causing index value; The process of obtaining the metal liquid flow evaluation value includes: Obtaining the metal liquid flow velocity of the hth bubble shrinkage defect area in each of the infrared images, and obtaining an average of the metal liquid flow velocities as the metal liquid flow velocity average; Calculating an average of the absolute values of the differences between the metal liquid flow velocities and the metal liquid flow velocities mean as a metal liquid flow velocity difference value; The metal liquid flow evaluation value is obtained by dividing the metal liquid flow velocity difference value by the metal liquid flow velocity average value.

2. The method for predicting shrinkage defects in side plate casting according to claim 1, wherein: The distinguishing between thick-walled areas and thin-walled areas in the infrared image includes: Dividing the infrared image into a second number of regions, obtaining a first mean brightness of the i-th region and a second mean brightness of regions adjacent to the i-th region, wherein the value of i ranges from 1 to the second number; Adding the first mean and the second mean to obtain a sum of the means, obtaining an absolute value of a difference between the first mean and the second mean, and dividing the difference by the sum of the means to obtain a spectral absorption evaluation value of the i-th region; When the spectral absorption evaluation value is not less than a preset spectral absorption threshold, the i-th region is the thin-wall region; otherwise, the i-th region is the thick-wall region.

3. The shrinkage defect prediction method for side plate casting according to claim 1, characterized in that: The process of obtaining the coordinates of the bubble to be corrected includes: Obtaining the current bubble coordinates, differentiating the bubble coordinates to obtain a velocity vector of the bubble, and further differentiating the velocity vector to obtain an acceleration vector of the bubble; After obtaining the velocity vector and the acceleration vector, the Euler method is used to predict the bubble coordinates to be corrected at the next moment.

4. The method for predicting shrinkage defects in side plate casting according to claim 1, wherein: The process of obtaining the corrected bubble coordinates includes: The metal liquid flow evaluation value is multiplied by the bubble coordinate to be corrected to obtain the corrected bubble coordinate.

5. A shrinkage defect prediction system for side plate castings, characterized in that: The system comprises: an acquisition module, configured to acquire infrared images of a preset first number of side plate castings during the casting process, and distinguish thick-walled areas and thin-walled areas in the infrared images; a shrinkage defect performance evaluation value module, configured to obtain a temperature drop stability value and a color distribution unevenness value based on the temperature of the thin-walled area, and obtain a shrinkage defect performance evaluation value based on the temperature drop stability value and the color distribution unevenness value; and when the shrinkage defect performance evaluation value is not less than a preset shrinkage defect performance evaluation threshold, the thin-walled area is designated as a shrinkage defect area, wherein the temperature of the thin-walled area is the average of the temperatures corresponding to the pixels in the thin-walled area; a bubble-induced index value module, configured to obtain the temperature of the shrinkage defect area in the infrared image obtained at each moment to form a temperature-time curve, and obtain a bubble-induced index value of the shrinkage defect area based on the temperature-time curve; when the bubble-induced index value is not less than a preset bubble-induced threshold value, the shrinkage defect area is a bubble-induced shrinkage defect area; A coordinate module to be corrected is used to establish a rectangular coordinate system with the center point of the infrared image as the coordinate origin, use the center point of the bubble shrinkage defect area as the bubble coordinate, construct a bubble movement trend curve, and obtain the bubble coordinates to be corrected at the next moment according to the bubble movement trend curve; a metal liquid flow evaluation value module, configured to obtain a metal liquid flow velocity of the bubble shrinkage defect area in each of the infrared images, and obtain a metal liquid flow evaluation value of the bubble shrinkage defect area according to the metal liquid flow velocity; a correction module, configured to correct the bubble coordinates to be corrected according to the metal liquid flow evaluation value to obtain corrected bubble coordinates; The process of obtaining the temperature drop stable value includes: In each of the infrared images, obtaining the temperature of the thin-walled area; Obtaining the absolute value of the temperature difference between the thin-walled area in two infrared images adjacent at a time, and obtaining an average value of the absolute values as an absolute average value; Calculating an average of the absolute values of the absolute values minus the absolute average value, and taking the reciprocal of the average value to obtain the temperature drop stable value; The process of obtaining the color distribution unevenness value includes: Calculating the variance of the temperature corresponding to the pixel points in the thin-walled area as the variance of the thin-walled area; Dividing the variance of the thin-walled area by the temperature of the thin-walled area to obtain the color distribution unevenness value; The process of obtaining the shrinkage defect performance evaluation value includes: Multiplying the temperature drop stability value by the normalized value of the color distribution unevenness value as the shrinkage defect performance evaluation value; The process of obtaining the bubble-induced index value includes: Obtaining the temperature of the mth shrinkage defect region and the average of the temperatures of the shrinkage defect regions when the metal liquid begins to fill the mold, and dividing the temperature of the mth shrinkage defect region by the average of the temperatures of the shrinkage defect regions to obtain a temperature ratio; Obtaining the curvature of each moment in the temperature-time curve corresponding to the m-th shrinkage defect region, and calculating an average value of the curvature; Calculating an average of the absolute values of the differences between the curvature at each moment and the average value of the curvature as a curvature difference value; The temperature ratio is divided by the curvature difference value and then normalized to obtain the bubble-causing index value; The process of obtaining the metal liquid flow evaluation value includes: Obtaining the metal liquid flow velocity of the hth bubble shrinkage defect area in each of the infrared images, and obtaining an average of the metal liquid flow velocities as the metal liquid flow velocity average; Calculating an average of the absolute values of the differences between the metal liquid flow velocities and the metal liquid flow velocities mean as a metal liquid flow velocity difference value; The metal liquid flow evaluation value is obtained by dividing the metal liquid flow velocity difference value by the metal liquid flow velocity average value.

Images