A mold automatic casting method, system and computer equipment based on visual guidance

Abstract

The application relates to a mold automatic casting method and system based on visual guidance and a computer device. The method comprises the following steps: in response to the completion of the casting shell cleaning of a target casting, acquiring and determining the geometric size of each feature position of the target casting based on image data of the target casting; the feature position is divided based on the casting size and the casting curvature of the target casting; the geometric size of each feature position is compared with the corresponding target size, and the feature position with a size difference exceeding a preset difference is marked as a defect position; the size parameters of each defect position are modified according to a compensation strategy, the modified size parameters are fed back to a wax mold injection section for wax mold reprocessing, and the target casting is cast through the reprocessed wax mold; the method can effectively avoid the problem that the user needs to try the mold and cast multiple times before the finished product meets the required specifications, thereby reducing the production efficiency.

Description

Technical Field

[0001] This application relates to the field of investment casting technology, and in particular to a vision-guided automatic mold casting method, system and computer equipment. Background Technology

[0002] Investment casting is a precision casting process that involves melting a wax model, wrapping it with refractory material to form a shell, and then pouring in metal. It has advantages such as high precision, minimal machining, wide material adaptability, and flexible production, and is widely used in the manufacture of precision components in fields such as pumps, valves, and kitchenware.

[0003] Currently, due to phenomena such as mold shell expansion during firing and metal cooling contraction, the finished product dimensions are prone to deviation from the theoretical dimensions. This requires users to conduct multiple trial castings before the finished product reaches the required specifications, resulting in reduced production efficiency. Summary of the Invention

[0004] Therefore, it is necessary to provide a vision-guided automatic mold casting method, system, and computer equipment to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a vision-guided automatic mold casting method, the method comprising:

[0006] In response to the completion of the casting shell cleaning of the target casting, the geometric dimensions of each feature position of the target casting are acquired and determined based on the image data of the target casting; the feature positions are divided based on the casting dimensions and casting curvature of the target casting.

[0007] The geometric dimensions of each feature location are compared with the corresponding target dimensions, and feature locations with size differences exceeding a preset difference are marked as defect locations.

[0008] The parameters of each defect location are corrected according to the compensation strategy. The corrected dimensional parameters are fed back to the wax pattern injection molding section for wax pattern reprocessing. The target casting is then cast using the reprocessed wax pattern. The compensation strategy includes compensation for the difference in defect location based on the casting compensation size determined by the dimensional difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell.

[0009] In one embodiment, the feature locations include casting locations with geometric dimensions larger than a preset size and casting locations with geometric dimensions smaller than the preset size; the step of compensating for the difference in defect locations based on the casting compensation size determined jointly by the dimensional difference of the feature locations, the shrinkage rate corresponding to the casting material type of the feature locations, and the firing expansion rate of the casting shell includes:

[0010] The firing expansion rate of the casting shell is determined based on the geometry of the casting shell.

[0011] When the defect location is a casting location with a geometric dimension larger than the preset dimension, the casting compensation dimension is determined according to the first casting shrinkage rate corresponding to the casting location larger than the preset dimension based on the dimension difference and the casting material type, as well as the firing expansion rate of the mold shell; when the defect location is a casting location smaller than the preset dimension, the casting compensation dimension is determined according to the second casting shrinkage rate corresponding to the casting location smaller than the preset dimension based on the dimension difference and the casting material type, as well as the firing expansion rate of the mold shell.

[0012] Compensation is made for the difference in defect location based on the compensation dimensions of the casting.

[0013] In one embodiment, the feature location further includes a casting location with a curvature greater than a preset curvature; the step of compensating for the difference in the defect location based on the casting compensation size determined jointly by the size difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell may further include:

[0014] When the defect location is a casting location with a geometric dimension larger than the preset dimension, the casting compensation dimension is determined according to the dimensional difference, the shrinkage rate of the third casting exceeding the preset curvature corresponding to the casting material type, the shrinkage rate of the first casting, and the calcination expansion rate. When the defect location is a casting location smaller than the preset dimension, the casting compensation dimension is determined according to the shrinkage rate of the third casting exceeding the preset curvature corresponding to the casting material type, the shrinkage rate of the second casting, and the calcination expansion rate. The shrinkage rate of the third casting is the shrinkage rate in the normal direction of the surface of the target casting.

[0015] In one embodiment, the method further includes:

[0016] Based on the target dimensions of each feature position of the target casting, a pre-set correlation model is invoked to predict the dimensional differences at each feature position. The correlation model is constructed with the casting dimensions and casting curvature in the pre-set historical casting data as inputs and the corresponding dimensional differences in the historical casting data as outputs.

[0017] The predicted dimensional discrepancies are corrected using a compensation strategy, and the wax model is then processed based on the corrected dimensional parameters.

[0018] In one embodiment, the image data includes images of a six-sided target casting and images of the end effector of a robotic arm gripping each face; the step of acquiring and determining the geometric dimensions of each feature location of the target casting based on the image data includes:

[0019] The acquired image data is subjected to a first preprocessing operation to obtain the contour of the end effector and the contour of the target casting; the first preprocessing operation includes contrast enhancement, filtering and contour extraction;

[0020] The pixel width is calculated based on the preset end effector contour dimensions, and the contour dimensions of the target casting on each face are determined based on the pixel width.

[0021] Based on the outline dimensions of the target casting on each face and the feature positions of the target casting marked on each face, the geometric dimensions of each feature position are calculated.

[0022] In one embodiment, the method further includes:

[0023] If, after casting the target part using a reworked wax model, there are still feature locations marked as defect locations, a casting process backtracking strategy is executed. The casting process backtracking strategy includes:

[0024] The image data of the defect location is retrieved and image analysis is performed to obtain the defect characteristics of the casting;

[0025] In response to the defect feature being a bulging defect, a first defect warning message is output; the first defect warning message is used to indicate the material parameters for the preparation of the shell.

[0026] In response to the defect characteristic being a concentrated porosity defect, a second defect warning message is output; the second defect warning message is used to indicate the inspection of the roasting process parameters and the gating and riser system design parameters; the roasting process parameters include roasting temperature and roasting time;

[0027] In response to the defect characteristic being shrinkage cavity, a third defect warning message is output; the third defect warning message is used to indicate the design parameters of the gating and riser system and the type and temperature of the casting material used for feeding.

[0028] In one embodiment, the step of calling image data of the defect location and performing image analysis to obtain the defect features of the casting includes:

[0029] The image data of the defect location is subjected to a second preprocessing operation to obtain the contour of the target casting; the second preprocessing operation includes contrast enhancement, filtering, binarization and contour extraction.

[0030] The feature comparison is performed between the defect features of bulging defects in the pre-set database and the contour of the target casting, and the comparison results are output.

[0031] In one embodiment, the step of calling image data of the defect location and performing image analysis to obtain the defect features of the casting further includes:

[0032] In response to the fact that the contour of the target casting includes the contour of holes, the feature comparison is performed between the feature of concentrated porosity defects and shrinkage defects in the preset database and the contour of the target casting, and the comparison result is output.

[0033] Secondly, this application also provides a vision-guided automatic mold casting system, the system comprising:

[0034] An acquisition device is used to acquire and determine the geometric dimensions of each feature position of the target casting based on the image data of the target casting in response to the completion of the casting shell cleaning of the target casting; the feature positions are divided based on the casting size and casting curvature of the target casting.

[0035] The parameter comparison device is used to compare the geometric dimensions of each feature location with the corresponding target dimensions, and to mark feature locations whose size differences exceed a preset difference as defect locations.

[0036] The parameter correction device is used to correct the parameters of each defect location according to the compensation strategy, feed back the corrected dimensional parameters to the wax pattern injection molding section for wax pattern reprocessing, and cast the target casting through the reprocessed wax pattern. The compensation strategy includes compensation for the difference in defect location based on the casting compensation size determined by the dimensional difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method provided in the first aspect of this application.

[0038] The aforementioned vision-guided automatic casting method, system, and computer equipment can automatically acquire image data of the target casting after the casting shell is cleaned to determine the geometric dimensions of feature positions with different casting sizes and curvatures. If it is determined that the casting size at any feature position differs from the required target size, the unqualified feature position is marked as a defect position. Then, a corresponding shrinkage rate is introduced for the casting size and curvature at the defect position. Combined with the size difference, casting material type, and the firing expansion rate of the casting shell, the compensation size of the casting is determined. Based on this, a compensation strategy is formulated to correct the defect position parameters. The corrected parameters are then fed back to the wax pattern injection molding section for wax pattern reprocessing, thereby guiding the subsequent series of casting processes. This provides an effective basis for parameter adjustment after the first casting of the target casting fails, avoiding the problem of users needing to conduct multiple trial castings before the finished product meets the required specifications, which leads to reduced production efficiency. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart of a method for correcting defect location parameters according to a compensation strategy in one embodiment;

[0041] Figure 2 This is a flowchart of a method for correcting prediction discrepancies according to a compensation strategy in one embodiment;

[0042] Figure 3 This is a structural block diagram of a vision-guided automatic mold casting system in one embodiment. Detailed Implementation

[0043] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0044] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0045] This application relates to the casting production process of investment casting, a precision casting process. The complete production process includes: wax pattern injection molding, where a wax material with a specific formula is heated to a molten state and injected into a metal mold cavity through a precision wax press. After cooling and solidification, the wax pattern is removed, obtaining a wax pattern with dimensions consistent with the final casting; wax pattern tree assembly, where these wax patterns are precisely assembled with a wax gating system (including pouring cups, sprues, runners, and ingates) using manual or automated equipment to form a structurally sound wax pattern tree; shell preparation, where the assembled wax pattern tree is immersed in a refractory slurry prepared with silica sol or ethyl silicate binder, and then refractory materials such as zircon sand and mullite of different particle sizes are sprinkled on top. After multiple cycles of immersion-sanding-drying, a multi-layered ceramic shell is finally formed. The wax material is completely melted out by high-pressure steam dewaxing or hot water dewaxing to obtain a ceramic shell with a complete cavity; and shell firing, where the shell is placed in a mold. Sintering is performed in a high-temperature baking furnace to completely remove residual wax and improve the strength of the mold shell. For casting, the casting material is melted according to material requirements and adjusted to a suitable chemical composition and pouring temperature. The molten metal casting material is then poured into the preheated mold shell. The mold shell is cleaned after the metal solidifies and cools, using methods such as mechanical vibration, high-pressure water jetting, or sandblasting. The casting is then post-processed, including shot peening, polishing, and machining. Because the mold shell expands due to heat during the baking stage and the molten metal shrinks during the casting stage, the dimensions of the cast product differ from the actual size. Therefore, existing technologies typically manufacture wax molds larger than the target casting, then allow for shrinkage and finishing to obtain the finished product. This method is difficult to control, leading to increased workload and potential defects in the target casting. Repeated trial castings further reduce work efficiency.

[0046] This application provides an embodiment of a vision-guided automatic mold casting system, such as... Figure 1 As shown, the method includes the following steps S102 to S106. Wherein:

[0047] In step S102, in response to the completion of the cleaning of the casting shell of the target casting, the geometric dimensions of each feature position of the target casting are acquired and determined based on the image data of the target casting; the feature positions are divided based on the casting size and casting curvature of the target casting.

[0048] Among them, the target casting is used to characterize the finished casting after the shell cleaning is completed.

[0049] Specifically, image data can be obtained by taking pictures of the target casting using an industrial camera positioned at the location where the six-axis robot transfers the casting after the shell cleaning is completed.

[0050] Specifically, since the shrinkage rates of different casting sizes and curvatures are different, the shrinkage rates of each casting location are distinguished by introducing characteristic locations.

[0051] Step S104: Compare the geometric dimensions of each feature location with the corresponding target dimensions, and mark the feature locations whose size differences exceed the preset difference as defect locations.

[0052] The target size is used to characterize the casting size required for the product order.

[0053] Specifically, the preset difference is set based on the precision requirements of casting.

[0054] Step S106: Correct the parameters of each defect location according to the compensation strategy, feed back the corrected dimensional parameters to the wax pattern injection molding section for wax pattern reprocessing, and cast the target casting through the reprocessed wax pattern; the compensation strategy includes compensation for the difference in defect location based on the dimensional difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell.

[0055] Specifically, since the casting process can include multiple steps such as wax pattern injection molding, wax pattern assembly, shell preparation, shell firing, and casting, the finished shape of the wax pattern directly determines the finished shape of the target casting. After considering the differences in the multiple casting processes, by applying all the compensation dimensions for the differences to the initial wax pattern size, the final finished shape of the target casting can be effectively guided.

[0056] Furthermore, when comparing defect locations, if compensation is directly based on the comparison results, the compensation value may again generate new errors due to the shrinkage rate, leading to further defects. Therefore, this embodiment calculates the compensation value by considering dimensional differences, the shrinkage rate of the feature location, the type of casting material, and the firing expansion rate of the mold shell, thus compensating for the shrinkage that may occur at the difference location. In addition, this embodiment avoids calculations based on the entire component, but only calculates for the differences. This allows for compensation and adjustment even when the size of the difference is relatively small compared to the overall size of the casting, effectively achieving millimeter-level compensation.

[0057] The aforementioned vision-guided automatic casting method for molds can automatically acquire image data of the target casting after the casting shell is cleaned to determine the geometric dimensions of feature positions with different casting sizes and curvatures. If it is determined that the casting size at any feature position differs from the required target size, the unqualified feature position is marked as a defect position. Then, a corresponding shrinkage rate is introduced for the casting size and curvature at the defect position. Combined with the size difference, casting material type, and the firing expansion rate of the casting shell, the compensation size of the casting is determined. Based on this, a compensation strategy is formulated to correct the defect position parameters. The corrected parameters are then fed back to the wax pattern injection molding section for wax pattern reprocessing, thereby guiding the subsequent series of casting processes. This provides an effective basis for parameter adjustment after the first casting of the target casting fails, avoiding the problem of users having to conduct multiple trial castings before the finished product meets the required specifications, which leads to a decrease in production efficiency.

[0058] In one exemplary embodiment, specifically, the difference compensation formula includes:

[0059] Compensation difference = Size difference × [(1+α)] 铸件收缩 ) / (1− β 焙烧膨胀 )]

[0060] In the formula: α 铸件收缩 This indicates the shrinkage rate of the casting for the corresponding material type. β 焙烧膨胀 This indicates the shell firing expansion rate for the corresponding material type.

[0061] In an exemplary embodiment, the feature locations include casting locations with geometric dimensions larger than a preset size and casting locations with geometric dimensions smaller than the preset size; the step of compensating for the difference in defect locations based on the casting compensation size jointly determined by the size difference of the feature locations, the shrinkage rate corresponding to the casting material type of the feature locations, and the firing expansion rate of the casting shell includes:

[0062] The firing expansion rate of the casting shell is determined based on the geometry of the casting shell.

[0063] When the defect location is a casting location with a geometric dimension larger than the preset dimension, the casting compensation dimension is determined according to the first casting shrinkage rate corresponding to the casting location larger than the preset dimension based on the dimension difference and the casting material type, as well as the firing expansion rate of the mold shell; when the defect location is a casting location smaller than the preset dimension, the casting compensation dimension is determined according to the second casting shrinkage rate corresponding to the casting location smaller than the preset dimension based on the dimension difference and the casting material type, as well as the firing expansion rate of the mold shell.

[0064] Compensation is made for the difference in defect location based on the compensation dimensions of the casting.

[0065] Among them, the casting positions with geometric dimensions larger than the preset dimensions are used to characterize the casting positions where either the thickness or width exceeds the preset value. In the specific implementation process, it was found that, taking steel materials as an example, when the thickness or width is lower than the preset value, the shrinkage is relatively small due to the faster cooling and the obstruction of shrinkage. However, when the thickness or width is greater than the preset value, the thermal shrinkage is significant and the shrinkage is significantly increased.

[0066] For example, the casting material can be stainless steel, and the preset size can be a width or thickness of less than 3 mm. In the specific implementation process, the shrinkage rate of the casting position larger than the preset size is about 0.95%, and the shrinkage rate of the casting position smaller than the preset size is about 0.37%.

[0067] In an exemplary embodiment, the feature location further includes a casting location with a curvature greater than a preset curvature; the step of compensating for the difference in the defect location based on the casting compensation size determined jointly by the dimensional difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell further includes:

[0068] When the defect location is a casting location with a geometric dimension larger than the preset dimension, the casting compensation dimension is determined according to the dimensional difference, the shrinkage rate of the third casting exceeding the preset curvature corresponding to the casting material type, the shrinkage rate of the first casting, and the calcination expansion rate. When the defect location is a casting location smaller than the preset dimension, the casting compensation dimension is determined according to the shrinkage rate of the third casting exceeding the preset curvature corresponding to the casting material type, the shrinkage rate of the second casting, and the calcination expansion rate. The shrinkage rate of the third casting is the shrinkage rate in the normal direction of the surface of the target casting.

[0069] Specifically, since the shrinkage rate of a curved surface is found to be 1% higher than that of a flat surface during the implementation process, it is necessary to add the shrinkage rate of the third casting to the shrinkage rate of the first casting, or add the shrinkage rate of the third casting to the shrinkage rate of the first casting, in order to prevent contour distortion.

[0070] In one exemplary embodiment, such as Figure 2 As shown, the method further includes the following steps S202 to S204. Wherein:

[0071] Step S202: Based on the target dimensions of each feature position of the target casting, a pre-set correlation model is called to predict the size difference of each feature position. The correlation model is constructed with the casting dimensions and casting curvature in the pre-set historical casting data as inputs and the corresponding size differences in the historical casting data as outputs.

[0072] Specifically, in order to provide reference values ​​for the first wax mold injection molding process, the potential differences in the target size in the current order information during the production process can be determined by the dimensional differences in historical casting data, and the differences can be compensated through a compensation strategy to shorten the trial molding cycle.

[0073] Since the casting dimensions in the order information may not exist in the historical casting data, the size difference at each feature location is calculated through a correlation model. The specific method for constructing the correlation model can refer to existing solutions, and will not be elaborated here.

[0074] Step S204: Correct the predicted dimensional difference using a compensation strategy, and perform wax model processing based on the corrected dimensional parameters.

[0075] In one exemplary embodiment, the image data includes images of a six-sided target casting and images of the end effector of a robotic arm gripping each face; the step of acquiring and determining the geometric dimensions of each feature location of the target casting based on the image data includes:

[0076] The acquired image data is subjected to a first preprocessing operation to obtain the contour of the end effector and the contour of the target casting; the first preprocessing operation includes contrast enhancement, filtering and contour extraction;

[0077] The pixel width is calculated based on the preset end effector contour dimensions, and the contour dimensions of the target casting on each face are determined based on the pixel width.

[0078] Based on the outline dimensions of the target casting on each face and the feature positions of the target casting marked on each face, the geometric dimensions of each feature position are calculated.

[0079] Specifically, after the casting is completed and the mold shell is cleaned, the embodiment of this application uses the end effector that grips the target casting as a reference to calculate the obtained image pixels, and determines the outline size through the pixels. Since the size of the end effector is known, it is easy to accurately determine the outline size of the target casting.

[0080] Furthermore, the reference position of the end effector can be selected from the position of the chuck section.

[0081] The embodiments of this application avoid using focal length calculation, which can ignore the minor errors in robot movements, thus making the calculated contour dimensions more accurate.

[0082] In one exemplary embodiment, the method further includes the following steps:

[0083] If, after casting the target part using a reworked wax model, there are still feature locations marked as defect locations, a casting process backtracking strategy is executed. The casting process backtracking strategy includes:

[0084] Image data of the defect location is retrieved and analyzed to obtain the defect characteristics of the casting. In response to the defect characteristic being a bulging defect, a first defect warning message is output. This first defect warning message indicates the parameters of the shell preparation material to be checked. In response to the defect characteristic being a concentrated porosity defect, a second defect warning message is output. This second defect warning message indicates the parameters of the baking process and the design parameters of the gating and riser system to be checked. The baking process parameters include baking temperature and baking time. In response to the defect characteristic being a shrinkage cavity defect, a third defect warning message is output. This third defect warning message indicates the design parameters of the gating and riser system and the type and temperature of the casting material used for feeding.

[0085] Specifically, during implementation, it was found that: bulging defects often appear as localized protrusions on the casting surface, frequently occurring when the strength of the casting shell is low under high or medium temperatures. Therefore, users need to be instructed to check the material parameters for the shell preparation to ensure that the shell material meets the corresponding processing requirements. Concentrated porosity defects often occur when the firing temperature is too low during the shell firing process, resulting in poor shell permeability and preventing effective venting, thus creating porosity. Another cause is improper gating and riser design leading to poor venting. Therefore, users are advised to check the firing process parameters and the gating and riser system design parameters. Shrinkage defects are irregular voids appearing in thick sections of the casting, often caused by improper gating and riser design or ineffective feeding. Therefore, users are instructed to check the gating and riser system design parameters and the type and temperature of the casting material used for feeding. The ideas in this application's embodiments can also be applied to other defects, such as shrinkage porosity defects. Specific adjustments to the camera accuracy are needed, and further examples are not provided here.

[0086] In an exemplary embodiment, the step of calling image data of the defect location and performing image analysis to obtain the defect features of the casting includes:

[0087] The image data of the defect location is subjected to a second preprocessing operation to obtain the contour of the target casting; the second preprocessing operation includes contrast enhancement, filtering, binarization and contour extraction.

[0088] The feature comparison is performed between the defect features of bulging defects in the pre-set database and the contour of the target casting, and the comparison results are output.

[0089] In an exemplary embodiment, the step of calling image data of the defect location and performing image analysis to obtain the defect features of the casting further includes:

[0090] In response to the fact that the contour of the target casting includes the contour of holes, the feature comparison is performed between the feature of concentrated porosity defects and shrinkage defects in the preset database and the contour of the target casting, and the comparison result is output.

[0091] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0092] Based on the same inventive concept, this application also provides a vision-guided automatic mold casting device for implementing the above-mentioned automatic mold casting method. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more vision-guided automatic mold casting device embodiments provided below can be found in the above-described limitations of the vision-guided automatic mold casting method, and will not be repeated here.

[0093] In one exemplary embodiment, such as Figure 3 As shown, a vision-guided automatic mold casting system 400 is provided, the system comprising:

[0094] The acquisition device 401 is used to acquire and determine the geometric dimensions of each feature position of the target casting based on the image data of the target casting in response to the completion of the casting shell cleaning of the target casting; the feature positions are divided based on the casting size and casting curvature of the target casting.

[0095] The parameter comparison device 402 is used to compare the geometric dimensions of each feature position with the corresponding target size, and mark the feature positions with size differences exceeding a preset difference as defect positions;

[0096] The parameter correction device 403 is used to correct the parameters of each defect location according to the compensation strategy, feed back the corrected dimensional parameters to the wax pattern injection molding section for wax pattern reprocessing, and cast the target casting through the reprocessed wax pattern; the compensation strategy includes compensation for the difference in defect location based on the casting compensation size determined by the dimensional difference of the feature location, the shrinkage rate corresponding to the casting material type of the feature location, and the firing expansion rate of the casting shell.

[0097] The various modules in the aforementioned vision-guided automatic mold casting device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0098] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0099] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0100] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described above.

[0101] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0102] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A vision-guided automatic mold casting method, characterized in that, The method includes: In response to the completion of the casting shell cleaning of the target casting, the geometric dimensions of each feature position of the target casting are acquired and determined based on the image data of the target casting; the feature positions are divided based on the casting dimensions and casting curvature of the target casting. The geometric dimensions of each feature location are compared with the corresponding target dimensions, and feature locations with size differences exceeding a preset difference are marked as defect locations. The parameters of each defect location are corrected according to the compensation strategy. The corrected dimensional parameters are fed back to the wax pattern injection molding section for wax pattern reprocessing, and the target casting is then cast using the reprocessed wax pattern. The compensation strategy includes compensating for the differences in the defect locations based on the dimensional differences of the feature locations, the shrinkage rate corresponding to the casting material type at the feature locations, and the firing expansion rate of the casting shell. The formula for the difference compensation includes: Compensation difference = Size difference × [(1+α)] 铸件收缩 ) / (1− β 焙烧膨胀 )] In the formula: α 铸件收缩 This indicates the shrinkage rate of the casting for the corresponding material type. β 焙烧膨胀 This indicates the shell firing expansion rate for the corresponding material type.

2. The method according to claim 1, characterized in that, The characteristic locations include casting locations with geometric dimensions larger than a preset size and casting locations with geometric dimensions smaller than a preset size; the step of compensating for the difference in the defect locations based on the casting compensation size determined jointly by the dimensional difference of the characteristic locations, the shrinkage rate corresponding to the casting material type of the characteristic locations, and the firing expansion rate of the casting shell includes: The firing expansion rate of the casting shell is determined based on the geometry of the casting shell. When the defect location is a casting location with a geometric dimension greater than a preset dimension, the casting compensation dimension is determined according to the dimension difference, the first casting shrinkage rate corresponding to the casting location greater than the preset dimension for the casting material type, and the firing expansion rate of the mold shell; when the defect location is a casting location less than the preset dimension, the casting compensation dimension is determined according to the dimension difference, the second casting shrinkage rate corresponding to the casting location less than the preset dimension for the casting material type, and the firing expansion rate of the mold shell. The difference in the location of the defect is compensated based on the compensation dimensions of the casting.

3. The method according to claim 2, characterized in that, The characteristic location also includes casting locations with curvature greater than a preset curvature; the step of compensating for the difference in the defect location based on the casting compensation size determined jointly by the size difference of the characteristic location, the shrinkage rate corresponding to the casting material type of the characteristic location, and the firing expansion rate of the casting shell further includes: When the defect location is a casting location with a geometric dimension larger than a preset dimension, the casting compensation dimension is determined according to the dimension difference, the third casting shrinkage rate exceeding the preset curvature corresponding to the casting material type, the first casting shrinkage rate, and the firing expansion rate; when the defect location is a casting location below the preset dimension, the casting compensation dimension is determined according to the third casting shrinkage rate exceeding the preset curvature corresponding to the casting material type, the second casting shrinkage rate, and the firing expansion rate; the third casting shrinkage rate is the shrinkage rate in the normal direction of the surface of the target casting.

4. The method according to claim 1, characterized in that, The method further includes: Based on the target dimensions of each feature position of the target casting, a preset correlation model is invoked to predict the size difference of each feature position; the correlation model is constructed with the casting dimensions and casting curvature in the preset historical casting data as input and the corresponding size difference in the historical casting data as output. The predicted dimensional difference is corrected using the compensation strategy, and wax model processing is performed based on the corrected dimensional parameters.

5. The method according to claim 1, characterized in that, The image data includes images of the target casting from all six sides, and images of the end effector of the robotic arm gripping each side; the step of acquiring and determining the geometric dimensions of each feature position of the target casting based on the image data of the target casting includes: The acquired image data is subjected to a first preprocessing operation to obtain the contour of the end effector and the contour of the target casting; the first preprocessing operation includes contrast enhancement, filtering, and contour extraction; The pixel width is calculated based on the preset contour dimensions of the end effector, and the contour dimensions of the target casting on each face are determined based on the pixel width. Based on the outline dimensions of the target casting on each face and the feature positions of the target casting marked on each face, the geometric dimensions of each feature position are calculated.

6. The method according to claim 1, characterized in that, The method further includes: If, after casting the target part using a reworked wax model, there are still characteristic locations marked as the defect locations, a casting process backtracking strategy is executed; the casting process backtracking strategy includes: The image data of the defect location is retrieved and image analysis is performed to obtain the defect features of the casting; In response to the defect feature being a bulging defect, a first defect warning message is output; the first defect warning message is used to indicate the material parameters for the preparation of the shell. In response to the defect characteristic being a concentrated porosity defect, a second defect warning message is output; the second defect warning message is used to indicate the inspection of the roasting process parameters and the gating and riser system design parameters; the roasting process parameters include roasting temperature and roasting time; In response to the defect feature being a shrinkage cavity defect, a third defect warning message is output; the third defect warning message is used to indicate the need to check the design parameters of the gating and riser system and the type and temperature of the casting material used for feeding.

7. The method according to claim 6, characterized in that, The step of calling the image data of the defect location and performing image analysis to obtain the defect features of the casting includes: The image data at the defect location is subjected to a second preprocessing operation to obtain the outline of the target casting; the second preprocessing operation includes contrast enhancement, filtering, binarization, and outline extraction. The feature comparison is performed between the defect features of bulging defects in the pre-set database and the contour of the target casting, and the comparison results are output.

8. The method according to claim 7, characterized in that, The step of calling the image data of the defect location and performing image analysis to obtain the defect features of the casting further includes: In response to the fact that the contour of the target casting includes a hole contour, the feature comparison is performed between the feature of concentrated porosity defects and shrinkage defects in the preset database and the contour of the target casting, and the comparison result is output.

9. A vision-guided automatic mold casting system, characterized in that, The system includes: An acquisition device is configured to acquire and determine the geometric dimensions of each feature position of the target casting based on image data of the target casting in response to the completion of the casting shell cleaning of the target casting; the feature positions are defined based on the casting dimensions and casting curvature of the target casting. The parameter comparison device is used to compare the geometric dimensions of each of the feature locations with the corresponding target dimensions, and to mark the feature locations whose size difference exceeds a preset difference as defect locations; A parameter correction device is used to correct the parameters of each defect location according to a compensation strategy, feed back the corrected dimensional parameters to the wax pattern injection molding section for wax pattern reprocessing, and then cast the target part using the reprocessed wax pattern. The compensation strategy includes compensating for the differences in the defect locations based on the dimensional differences of the feature locations, the shrinkage rate corresponding to the casting material type at the feature locations, and the firing expansion rate of the casting shell. The formula for the difference compensation includes: Compensation difference = Size difference × [(1+α)] 铸件收缩 ) / (1− β 焙烧膨胀 )] In the formula: α 铸件收缩 This indicates the shrinkage rate of the casting for the corresponding material type. β 焙烧膨胀 This indicates the shell firing expansion rate for the corresponding material type.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images