Die casting device suitable for automatic production

Abstract

The invention relates to the technical field of metal casting, in particular to a die-casting device suitable for automatic production, comprising: a die opening and closing control module for controlling die opening and closing of a forming structure, the forming structure comprising a movable die, a fixed die and a plurality of guide pillars; the image acquisition module is used for acquiring a mold opening image and a mold closing image containing a forming structure after the mold is opened and closed in place; the mold analysis module is used for analyzing parting surfaces of the movable mold and the fixed mold according to the mold opening image and judging whether the parting surfaces are abnormal or not; the processor is further used for analyzing a gap between the movable mold and the fixed mold according to the mold closing image and judging whether mold closing is abnormal or not; the guide analysis module is used for analyzing movement information of the guide columns according to the mold closing image and judging whether movement abnormity occurs or not; and the comprehensive analysis module is used for generating and sending an abnormal signal according to one or more of parting surface abnormity, mold closing abnormity and movement abnormity. By adopting the scheme, the technical problem that abnormal conditions in mold opening and closing are difficult to find in time in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal casting, and specifically provides a die-casting device suitable for automated production. Background Art

[0002] Die casting is an advanced forming process in which molten metal is injected into a precision die cavity at high speed under high pressure, and high-precision metal parts are obtained after cooling and solidification. With its high efficiency, high surface quality, and the ability to form complex structures, this process is widely used in the fields of automotive, aerospace, and electronic device manufacturing. The main structures involved in the die-casting forming process include the gating system, the forming system, etc. Among them, the forming structure includes a moving die, a fixed die, a guiding structure, and a pushing structure. During use, the guiding structure controls the movement of the moving die along a specified direction, causing the moving die and the fixed die to abut against each other to achieve mold clamping and form a die-casting forming mold. Molten metal is injected into the die-casting forming mold through the gating system. After the molten metal cools, the guiding structure controls the reverse movement of the moving die to open the mold, and the pushing mechanism pushes the workpiece formed by cooling the molten metal out of the moving die and the fixed die.

[0003] During the use process, the gating system needs to inject molten metal into the die-casting forming mold under a certain pressure. To prevent the die-casting forming mold from breaking through the parting surface, that is, mold swelling, the die-casting forming mold needs to withstand a continuous clamping pressure of up to 50 - 150 MPa. However, under long-term high-pressure working conditions, the forming structure is prone to cumulative fatigue damage: First, during the long-term and high-frequency use process of the guiding structure, the clearance between the guide pillar and the guide sleeve of the guiding structure increases due to surface fretting wear, resulting in axial deflection of the moving die during movement, interfering with the synchronism of the movement of each part of the moving die. Second, during the long-term and high-pressure use process, the contact stress distribution on the parting surface between the moving die and the fixed die is uneven, resulting in local plastic deformation of the moving die and the fixed die, destroying the parallelism between the moving die and the fixed die. Once abnormal situations such as non-synchronous mold opening and closing or inclination of the parting surface between the moving die and the fixed die occur, the formed workpiece will have differences in flash size and unbalanced cavity filling, directly affecting the dimensional accuracy and surface density of the formed workpiece, and greatly reducing the qualified rate of the formed workpiece and the subsequent processing difficulty. Therefore, there is an urgent need for an automated production die-casting device that can timely detect non-synchronous situations during mold opening and closing. Summary of the Invention

[0004] The present invention aims to provide a die-casting device suitable for automated production to solve the technical problem in the prior art that it is difficult to timely detect abnormal situations during mold opening and closing.

[0005] The present invention provides the following basic solution:

[0006] A die-casting device suitable for automated production, comprising:

[0007] A mold opening and closing control module for controlling the mold opening and closing of a molding structure, where the molding structure includes a moving mold, a fixed mold, and a plurality of guide pillars;

[0008] It further includes:

[0009] An image acquisition module for acquiring an open mold image and a closed mold image containing the molding structure after the mold opening and closing are in place;

[0010] A mold analysis module for analyzing the parting surface between the moving mold and the fixed mold according to the open mold image to determine whether there is an abnormal parting surface; and also for analyzing the gap between the moving mold and the fixed mold according to the closed mold image to determine whether there is an abnormal mold closing;

[0011] A guide analysis module for analyzing the movement information of the plurality of guide pillars according to the closed mold image to determine whether there is an abnormal movement;

[0012] A comprehensive analysis module for generating and sending an abnormal signal according to one or more of the abnormal parting surface, abnormal mold closing, and abnormal movement.

[0013] Furthermore, the number of image acquisition modules is at least two. The image acquisition modules are respectively oriented towards the moving mold and the fixed mold. The image acquired by the image acquisition module oriented towards the moving mold contains the parting surface of the moving mold, and the image acquired by the image acquisition module oriented towards the fixed mold contains the parting surface of the fixed mold.

[0014] Furthermore, the mold analysis module presets a standard parting surface for the moving mold and the fixed mold. The mold analysis module is used to match the standard parting surface with the open mold image. When the standard parting surface does not match the parting surface in the open mold image, it is determined that there is an abnormal parting surface.

[0015] Furthermore, the mold analysis module is also used to perform image recognition according to the closed mold image, mark the parting surface edges of the moving mold and the fixed mold, divide the edge lines into edge line groups according to the angles of the parting surface edges, and the number of parting surface edges in the same edge line group is not more than two;

[0016] It is also used to calculate the straight-line distance between the parting surface edges in each edge line group and determine whether the straight-line distance is greater than a preset gap threshold. When the straight-line distance is greater than the gap threshold, it is determined that there is an abnormal mold closing.

[0017] Furthermore, the guide analysis module is used to perform image recognition according to the closed mold image, analyze the movement information of each guide pillar, calculate the guide pillar fluctuation range according to the movement information, and determine whether the movement information of each guide pillar is within the guide pillar fluctuation range. When the movement information of any one guide pillar is outside the guide pillar fluctuation range, it is determined that there is an abnormal movement.

[0018] Furthermore, the image acquisition module is also used to acquire dynamic images during the mold closing process, and the dynamic images contain the molding structure;

[0019] The guiding analysis module is also used to analyze the dynamic information of multiple guide pillars based on the dynamic image, determine whether the guide pillars have asynchronous movement, and when asynchronous movement occurs, it is determined that abnormal movement has occurred.

[0020] Furthermore, the guiding analysis module is also used to perform image recognition based on the dynamic image, analyze the dynamic information of each guide pillar at the same moment, calculate the dynamic fluctuation range according to the dynamic information, and determine whether the dynamic information of each guide pillar is within the dynamic fluctuation range. When the dynamic information of any guide pillar is outside the dynamic fluctuation range, it is determined that the guide pillars have asynchronous movement.

[0021] Furthermore, it further includes:

[0022] A storage module for storing the mold opening image, mold closing image, and dynamic image;

[0023] A model training module for, when the comprehensive analysis module generates an abnormal signal, calling the mold opening image, mold closing image, dynamic image, and abnormal signal stored in the storage module to generate and train an abnormal prediction model;

[0024] A model prediction module for calling the abnormal prediction model and predicting the mold maintenance information according to the abnormal prediction model.

[0025] Furthermore, the comprehensive analysis module presets a maintenance threshold, and the comprehensive analysis module is also used to determine whether the mold maintenance information is less than the maintenance threshold. When the mold maintenance information is less than the maintenance threshold, a maintenance signal is generated and sent.

[0026] Furthermore, the image acquisition module is also used to acquire the finished product image after mold opening;

[0027] It further includes:

[0028] A model trigger module for determining whether the die-casting product is qualified according to the finished product image;

[0029] The model prediction module is used to call the abnormal prediction model when the model trigger module determines that the die-casting product is unqualified.

[0030] Beneficial effects:

[0031] 1. In this solution, the mold analysis module is set to analyze the static mold opening image and mold closing image, so as to determine whether there is an abnormality in the parting surface between the moving mold and the fixed mold. For example, due to fatigue damage, surface abnormalities that affect the quality of the finished product may occur, or due to the inclination of the moving mold, the parting surface is different from the standard parting surface. At the same time, when the mold is closed, the distance between the parting surfaces of the moving mold and the fixed mold is used to determine whether the moving mold or the fixed mold has fatigue damage or inclination. The opening and closing of the moving mold are controlled by guide pillars. The guiding analysis module is set to analyze the length of the guide pillars on the side of the moving mold away from the fixed mold when the mold is closed, and determine whether the lengths of the guide pillars are different, so as to determine whether there is an abnormality in the movement of the guide pillars controlling the moving mold.

[0032] In this solution, various determinations are made through the mold opening image and the mold closing image to judge from different angles whether there are abnormalities such as asynchronous mold opening and closing or inclination of the parting surface between the moving mold and the fixed mold in the die-casting device. First, it is to timely detect abnormalities in the mold opening and closing and solve them in time to avoid affecting the quality of the formed workpieces and at the same time reduce waste of raw materials. Second, the diversified judgment improves the recognition accuracy of abnormal situations and timely discovers abnormal situations.

[0033] 2. In this solution, the storage module is set to store data during the use of the die-casting device. The model training module is set to generate and train an abnormal prediction model according to the data stored in the storage module when an abnormality occurs in the mold opening and closing. The abnormal prediction model predicts the duration until the next abnormal situation of the mold opening and closing based on the input mold opening image, mold closing image, and dynamic image, that is, the mold maintenance information. Before the next abnormal situation arrives during the mold opening and closing, a maintenance signal is generated in advance to remind the staff to increase the manual attention to the die-casting device or perform maintenance in advance to avoid the occurrence of abnormal situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is the logic block diagram of the first embodiment of a die-casting device suitable for automated production according to the present invention;

[0035] Figure 2 It is the logic block diagram of the second embodiment of a die-casting device suitable for automated production according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following is a more detailed description through specific embodiments:

[0037] Embodiment 1

[0038] A die-casting device suitable for automated production, as shown in the attached Figure 1 figure, includes:

[0039] A mold opening and closing control module for controlling the mold opening and closing of the forming structure. The forming structure includes a moving mold, a fixed mold, and a plurality of guide pillars;

[0040] An image acquisition module for acquiring a mold opening image and a mold closing image containing the forming structure after the mold opening and closing are in place; it is also used to acquire a dynamic image during the mold closing process, and the dynamic image contains the forming structure;

[0041] A mold analysis module for analyzing the parting surface between the moving mold and the fixed mold according to the mold opening image to judge whether there is an abnormal parting surface; it is also used to analyze the gap between the moving mold and the fixed mold according to the mold closing image to judge whether there is an abnormal mold closing.

[0042] A guiding analysis module is used to analyze the movement information of multiple guide pillars according to the mold closing image to determine whether there is abnormal movement; it is also used to analyze the dynamic information of multiple guide pillars according to the dynamic image to determine whether the guide pillars have asynchronous movement. When asynchronous movement occurs, it is determined that there is abnormal movement.

[0043] A comprehensive analysis module is used to generate and send an abnormal signal according to one or more of parting surface abnormality, mold closing abnormality, and movement abnormality.

[0044] Specifically:

[0045] This solution controls the existing die-casting device. The existing die-casting device includes a pouring system, a molding system, etc. Among them, the molding structure includes a moving mold, a fixed mold, a guiding structure, and a pushing structure. During use, the guiding structure controls the moving mold to move along a specified direction, so that the moving mold and the fixed mold are abutted against each other to realize mold closing and form a die-casting molding mold. Molten metal is injected into the die-casting molding mold through the pouring system. After the molten metal cools, the guiding structure controls the moving mold to move in the reverse direction to open the mold, and the pushing mechanism pushes the workpiece formed by cooling the molten metal out of the moving mold and the fixed mold.

[0046] The molding structure in this solution is the same as the molding structure in the existing die-casting device. The molding structure includes a moving mold, a fixed mold, and multiple guide pillars. The moving mold is sleeved on multiple guide pillars, that is, multiple guide pillars pass through the same moving mold. In this embodiment, the number of guide pillars is four, and the four guide pillars are rectangularly distributed. Driven by power, the moving mold moves along the axial direction of the guide pillar until the parting surface of the moving mold abuts against the parting surface of the fixed mold, realizing the mold closing of the moving mold and the fixed mold. The cavity between the moving mold and the fixed mold is the position where the die-casting workpiece is located. The mold opening and closing control module is used to control the mold opening and closing of the molding structure. It is a system installed in the existing die-casting device and is used to control the mold opening and closing of the moving mold and the fixed mold through the guiding structure. The die-casting device is a very mature existing technology, so its structure, principle, and control process will not be elaborated here.

[0047] The number of image acquisition modules is at least two. The image acquisition modules are respectively oriented towards the moving mold and the fixed mold. The image acquired by the image acquisition module oriented towards the moving mold contains the parting surface of the moving mold, and the image acquired by the image acquisition module oriented towards the fixed mold contains the parting surface of the fixed mold. Taking the axial direction of the guide pillar as horizontal as an example, it is defined that the moving mold is on the left and the fixed mold is on the right. The moving mold moves to the right until it abuts against the fixed mold. At this time, the mold closing is in place, and this process is the mold closing process; the moving mold moves to the left until the moving mold is reset. At this time, the mold opening is in place, and this process is the mold opening process. In this embodiment, the image acquisition module uses a camera, and the number of image acquisition modules is two. One image acquisition module is located on the left and is oriented towards the fixed mold. The image it captures contains the parting surface of the fixed mold, the guide pillar, and the side surface of the moving mold away from the fixed mold; the other image acquisition module is located on the right and is oriented towards the moving mold. The image it captures contains the parting surface of the moving mold and the side surface of the fixed mold away from the moving mold.

[0048] The image acquisition module is used to acquire the mold-closing image containing the formed structure after the mold is closed in place; it is also used to acquire the mold-opening image containing the formed structure after the mold is opened in place; and it is also used to acquire the dynamic image containing the formed structure during the mold-closing process. The mold-closing image is the image acquired by the two image acquisition modules at the moment when the mold is closed in place, the mold-opening image is the image acquired by the two image acquisition modules at the moment when the mold is opened in place, and the dynamic image is the continuous image acquired by the image acquisition module facing the fixed mold during the mold-closing process.

[0049] The mold analysis module presets the standard parting surfaces of the moving mold and the fixed mold, as well as the matching error. The standard parting surface is the shape of the parting surface set in advance under normal conditions, and the matching error is the allowable error when presetting the matching parting surface. The mold analysis module is used to perform image recognition based on the mold-opening image to mark the parting surface of the moving mold and the parting surface of the fixed mold; it is also used to call the standard parting surface of the moving mold, calculate the shape similarity between the standard parting surface of the moving mold and the marked parting surface of the moving mold, and determine whether the shape similarity is within the matching error. If it is, it is determined that the standard parting surface of the moving mold and the marked parting surface of the moving mold match, otherwise it is determined that the standard parting surface of the moving mold and the marked parting surface of the moving mold do not match. Similarly, the standard parting surface of the fixed mold and the marked parting surface of the fixed mold are matched. When any standard parting surface of the moving mold and the fixed mold does not match the marked parting surface, that is, when it is considered that the standard parting surface does not match the parting surface in the mold-opening image, it is determined that there is an abnormality in the parting surface.

[0050] The calculation of the shape similarity adopts the existing image similarity calculation method. When determining whether the shape similarity is within the matching error, the shape similarity and the matching error are percentages not greater than 1. The shape similarity is δ, and the matching error is σ. When 1 - δ ≤ σ, it is determined whether the shape similarity is within the matching error. Within the error range, if the parting surfaces of the moving mold and the fixed mold in the mold-opening image match the standard parting surfaces, it means that there is no cumulative fatigue damage in the moving mold and the fixed mold, and at the same time, there is no out-of-sync movement in the moving mold, so as to realize the identification of abnormal conditions during the mold-closing and mold-opening processes of the die-casting device.

[0051] The mold analysis module is also preset with a gap threshold, which is used to determine whether the gap between the movable mold and the movable mold parting surface is within an acceptable error when the mold is in place. The mold analysis module is also used to perform image recognition based on the mold closing image, mark the parting surface edges of the movable mold and the fixed mold, calculate the angle of each parting surface edge, and divide the edge groups according to the angle of the parting surface edge. The parting surface edges with the same angle are in the same group. The mold analysis module is also used to calculate the number of parting surface edges in the same edge group. When the number is greater than two, calculate the straight-line distance between the parting surface edges in the group, select the parting surface edge corresponding to the minimum straight-line distance, and eliminate the remaining parting surface edges in the group. The number of parting surface edges in the same edge group is not greater than two. The mold analysis module is also used to calculate the straight-line distance of the parting surface edges in each edge group, and determine whether the straight-line distance corresponding to each edge group is greater than the gap threshold one by one. When the straight-line distance corresponding to any edge group is greater than the gap threshold, it is determined that a mold closing abnormality occurs.

[0052] The parting surface between the movable die and the fixed die is a key component of die casting. The gap threshold is used to determine whether the parallelism between the parting surfaces of the movable die and the fixed die meets the requirements, thereby identifying abnormal situations in the opening and closing process of the die casting device.

[0053] The guide analysis module is used to perform image recognition based on the mold closing image and analyze the movement information of each guide pin, that is, to calculate the distance that the movable mold moves along the axial direction of the guide pin when the mold is in place. The movement information includes the movement distances corresponding to the four guide pins.

[0054] The guide analysis module is also used to calculate the guide column fluctuation range based on the movement information. The guide column fluctuation range calculation formula is as follows:

[0055]

[0056] In the formula, n is the number of guide pillars, the value of n is 4, x n is the movement information of the four guide pillars, α is the control accuracy of the preset die-casting device opening and closing, and the fluctuation range of the guide pillars is [Q min ,Q max ].

[0057] The guide analysis module is also used to determine whether the movement information of each guide post is within the fluctuation range of the guide post. When the movement information of any guide post is outside the fluctuation range of the guide post, it is determined that the movement is abnormal. The guide analysis module analyzes and determines the movement of the movable mold along the guide post, thereby determining whether the movable mold is tilted during mold closing and whether the movable mold moves synchronously, thereby realizing the identification of abnormal conditions during the opening and closing process of the die-casting device.

[0058] The guiding analysis module is also used to perform image recognition based on the dynamic image and analyze the dynamic information of each guide post at the same moment. Specifically, the dynamic image is continuous, and the guiding analysis module is used to randomly extract image frames from the dynamic image and analyze the dynamic information of each guide post in the same image frame. The dynamic information is the distance that the moving die moves along the axial direction of each guide post in the image frame. In this embodiment, the dynamic fluctuation range adopts the same calculation formula as the guide post fluctuation range, and the dynamic fluctuation range is calculated according to the dynamic information. Each image frame corresponds to a dynamic fluctuation range. The guiding analysis module is also used to determine whether the dynamic information of each guide post in the same image frame is within the dynamic fluctuation range, and judge each extracted image frame one by one. When the dynamic information of any guide post is outside the dynamic fluctuation range, it is determined that the guide post has non-synchronous movement.

[0059] At the same time, the setting of the guiding analysis module also recognizes the movement of the moving die during the mold closing process. Through the movement of the moving die on each guide post, it is judged again whether the moving die is tilted and whether the movement of the moving die is synchronous during mold closing, so as to realize the recognition of abnormal conditions during the mold opening and closing process of the die-casting device. In this solution, the dynamic image is the image collected during the mold closing process. Using the mold closing process instead of the mold opening process reduces the interference of the die-cast molded workpiece on image recognition, thereby improving the accuracy of abnormal condition recognition.

[0060] The comprehensive analysis module is used to generate an abnormal signal according to one or more of the parting surface abnormality, mold closing abnormality, and movement abnormality. The abnormal signal is marked differently according to the differences in the parting surface abnormality, mold closing abnormality, and movement abnormality. The type of abnormality can be identified through the abnormal signal. The comprehensive analysis module is also used to send the abnormal signal to the working end used by the staff. The working end can be a computer set in the maintenance center or a mobile device set for the staff, which is determined according to the actual situation.

[0061] Adopting this solution, multiple determinations are made through the mold opening image, mold closing image, and dynamic image, and it is judged from different angles whether there are abnormal conditions such as non-synchronous mold opening and closing or inclination of the parting surface between the moving die and the fixed die of the die-casting device. First, the abnormal conditions during the mold opening and closing process can be discovered in time and solved in time to avoid affecting the quality of the molded workpiece and reduce the waste of raw materials at the same time. Second, the diversified judgment improves the recognition accuracy of abnormal conditions and discovers abnormal conditions in time.

[0062] Embodiment 2

[0063] The difference between this embodiment and Embodiment 1 is as follows:

[0064] In a die-casting device suitable for automated production, the image acquisition module is also used to collect the finished product image after mold opening. The finished product image includes the die-cast molded workpiece.

[0065] As shown in the appendix Figure 2 It also includes:

[0066] A storage module for storing mold opening images, mold closing images, and dynamic images. The mold opening images, mold closing images, and dynamic images have image attributes, which include image acquisition time, acquisition device identifier (i.e., the identifier of the image acquisition module), and installation device identifier (i.e., the identifier of the die-casting device). When used for maintenance, the storage module stores the mold opening images, mold closing images, and dynamic images as historical records, and stores the mold opening images, mold closing images, and dynamic images collected after maintenance as current records. By triggering an update, the mold opening images, mold closing images, and dynamic images generated during the previous maintenance process are used as current records, which facilitates the calling of data during analysis and enables rapid data feedback.

[0067] A model training module for, when an anomaly signal is generated by the comprehensive analysis module, calling the mold opening images, mold closing images, dynamic images, and anomaly signals stored in the storage module to generate and train an anomaly prediction model. Specifically, the model training module is preset with a neural network model. When the comprehensive analysis module generates an anomaly signal, it indicates that operations such as maintenance and calibration need to be performed on the die-casting device. The anomaly signal includes an anomaly device identifier (i.e., the identifier of the die-casting device where the anomaly occurs), anomaly type, and anomaly time. The anomaly types include parting surface anomaly, mold closing anomaly, and movement anomaly.

[0068] The model training module is used to call the mold opening images, mold closing images, dynamic images, and anomaly signals in the current record, and train the neural network model based on the current record and the anomaly signal to generate an anomaly prediction model. Each anomaly prediction model corresponds to a die-casting device. Therefore, when calling the current record, the current record of the same die-casting device is called according to the anomaly device identifier and the installation device identifier, and an anomaly prediction model exclusive to the die-casting device is trained to improve the prediction accuracy of the anomaly prediction model.

[0069] The model trigger module is used to determine whether the die-cast product is qualified based on the finished product image. Specifically, the model trigger module is used to perform image recognition on the finished product image and determine whether the die-cast product is qualified based on the image recognition result. That is, it is determined whether problems such as surface pores and burrs appear on the die-cast product. If so, the die-cast product is determined to be unqualified. During the die-casting process, occasionally the product may be unqualified due to manual, material, or other problems. This situation occurs occasionally and does not represent an anomaly in the die-casting device.

[0070] The model prediction module is used to call the anomaly prediction model and predict the die maintenance information according to the anomaly prediction model when the model trigger module determines that the die-casting product is unqualified. Specifically, the anomaly prediction model is used to input one or more of the mold opening image, mold closing image, and dynamic image, and output the predicted anomaly prediction type and anomaly prediction time (i.e., die maintenance information). The model prediction module is used to call the anomaly prediction model when the model trigger module determines that the die-casting product is unqualified, and call the mold opening image, mold closing image, and dynamic image that are the closest to the current time in the current record from the storage module; the model prediction module is also used to obtain the die maintenance information output by the anomaly prediction model according to the called current record. The possible anomalies of the die-casting device are predicted through the anomaly prediction model, and the time when the anomaly appears is predicted, so as to carry out prevention in advance.

[0071] The comprehensive analysis module is preset with a maintenance threshold, which is used to judge whether to increase the manual attention to the die-casting equipment or to perform early maintenance on the die-casting equipment.

[0072] The comprehensive analysis module is also used to judge whether the die maintenance information is less than the maintenance threshold. When the die maintenance information is less than the maintenance threshold, a maintenance signal is generated and sent. Specifically, the comprehensive analysis module is used to call the corresponding mold opening image, mold closing image, and dynamic image when the model prediction module generates the die maintenance information, obtain the corresponding installation equipment identifier according to the mold opening image, mold closing image, and dynamic image, generate a maintenance signal according to the installation equipment identifier and the die maintenance information, and send the maintenance signal to the working end. This helps the staff quickly locate the die-casting device, abnormal time, and abnormal type where abnormal conditions may occur, facilitating the staff to quickly carry out the maintenance and management of the device.

[0073] Adopting this solution, through the prediction of the anomaly prediction model, a maintenance signal is generated in advance before the next mold opening and closing when the abnormal situation arrives, and the maintenance signal is used to remind the staff to increase the manual attention to the die-casting device or perform early maintenance to avoid the occurrence of abnormal situations.

[0074] The above are only embodiments of the present invention. Specific structures and common knowledge such as characteristics that are well-known in the art are not described in detail herein. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention pertains before the filing date or the priority date, can learn all the prior art in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, complete and implement this solution in combination with their own abilities. Some typical well-known structures or well-known methods should not become obstacles for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. A die-casting device applicable to automated production, comprising: A mold opening and closing control module for controlling the opening and closing of a molding structure, the molding structure including a moving mold, a fixed mold, and a plurality of guide pillars; Characterized in that it further includes: An image acquisition module for acquiring an opening mold image and a closing mold image containing the molding structure after the mold opening and closing are in place; A mold analysis module for analyzing the parting surface between the moving mold and the fixed mold according to the opening mold image to determine whether there is an abnormality in the parting surface; and also for analyzing the gap between the moving mold and the fixed mold according to the closing mold image to determine whether there is a closing mold abnormality; A guiding analysis module for analyzing the movement information of a plurality of guide pillars according to the closing mold image to determine whether there is a movement abnormality; A comprehensive analysis module for generating and sending an abnormal signal according to one or more of the parting surface abnormality, closing mold abnormality, and movement abnormality.

2. The die-casting device applicable to automated production according to claim 1, wherein: The number of image acquisition modules is at least two. The image acquisition modules are respectively oriented towards the moving mold and the fixed mold. The image acquired by the image acquisition module oriented towards the moving mold contains the parting surface of the moving mold, and the image acquired by the image acquisition module oriented towards the fixed mold contains the parting surface of the fixed mold.

3. A die-casting device applicable to automated production according to claim 2, characterized in that: The mold analysis module presets the standard parting surface of the moving mold and the fixed mold. The mold analysis module is used to match the standard parting surface with the opening mold image. When the standard parting surface does not match the parting surface in the opening mold image, it is determined that there is an abnormality in the parting surface.

4. A die-casting device applicable to automated production according to claim 3, characterized in that: The mold analysis module is also used for image recognition according to the closing mold image, marking the parting surface edges of the moving mold and the fixed mold, dividing the edge lines into edge line groups according to the angles of the parting surface edge lines, and the number of parting surface edge lines in the same edge line group is not greater than two; It is also used to calculate the straight-line distance of the parting surface edge lines in each edge line group, and determine whether the straight-line distance is greater than a preset gap threshold. When the straight-line distance is greater than the gap threshold, it is determined that there is a closing mold abnormality.

5. A die-casting device applicable to automated production according to claim 4, characterized in that: The guiding analysis module is used for image recognition according to the closing mold image, analyzing the movement information of each guide pillar, calculating the guide pillar fluctuation range according to the movement information, and determining whether the movement information of each guide pillar is within the guide pillar fluctuation range. When the movement information of any one guide pillar is outside the guide pillar fluctuation range, it is determined that there is a movement abnormality.

6. A die-casting device applicable to automated production according to any one of claims 1-5, characterized in that: The image acquisition module is also used for acquiring dynamic images during the closing mold process, and the dynamic images contain the molding structure; The guiding analysis module is also used for analyzing the dynamic information of a plurality of guide pillars according to the dynamic images to determine whether the guide pillars have asynchronous movement. When asynchronous movement occurs, it is determined that there is a movement abnormality.

7. A die-casting device applicable to automated production according to claim 6, characterized in that: The guiding analysis module is also used for image recognition according to the dynamic images, analyzing the dynamic information of each guide pillar at the same moment, calculating the dynamic fluctuation range according to the dynamic information, and determining whether the dynamic information of each guide pillar is within the dynamic fluctuation range. When the dynamic information of any one guide pillar is outside the dynamic fluctuation range, it is determined that the guide pillars have asynchronous movement.

8. A die-casting device applicable to automated production according to claim 7, characterized in that, It further includes: A storage module for storing the opening mold image, the closing mold image, and the dynamic images; A model training module for, when the comprehensive analysis module generates an abnormal signal, calling the opening mold image, the closing mold image, the dynamic images, and the abnormal signal stored in the storage module to generate and train an abnormal prediction model; A model prediction module for calling the abnormal prediction model and predicting mold maintenance information according to the abnormal prediction model.

9. A die-casting device applicable to automated production according to claim 8, characterized in that: The comprehensive analysis module is preset with a maintenance threshold, and the comprehensive analysis module is also used to determine whether the die maintenance information is less than the maintenance threshold. When the die maintenance information is less than the maintenance threshold, a maintenance signal is generated and sent.

10. A die-casting device applicable to automated production according to claim 9, characterized in that: The image acquisition module is also used to acquire the finished product image after the mold is opened; It also includes: The model trigger module is used to judge whether the die-casting product is qualified according to the finished product image; The model prediction module is used to call the abnormal prediction model when the model trigger module determines that the die-casting product is unqualified.

Images