CN117094990B - Defect detection method and system for composite stamping die - Google Patents

Abstract

The disclosure provides a defect detection method and system for a composite stamping die, and relates to a die quality detection technology, wherein the method comprises the following steps: collecting a three-dimensional simulation model of a mold, wherein the three-dimensional simulation model comprises a mold opening three-dimensional model and a mold closing three-dimensional model; matching the standard mold opening three-dimensional model with the standard mold closing three-dimensional model; slicing the standard mold-opening three-dimensional model, wherein a plurality of mold-opening standard slices of the model are provided with a plurality of groups of mold-opening characteristic region standard coordinates; generating a plurality of model closing standard slices, a plurality of model opening slices and a plurality of model closing slices; and comparing the coordinates of the multi-group die clamping characteristic region with the standard coordinates of the multi-group die clamping characteristic region to generate a die clamping defect region, generating a die opening defect region and detecting a defect entity. The method can solve the technical problem that the existing method for detecting the defects of the die directly through image processing has lower detection efficiency, and can save the positioning time of the die defect detection position, thereby improving the die defect detection efficiency.

Description

Defect detection method and system for composite stamping die

Technical Field

The present disclosure relates to die quality inspection techniques, and more particularly, to a defect inspection method and system for a composite stamping die.

Background

The composite press mold means a mold capable of performing two or more steps on one set of mold, for example: compared with the traditional single-procedure stamping die, the lap joint type composite die, the co-stamping type composite die, the turning type composite die and the like have the advantages of high production efficiency, low energy consumption and the like, and are widely applied to the stamping industry.

The existing composite stamping die defect detection method generally comprises the steps of collecting images of dies and carrying out defect detection analysis on the die images by utilizing an image processing algorithm, wherein the defect detection method is low in die defect detection efficiency because the defect detection positions are not clear and all the die images need to be detected, and meanwhile, the composite stamping die is complex in structure and needs a large amount of detection time.

The existing method for detecting the defects of the composite stamping die directly through image processing has the following defects: because the structure of the composite stamping die is complex, the detection efficiency is low.

Disclosure of Invention

Therefore, in order to solve the above technical problems, the technical solution adopted in the embodiments of the present disclosure is as follows:

the defect detection method for the composite stamping die comprises the following steps of: collecting a three-dimensional simulation model of a composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model; according to the model of the composite stamping die, matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library; slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of mold opening standard slices of the model, wherein the mold opening standard slices of the model are provided with a plurality of groups of mold opening characteristic region standard coordinates; slicing the standard die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing standard slices, wherein the model die-closing standard slices have standard coordinates of a multi-group die-closing characteristic region; slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-opening slices, wherein the plurality of model die-opening slices are provided with a plurality of groups of die-opening characteristic region coordinates; slicing the die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly slices, wherein the model die assembly slices have multiple groups of die assembly characteristic region coordinates; and comparing the multi-group die clamping characteristic region coordinates with the multi-group die clamping characteristic region standard coordinates to generate a die clamping defect region, comparing the multi-group die opening characteristic region coordinates with the multi-group die opening characteristic region standard coordinates to generate a die opening defect region, and detecting a defect entity based on the die clamping defect region and the die opening defect region.

A defect detection system for a composite stamping die, comprising: the three-dimensional simulation model acquisition module is used for acquiring a three-dimensional simulation model of the composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model; the standard three-dimensional model matching module is used for matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library according to the model of the composite stamping die; the model mold opening standard slice generation module is used for slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model mold opening standard slices, wherein the plurality of model mold opening standard slices are provided with a plurality of groups of mold opening characteristic region standard coordinates; the model die assembly standard slice generation module is used for slicing the standard die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly standard slices, wherein the plurality of model die assembly standard slices are provided with a plurality of groups of die assembly characteristic region standard coordinates; the die-opening and die-cutting sheet generating module is used for slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of die-opening and die-cutting sheets, wherein the die-opening and die-cutting sheets are provided with a plurality of groups of die-opening characteristic region coordinates; the model die-closing slice generation module is used for slicing the die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing slices, wherein the plurality of model die-closing slices are provided with a plurality of groups of die-closing characteristic region coordinates; the defect entity detection module is used for comparing the coordinates of the multi-group die clamping characteristic areas with the standard coordinates of the multi-group die clamping characteristic areas to generate die clamping defect areas, comparing the coordinates of the multi-group die opening characteristic areas with the standard coordinates of the multi-group die opening characteristic areas to generate die opening defect areas, and detecting the defect entity based on the die clamping defect areas and the die opening defect areas.

By adopting the technical method, compared with the prior art, the technical progress of the present disclosure has the following points:

(1) The method can solve the technical problem that the existing method for detecting the defects of the composite stamping die directly through image processing has lower detection efficiency, and firstly, the composite stamping die is subjected to three-dimensional simulation modeling in two states of die opening and die closing through a digital twin technology to obtain a die opening three-dimensional model and a die closing three-dimensional model; then matching a standard open-die three-dimensional model with a standard closed-die three-dimensional model from a standard three-dimensional model library according to the model of the composite stamping die; obtaining a first preset interval, slicing the standard mold opening three-dimensional model from the top surface of an upper die holder to the bottom surface of a lower die holder according to the first preset interval to obtain a plurality of mold opening standard slices, wherein the mold opening standard slices are provided with a plurality of groups of mold opening characteristic region standard coordinates; the standard die-closing three-dimensional model, the die-opening three-dimensional model and the die-closing three-dimensional model are sequentially sliced according to the first preset interval to obtain a plurality of model die-closing standard slices, a plurality of model die-opening slices and a plurality of model die-closing slices, wherein the plurality of model die-closing standard slices have multi-set die-closing characteristic region standard coordinates, the plurality of model die-opening slices have multi-set die-opening characteristic region coordinates and the plurality of model die-closing slices have multi-set die-closing characteristic region coordinates; and comparing the multi-group die clamping characteristic region coordinates with the multi-group die clamping characteristic region standard coordinates to generate a die clamping defect region, comparing the multi-group die opening characteristic region coordinates with the multi-group die opening characteristic region standard coordinates to generate a die opening defect region, and finally detecting a defect entity based on the die clamping defect region and the die opening defect region. The defect position of the composite stamping die is accurately positioned by utilizing the model slicing concept, so that the positioning time of the defect detection position of the die can be saved, the efficiency and the accuracy of the defect detection position positioning are improved, and the defect detection efficiency of the composite stamping die is improved.

(2) By utilizing the digital twin technology to carry out three-dimensional simulation modeling on the composite stamping die, the digital twin technology has the advantages of interoperability, instantaneity, fidelity and the like, so that the efficiency and accuracy of model slice comparison can be improved, and the efficiency and quality of defect detection of the composite stamping die can be improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings that are used in the description of the embodiments will be briefly described below.

Fig. 1 is a schematic flow chart of a defect detection method for a composite stamping die;

fig. 2 is a schematic flow chart of detecting a defect entity based on a mold closing defect area and a mold opening defect area in a defect detection method for a composite stamping mold;

fig. 3 is a schematic structural diagram of a defect detection system for a composite stamping die according to the present application.

Detailed Description

The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments in this disclosure are intended to be within the scope of this disclosure.

Based on the above description, as shown in fig. 1, the present disclosure provides a defect detection method for a composite stamping die, including:

the composite stamping die is a die capable of completing a plurality of stamping processes, has the advantages of high production efficiency, low energy consumption and the like, and is complex in structure and difficult to manufacture. The method is used for detecting the defects of the composite stamping die to achieve the aim of improving the die defect detection efficiency, and the method is specifically implemented in a defect detection system for the composite stamping die.

Collecting a three-dimensional simulation model of a composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model;

in the embodiment of the application, firstly, modeling data acquisition is performed on a composite stamping die to be subjected to defect detection in two states of die opening and die closing, wherein the modeling data comprise data of structures, shapes, sizes, materials and the like of all parts of the composite stamping die, and the composite stamping die comprises parts such as an upper die base, an upper die base plate, an upper clamping plate, a stripping base plate, a stripping plate, a lower die plate, a lower base plate and a lower die base.

Then based on a digital twin technology, according to modeling data of the composite stamping die in two states of die opening and die closing, utilizing three-dimensional simulation modeling software to perform three-dimensional simulation modeling on the composite stamping die, wherein the digital twin technology refers to a process of performing virtual modeling on the die in a digital form, and has the advantages of interoperability, expandability, instantaneity, fidelity, closed loop and the like; the commonly used three-dimensional simulation modeling software of the stamping die comprises Ansys Forming, DYNAFORM, AI-FORM and other software, and a person skilled in the art can select the adaptive simulation modeling software according to actual requirements; and generating an open three-dimensional model and a closed three-dimensional model of the composite stamping die.

By generating the mold opening three-dimensional model and the mold closing three-dimensional model, support is provided for the next step of model slicing operation and the subsequent model slicing comparison analysis.

According to the model of the composite stamping die, matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library;

in the embodiment of the application, firstly, a mold model of a composite stamping mold to be subjected to defect detection is obtained, then the mold model is input into a standard three-dimensional model library for model matching, the standard three-dimensional model library is a database for storing standard three-dimensional models of the composite stamping mold, a plurality of mold models and a plurality of mold standard three-dimensional models are stored in the database, the mold models and the mold standard three-dimensional models have a corresponding relationship, and the standard mold opening three-dimensional model and the standard mold closing three-dimensional model of the composite stamping mold are obtained. And by obtaining the standard mold opening three-dimensional model and the standard mold closing three-dimensional model, support is provided for next step of mold slicing comparison and obtaining a mold closing defect area and a mold opening defect area.

Slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of mold opening standard slices of the model, wherein the mold opening standard slices of the model are provided with a plurality of groups of mold opening characteristic region standard coordinates;

in this embodiment of the present application, a first preset pitch is obtained, where the first preset pitch is a pitch used for cutting a mold model, and a person skilled in the art may set according to an actual size of the mold model, for example: when the mold pattern length is 50 cm, the first preset spacing may be set to 5 cm.

And then slicing the standard open-die three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to the first preset interval to obtain a plurality of model open-die standard slices, wherein each model open-die standard slice has standard coordinates of an open-die characteristic region, and the open-die characteristic region refers to a region with a stamping function in each part of the composite stamping die, for example: a feeding punch, a forming punch, an inner guide sleeve and the like in the upper clamping plate; bending inserts, pressing and unloading inserts, guide pins and the like in the stripper plate; the standard coordinates of the mold opening feature region refer to the position coordinates of the mold opening feature region in the mold opening standard slice of the mold, and a plurality of groups of standard coordinates of the mold opening feature region of the mold opening standard slices of the mold are obtained. And by obtaining a plurality of groups of standard coordinates of the mold opening characteristic regions, support is provided for next mold opening three-dimensional model comparison and mold opening defect region generation.

Slicing the standard die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing standard slices, wherein the model die-closing standard slices have standard coordinates of a multi-group die-closing characteristic region;

in this embodiment of the present application, slicing the standard mold-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to the first preset interval, so as to obtain a plurality of mold-closing standard slices, where the plurality of mold-closing standard slices have multiple sets of mold-closing feature area standard coordinates, and the mold-closing standard slices have a corresponding relationship with a set of mold-closing feature area standard coordinates.

Slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-opening slices, wherein the plurality of model die-opening slices are provided with a plurality of groups of die-opening characteristic region coordinates;

in this embodiment of the present application, the open-mold three-dimensional model is sliced from the top surface of the upper die base to the bottom surface of the lower die base according to the first preset interval, so as to generate a plurality of open-mold slices of the model, where the plurality of open-mold slices of the model have a plurality of groups of open-mold feature region coordinates, and the open-mold slices of the model have a corresponding relationship with a group of open-mold feature region coordinates.

Slicing the die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly slices, wherein the model die assembly slices have multiple groups of die assembly characteristic region coordinates;

in this embodiment of the present application, slicing the mold-closing three-dimensional model from the top surface of the upper mold base to the bottom surface of the lower mold base according to the first preset interval, and generating a plurality of mold-closing slices, where the plurality of mold-closing slices have a plurality of sets of mold-closing feature area coordinates, and the mold-closing slices have a correspondence with a set of mold-closing feature area coordinates. The method provides support for accurately positioning the defect position of the composite stamping die in the next step by obtaining a plurality of groups of die opening characteristic region standard coordinates, a plurality of groups of die closing characteristic region standard coordinates, a plurality of groups of die opening characteristic region coordinates and a plurality of groups of die closing characteristic region coordinates.

And comparing the multi-group die clamping characteristic region coordinates with the multi-group die clamping characteristic region standard coordinates to generate a die clamping defect region, comparing the multi-group die opening characteristic region coordinates with the multi-group die opening characteristic region standard coordinates to generate a die opening defect region, and detecting a defect entity based on the die clamping defect region and the die opening defect region.

In one embodiment, the method further comprises:

extracting a first die clamping characteristic region standard coordinate set of a first die clamping standard section according to the multi-set die clamping characteristic region standard coordinates, and extracting a first die clamping characteristic region coordinate set of the first die clamping section from the multi-set die clamping characteristic region coordinates, wherein the first die clamping characteristic region coordinate set is the characteristic region coordinate with the largest overlapping coordinate number with the first die clamping characteristic region standard coordinate set in the first die clamping section;

counting the number of non-coincident coordinates and the number of coincident coordinates of the standard coordinate set of the first die-closing feature area and the coordinate set of the first die-closing feature area, calculating the ratio of the number of non-coincident coordinates to the total number of coordinates, and setting the ratio as the defect probability of the first die-closing feature area, wherein the total number of coordinates is equal to the sum of the number of non-coincident coordinates and the number of coincident coordinates;

in this embodiment of the present application, the coordinates of the multiple-unit mold clamping feature area and the standard coordinates of the multiple-unit mold clamping feature area are compared to generate a mold clamping defect area. Firstly, randomly extracting a standard coordinate of a combined die characteristic region from the standard coordinates of the combined die characteristic region to serve as a first standard coordinate set of the combined die characteristic region of a first combined die standard slice; then sequentially matching the first die closing feature region standard coordinate set with the multiple sets of die closing feature region coordinates, and counting the number of coincident coordinates of each set of matching results to obtain multiple sets of coincident coordinate number counting results; and then taking the combined die characteristic region coordinate with the largest number of the overlapped coordinates with the first combined die characteristic region standard coordinate set in the combined die characteristic region coordinate as a first combined die characteristic region coordinate set of a first combined die slice to obtain a first combined die characteristic region coordinate set.

Respectively counting the number of non-coincident coordinates and coincident coordinates in the first die closing feature area standard coordinate set and the first die closing feature area coordinate set to obtain the number of non-coincident coordinates and the number of coincident coordinates, adding and summing the number of non-coincident coordinates and the number of coincident coordinates, and taking the sum result as the total number of coordinates; and then setting the ratio of the non-coincident coordinate number to the total coordinate number as the defect probability of the first mold closing characteristic region, and obtaining the defect probability of the first mold closing characteristic region.

When the defect probability of the first die clamping characteristic region is greater than or equal to a defect probability threshold value, adding a non-coincident coordinate distribution region of the first die clamping characteristic region of the first die clamping slice into the die clamping defect region;

in one embodiment, the method further comprises:

performing precision calibration by traversing the standard coordinates of the multi-group die clamping characteristic areas to generate a plurality of characteristic area precision thresholds, wherein the characteristic area precision thresholds are in one-to-one correspondence with the standard coordinates of the die clamping characteristic areas;

in one embodiment, the method further comprises:

traversing the standard coordinates of the multi-combination die characteristic area to obtain a first characteristic area element type and a second characteristic area element type until an Nth characteristic area element type;

In the embodiment of the application, first, element information extraction is sequentially performed on the standard coordinates of the multi-set mold clamping characteristic region, wherein the elements refer to parts corresponding to the standard coordinates of the mold clamping characteristic region; and obtaining the first characteristic area element type, the second characteristic area element type and the N characteristic area element type.

Traversing the first characteristic area element type and the second characteristic area element type until the Nth characteristic area element type is subjected to punching control precision analysis, and generating a first characteristic area punching precision set, and a second characteristic area punching precision set until the Nth characteristic area punching precision set;

in one embodiment, the method further comprises:

searching a plurality of first characteristic area stamping control record data by taking the first characteristic area element type as an area constraint condition, wherein the plurality of first characteristic area stamping control record data have a plurality of stamping control precision record values;

clustering and grouping the stamping control precision record values according to the precision consistency deviation to generate a plurality of groups of stamping control precision record values;

removing groups with the number of stamping control precision recorded values smaller than or equal to the threshold value of the number of recorded values from the plurality of groups of stamping control precision recorded values, and generating a stamping precision set of the first characteristic region;

And traversing the second characteristic area element type until the Nth characteristic area element type is subjected to stamping control precision analysis, and generating a second characteristic area stamping precision set until the Nth characteristic area stamping precision set.

In this embodiment of the present application, the first feature area element type, the second feature area element type, and up to the nth feature area element type are analyzed for stamping control accuracy, first, the first feature area element type is used as an area constraint condition to perform historical stamping control search, and a plurality of first feature area stamping control record data are obtained, where the stamping control record data are control data in a first feature area historical stamping process, and the plurality of first feature area stamping control record data have a plurality of stamping control accuracy record values, where the first feature area stamping control record data and the stamping control accuracy record values have a corresponding relationship, where the stamping control accuracy is used to characterize the stamping control accuracy, and the greater the stamping control accuracy is, the higher the stamping control accuracy is.

Setting a precision consistency deviation, wherein the precision consistency deviation can be set by a person skilled in the art according to actual numerical values of a plurality of stamping control precision recorded values, for example: setting the precision consistency deviation to be 1%, namely gathering stamping control precision record values with the precision deviation within 1% into one category, for example: the precision values with the precision between 90% and 91% are gathered into one class. And clustering and grouping the stamping control precision record values according to the precision consistency bias to obtain a plurality of groups of stamping control precision record values.

And sequentially carrying out numerical statistics on each group of stamping control precision record values to obtain the number of each group of record values in the stamping control precision record values. And obtaining a recorded value quantity threshold, wherein the recorded value quantity threshold can be set according to actual demands, judging the quantity of each recorded value in a plurality of groups of stamping control precision recorded values according to the recorded value quantity threshold, eliminating groups with the recorded value quantity smaller than or equal to the recorded value quantity threshold in the plurality of groups of stamping control precision recorded values, and forming a stamping precision set of a first characteristic area according to the remaining stamping control precision recorded values.

The record value quantity threshold is set to screen a plurality of groups of stamping control precision record values, accidental values with lower occurrence frequency in the stamping control precision record values can be removed, so that the accuracy of obtaining the stamping precision set of the first characteristic region can be improved, and the accuracy of setting the precision threshold of the characteristic region is further improved.

And then sequentially carrying out stamping control precision analysis on the second characteristic region element types to the Nth characteristic region element types by using the same method to obtain a second characteristic region stamping precision set to the Nth characteristic region stamping precision set.

Traversing the first characteristic region stamping precision set, respectively extracting precision maximum values from the second characteristic region stamping precision set to the Nth characteristic region stamping precision set, and adding the precision maximum values into the characteristic region precision threshold values.

In this embodiment of the present application, extracting, sequentially, the precision maximum values in the first feature area stamping precision set and the second feature area stamping precision set up to the nth feature area stamping precision set, to obtain a plurality of stamping precision maximum values, and using the plurality of stamping precision maximum values as feature area precision thresholds to obtain a plurality of feature area precision thresholds. The feature region precision thresholds and the die closing feature region standard coordinates have a one-to-one correspondence.

In one embodiment, the method further comprises:

extracting a maximum size deviation of the non-coincident coordinate distribution region of the first mold closing feature region of the first mold closing slice when the defect probability of the first mold closing feature region is greater than or equal to the defect probability threshold;

when the maximum size deviation is larger than a characteristic region precision threshold value of the first die clamping characteristic region, adding a non-coincident coordinate distribution region of the first die clamping characteristic region of the first die clamping slice into the die clamping defect region;

And when the maximum size deviation is smaller than or equal to a characteristic region precision threshold value of the first die clamping characteristic region, performing defect detection qualification identification on the first die clamping characteristic region of the first die clamping section.

In this embodiment of the present application, first, a defect probability threshold is obtained, where the defect probability threshold can be set by a person skilled in the art according to an actual defect detection accuracy requirement, where the higher the defect detection accuracy requirement is, the smaller the defect probability threshold is, for example: the defect probability threshold was set to 3%. And judging the defect probability of the first die closing feature region according to the defect probability threshold, and extracting the maximum size deviation of the non-coincident coordinate distribution region of the first die closing feature region of the first die closing slice when the defect probability of the first die closing feature region is larger than or equal to the defect probability threshold. Judging the maximum size deviation according to the characteristic region precision threshold value of the first die closing characteristic region, and marking a non-coincident coordinate distribution region of the first die closing characteristic region of the first die closing slice as a die closing defect region when the maximum size deviation is larger than the characteristic region precision threshold value of the first die closing characteristic region; and when the maximum size deviation is smaller than or equal to a characteristic region precision threshold value of the first die clamping characteristic region, performing defect detection qualification identification on the first die clamping characteristic region of the first die clamping section.

And when the defect probability of the first die clamping characteristic region is smaller than the defect probability threshold, performing defect detection qualification identification on the first die clamping characteristic region of the first die clamping slice.

In this embodiment of the present application, when the defect probability of the first mold closing feature area is smaller than the defect probability threshold, the first mold closing feature area is characterized to meet a quality detection requirement, and defect detection qualification identification is performed on the first mold closing feature area of the first mold closing slice, that is, the first mold closing slice is characterized to be a defect detection qualification area. And then comparing the coordinates of the multiple groups of mold opening characteristic areas with the standard coordinates of the multiple groups of mold opening characteristic areas by using the same method for generating the mold closing defect areas, so as to generate the mold opening defect areas. By generating the mold closing defect area and the mold opening defect area, support is provided for accurate positioning of the defect position of the next step.

In one embodiment, the method further comprises:

when the number of the mold closing defect areas is equal to 0, performing slice defect detection on the standard mold closing three-dimensional model and the mold closing three-dimensional model according to a second preset interval, wherein the second preset interval is one half of the first preset interval;

When the number of the open mold defect areas is equal to 0, performing slice defect detection on the standard open mold three-dimensional model and the open mold three-dimensional model according to a second preset interval, wherein the second preset interval is one half of the first preset interval;

when the M preset interval is smaller than or equal to a preset interval threshold value, if the number of the mold closing defect areas is equal to 0, generating a mold closing defect detection qualified mark;

and when the M preset distance is smaller than or equal to a preset distance threshold value, if the number of the open-mould defect areas is equal to 0, generating an open-mould defect detection qualified mark.

In this embodiment of the present application, the number of the mold closing defect areas and the mold opening defect areas is counted, when the number of the mold closing defect areas is equal to 0, the defect detection accuracy is low and the defect detection result is inaccurate due to the fact that the first preset interval is set too large, the second preset interval is obtained according to the first preset interval, the second preset interval is one half of the first preset interval, and then the slice defect detection is performed on the standard mold closing three-dimensional model and the mold closing three-dimensional model according to the second preset interval; and when the number of the open mold defect areas is equal to 0, performing slice defect detection on the standard open mold three-dimensional model and the open mold three-dimensional model according to the second preset interval.

Obtaining a preset spacing threshold, where the preset spacing threshold can be set by a person skilled in the art according to an actual defect detection requirement, where the larger the actual defect detection requirement is, the smaller the preset spacing threshold is, for example: the preset spacing threshold is set to 1 millimeter. And when the number of the mold closing defect areas or the number of the mold opening defect areas for slice defect detection according to the second preset interval is equal to 0, performing preset interval iteration until an Mth preset interval is obtained, wherein N is an integer larger than 2, and the specific value of N can be set according to actual conditions. Judging the Mth preset interval according to the preset interval threshold, and if the Mth preset interval is smaller than or equal to the preset interval threshold and the number of the die closing defect areas is equal to 0, performing die closing defect detection qualification identification on the detected die closing feature areas; and when the M preset distance is smaller than or equal to the preset distance threshold value and the number of the mold opening defect areas is equal to 0, detecting qualified marks of the mold opening defects of the detected mold opening feature areas.

The accuracy and the efficiency of slice defect detection can be improved by sequentially reducing the preset spacing according to the proportion to detect the slice defects, so that the accuracy and the efficiency of defect position positioning can be improved.

As shown in fig. 2, in one embodiment, the method further comprises:

extracting a die closing defect slice sequence number set and a die opening defect slice sequence number set of the first characteristic region;

extracting a first slice clamping defect area and a second slice clamping defect area of adjacent serial number slices of the clamping defect slice serial number set;

counting the number of slice plane coincident coordinates of the first slice die-closing defect area and the second slice die-closing defect area, calculating the ratio of the number of slice plane coincident coordinates to the total number of slice plane coordinates of the defect area, and setting the ratio as a defect area azimuth coincident coefficient, wherein the total number of slice plane coordinates of the defect area is equal to the sum of the number of slice plane coincident coordinates and the number of slice plane non-coincident coordinates;

when the azimuth consistency coefficient of the defect area is larger than or equal to an azimuth consistency coefficient threshold value, communicating the first slice mold-closing defect area and the second slice mold-closing defect area, and growing the mold-closing defect area to generate a mold-closing defect area expansion result;

performing defect region growth according to the open-mould defect slice sequence number set to generate an open-mould defect region expansion result;

and detecting a defect entity according to the mold closing defect area expansion result and the mold opening defect area expansion result.

In this embodiment of the present application, defect entity detection is performed according to the mold closing defect area and the mold opening defect area, and first, a mold closing defect slice sequence number set and a mold opening defect slice sequence number set of the first feature area are extracted to obtain a mold closing defect slice sequence number set and a mold opening defect slice sequence number set. And randomly extracting a first slice clamping defect area and a second slice clamping defect area from the clamping defect slice sequence number set, wherein the first slice clamping defect area and the second slice clamping defect area are slices with adjacent sequence numbers.

And counting the number of the slice plane coincident coordinates and the slice plane non-coincident coordinates of the first slice die-closing defect area and the second slice die-closing defect area to obtain the number of the slice plane coincident coordinates and the number of the slice plane non-coincident coordinates, adding and summing the number of the slice plane coincident coordinates and the number of the slice plane non-coincident coordinates, and taking the sum of the two as the total number of the slice plane coordinates of the defect area. And then calculating the ratio of the number of the coincident coordinates of the slice plane to the total number of the coordinates of the slice plane in the defect area, and setting the ratio of the number of the coincident coordinates of the slice plane to the total number of the coordinates of the slice plane as a direction coincidence coefficient of the defect area.

And acquiring an azimuth consistency coefficient threshold value, wherein the azimuth consistency coefficient threshold value can be set according to the actual defect detection requirement by a person skilled in the art, and the higher the actual defect detection requirement is, the smaller the azimuth consistency coefficient threshold value is. Judging the azimuth consistency coefficient of the defect area according to the azimuth consistency coefficient threshold, and when the azimuth consistency coefficient of the defect area is larger than or equal to the azimuth consistency coefficient threshold, communicating the first slice die-closing defect area with the second slice die-closing defect area to grow the die-closing defect area, wherein the die-closing defect area growth refers to that when any one of the first slice die-closing defect area and the second slice die-closing defect area is the defect area, the other adjacent area is also marked as the defect area, and a die-closing defect area expansion result is obtained. By growing and expanding the sequence number set of the combined-die defect slice and the sequence number set of the open-die defect slice, missing defect areas can be avoided, and therefore the defect detection accuracy of the composite stamping die is improved.

And carrying out defect region growth on the open-mould defect slice sequence number set by using the same method as that for obtaining the mould closing defect region expansion result to obtain the open-mould defect region expansion result. And finally, determining a plurality of defect detection positions according to the mold closing defect region expansion result and the mold opening defect region expansion result, and carrying out defect detection on the composite stamping mold according to the plurality of defect detection positions. The method can solve the technical problem of low detection efficiency of the existing method for detecting the defects of the composite stamping die directly through image processing, and can save the positioning time of the defect detection position of the die by accurately positioning the defect position of the composite stamping die by utilizing the model slicing concept, thereby improving the defect detection efficiency of the composite stamping die.

In one embodiment, as shown in fig. 3, there is provided a defect detection system for a composite stamping die, comprising:

the three-dimensional simulation model acquisition module is used for acquiring a three-dimensional simulation model of the composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model;

the standard three-dimensional model matching module is used for matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library according to the model of the composite stamping die;

the model mold opening standard slice generation module is used for slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model mold opening standard slices, wherein the plurality of model mold opening standard slices are provided with a plurality of groups of mold opening characteristic region standard coordinates;

the model die assembly standard slice generation module is used for slicing the standard die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly standard slices, wherein the plurality of model die assembly standard slices are provided with a plurality of groups of die assembly characteristic region standard coordinates;

The die-opening and die-cutting sheet generating module is used for slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of die-opening and die-cutting sheets, wherein the die-opening and die-cutting sheets are provided with a plurality of groups of die-opening characteristic region coordinates;

the model die-closing slice generation module is used for slicing the die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing slices, wherein the plurality of model die-closing slices are provided with a plurality of groups of die-closing characteristic region coordinates;

the defect entity detection module is used for comparing the coordinates of the multi-group die clamping characteristic areas with the standard coordinates of the multi-group die clamping characteristic areas to generate die clamping defect areas, comparing the coordinates of the multi-group die opening characteristic areas with the standard coordinates of the multi-group die opening characteristic areas to generate die opening defect areas, and detecting the defect entity based on the die clamping defect areas and the die opening defect areas.

In one embodiment, the system further comprises:

the slice defect detection module is used for detecting slice defects of the standard die-closing three-dimensional model and the die-closing three-dimensional model according to a second preset interval when the number of the die-closing defect areas is equal to 0, wherein the second preset interval is one half of the first preset interval;

The die-opening sheet defect detection module is used for detecting slicing defects of the standard die-opening three-dimensional model and the die-opening three-dimensional model according to a second preset interval when the number of the die-opening defect areas is equal to 0, wherein the second preset interval is one half of the first preset interval;

the mold closing defect detection qualified mark generation module is used for generating a mold closing defect detection qualified mark if the number of the mold closing defect areas is equal to 0 when the M preset distance is smaller than or equal to a preset distance threshold value;

and the mold opening defect detection qualified mark generation module is used for generating mold opening defect detection qualified marks if the number of the mold opening defect areas is equal to 0 when the M < th > preset interval is smaller than or equal to the preset interval threshold value.

In one embodiment, the system further comprises:

the first die closing feature region coordinate set extraction module is used for extracting a first die closing feature region standard coordinate set of a first die closing standard slice according to the multi-set die closing feature region standard coordinate set, and extracting a first die closing feature region coordinate set of the first die closing slice from the multi-set die closing feature region coordinate set, wherein the first die closing feature region coordinate set is the feature region coordinate with the largest overlapping coordinate number with the first die closing feature region standard coordinate set in the first die closing slice;

The first die closing feature region defect probability setting module is used for counting the number of non-coincident coordinates and the number of coincident coordinates of the first die closing feature region standard coordinate set and the first die closing feature region coordinate set, calculating the ratio of the number of non-coincident coordinates to the total number of coordinates, and setting the ratio as the first die closing feature region defect probability, wherein the total number of coordinates is equal to the sum of the number of non-coincident coordinates and the number of coincident coordinates;

the mold clamping defect region adding module is used for adding a non-coincident coordinate distribution region of the first mold clamping characteristic region of the first mold clamping slice into the mold clamping defect region when the defect probability of the first mold clamping characteristic region is larger than or equal to a defect probability threshold;

and the defect detection qualified identification module is used for carrying out defect detection qualified identification on the first die clamping characteristic region of the first die clamping section when the defect probability of the first die clamping characteristic region is smaller than the defect probability threshold value.

In one embodiment, the system further comprises:

The characteristic region precision threshold generating module is used for traversing the multi-group die closing characteristic region standard coordinates to carry out precision calibration and generating a plurality of characteristic region precision thresholds, wherein the characteristic region precision thresholds are in one-to-one correspondence with the die closing characteristic region standard coordinates;

a maximum size deviation extraction module, configured to extract a maximum size deviation of the non-coincident coordinate distribution region of the first mold-clamping feature region of the first mold-clamping slice when the defect probability of the first mold-clamping feature region is greater than or equal to the defect probability threshold;

a mold clamping defect region adding module, configured to add a non-overlapping coordinate distribution region of the first mold clamping feature region of the first mold clamping slice to the mold clamping defect region when the maximum size deviation is greater than a feature region precision threshold of the first mold clamping feature region;

and the defect detection qualified identification module is used for carrying out defect detection qualified identification on the first die clamping characteristic region of the first die clamping section when the maximum size deviation is smaller than or equal to the characteristic region precision threshold value of the first die clamping characteristic region.

In one embodiment, the system further comprises:

the characteristic region element type acquisition module is used for traversing the standard coordinates of the multi-group die-closing characteristic region to acquire a first characteristic region element type, a second characteristic region element type and an Nth characteristic region element type;

the characteristic region stamping precision set generation module is used for traversing the first characteristic region element type and the second characteristic region element type until the Nth characteristic region element type is subjected to stamping control precision analysis to generate a first characteristic region stamping precision set and a second characteristic region stamping precision set until the Nth characteristic region stamping precision set;

the precision maximum value extraction module is used for traversing the first characteristic region stamping precision set, extracting precision maximum values from the second characteristic region stamping precision set to the Nth characteristic region stamping precision set respectively, and adding the precision maximum values into the characteristic region precision thresholds.

In one embodiment, the system further comprises:

the punching control record data retrieval module is used for retrieving a plurality of first characteristic area punching control record data by taking the first characteristic area element type as an area constraint condition, wherein the plurality of first characteristic area punching control record data have a plurality of punching control precision record values;

The stamping control precision record value generation module is used for clustering and grouping the stamping control precision record values according to precision consistency deviation to generate a plurality of groups of stamping control precision record values;

the first characteristic region stamping precision set generation module is used for eliminating groups with the number of stamping control precision recorded values smaller than or equal to the threshold value of the number of recorded values from the plurality of groups of stamping control precision recorded values to generate the first characteristic region stamping precision set;

the stamping control precision analysis module is used for traversing the second characteristic area element type until the Nth characteristic area element type is subjected to stamping control precision analysis, and generating a second characteristic area stamping precision set until the Nth characteristic area stamping precision set.

In one embodiment, the system further comprises:

the defect slice sequence number set extraction module is used for extracting a die closing defect slice sequence number set and a die opening defect slice sequence number set of the first characteristic region;

the slice mold closing defect area extraction module is used for extracting a first slice mold closing defect area and a second slice mold closing defect area of adjacent serial number slices of the mold closing defect slice serial number set;

The defect area azimuth consistency factor setting module is used for counting the number of slice plane coincident coordinates of the first slice die-closing defect area and the second slice die-closing defect area, calculating the ratio of the number of the slice plane coincident coordinates to the total number of the slice plane coordinates of the defect area, and setting the ratio as the defect area azimuth consistency factor, wherein the total number of the slice plane coordinates of the defect area is equal to the sum of the number of the slice plane coincident coordinates and the number of the slice plane non-coincident coordinates;

the mold closing defect area expansion result generation module is used for communicating the first slice mold closing defect area and the second slice mold closing defect area when the azimuth consistency coefficient of the defect area is larger than or equal to the azimuth consistency coefficient threshold value, growing the mold closing defect area and generating a mold closing defect area expansion result;

the open-mould defect area expansion result generation module is used for carrying out defect area growth according to the open-mould defect slice sequence number set to generate an open-mould defect area expansion result;

And the defect entity detection module is used for detecting the defect entity according to the mold closing defect area expansion result and the mold opening defect area expansion result.

In summary, compared with the prior art, the embodiments of the present disclosure have the following technical effects:

(1) The method solves the technical problem that the existing method for directly carrying out defect detection on the composite stamping die through image processing has lower detection efficiency, and can save the positioning time of the defect detection position of the die and improve the efficiency and accuracy of the defect detection position positioning by accurately positioning the defect position of the composite stamping die by utilizing the model slicing concept, thereby improving the defect detection efficiency of the composite stamping die.

(2) The recorded value quantity threshold is set to screen a plurality of groups of stamping control precision recorded values, accidental values with lower occurrence frequency in the stamping control precision recorded values can be removed, so that the accuracy of obtaining the stamping precision set of the first characteristic region can be improved, and the accuracy of setting the precision threshold of the characteristic region is further improved.

(3) The accuracy and the efficiency of slice defect detection can be improved by sequentially reducing the preset spacing according to the proportion to detect the slice defects, so that the accuracy and the efficiency of defect position positioning can be improved; by growing and expanding the sequence number set of the combined mould defect slice and the sequence number set of the open mould defect slice, missing defect areas can be avoided, and the accuracy of mould defect detection is improved.

The above examples merely represent a few embodiments of the present disclosure and are not to be construed as limiting the scope of the invention. Accordingly, various alterations, modifications and variations may be made by those having ordinary skill in the art without departing from the scope of the disclosed concept as defined by the following claims and all such alterations, modifications and variations are intended to be included within the scope of the present disclosure.

Claims (8)

1. The defect detection method for the composite stamping die is characterized by comprising the following steps of:

collecting a three-dimensional simulation model of a composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model;

according to the model of the composite stamping die, matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library;

slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of mold opening standard slices of the model, wherein the mold opening standard slices of the model are provided with a plurality of groups of mold opening characteristic region standard coordinates;

slicing the standard die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing standard slices, wherein the model die-closing standard slices have standard coordinates of a multi-group die-closing characteristic region;

Slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-opening slices, wherein the plurality of model die-opening slices are provided with a plurality of groups of die-opening characteristic region coordinates;

slicing the die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly slices, wherein the model die assembly slices have multiple groups of die assembly characteristic region coordinates;

and comparing the multi-group die clamping characteristic region coordinates with the multi-group die clamping characteristic region standard coordinates to generate a die clamping defect region, comparing the multi-group die opening characteristic region coordinates with the multi-group die opening characteristic region standard coordinates to generate a die opening defect region, and detecting a defect entity based on the die clamping defect region and the die opening defect region.

2. The method as recited in claim 1, further comprising:

when the number of the mold closing defect areas is equal to 0, performing slice defect detection on the standard mold closing three-dimensional model and the mold closing three-dimensional model according to a second preset interval, wherein the second preset interval is one half of the first preset interval;

When the number of the open mold defect areas is equal to 0, performing slice defect detection on the standard open mold three-dimensional model and the open mold three-dimensional model according to a second preset interval, wherein the second preset interval is one half of the first preset interval;

when the M preset interval is smaller than or equal to a preset interval threshold value, if the number of the mold closing defect areas is equal to 0, generating a mold closing defect detection qualified mark;

and when the M preset distance is smaller than or equal to a preset distance threshold value, if the number of the open-mould defect areas is equal to 0, generating an open-mould defect detection qualified mark.

3. The method of claim 1, wherein comparing the multi-set clamp characteristic region coordinates to the multi-set clamp characteristic region standard coordinates to generate a clamp defect region comprises:

extracting a first die clamping characteristic region standard coordinate set of a first die clamping standard section according to the multi-set die clamping characteristic region standard coordinates, and extracting a first die clamping characteristic region coordinate set of the first die clamping section from the multi-set die clamping characteristic region coordinates, wherein the first die clamping characteristic region coordinate set is the characteristic region coordinate with the largest overlapping coordinate number with the first die clamping characteristic region standard coordinate set in the first die clamping section;

Counting the number of non-coincident coordinates and the number of coincident coordinates of the standard coordinate set of the first die-closing feature area and the coordinate set of the first die-closing feature area, calculating the ratio of the number of non-coincident coordinates to the total number of coordinates, and setting the ratio as the defect probability of the first die-closing feature area, wherein the total number of coordinates is equal to the sum of the number of non-coincident coordinates and the number of coincident coordinates;

when the defect probability of the first die clamping characteristic region is greater than or equal to a defect probability threshold value, adding a non-coincident coordinate distribution region of the first die clamping characteristic region of the first die clamping slice into the die clamping defect region;

and when the defect probability of the first die clamping characteristic region is smaller than the defect probability threshold, performing defect detection qualification identification on the first die clamping characteristic region of the first die clamping slice.

4. The method of claim 3, wherein adding a non-coincident coordinate distribution region of the first clamp signature region of the first clamp slice into the clamp defect region when the first clamp signature region defect probability is greater than or equal to a defect probability threshold, further comprising:

performing precision calibration by traversing the standard coordinates of the multi-group die clamping characteristic areas to generate a plurality of characteristic area precision thresholds, wherein the characteristic area precision thresholds are in one-to-one correspondence with the standard coordinates of the die clamping characteristic areas;

Extracting a maximum size deviation of the non-coincident coordinate distribution region of the first mold closing feature region of the first mold closing slice when the defect probability of the first mold closing feature region is greater than or equal to the defect probability threshold;

when the maximum size deviation is larger than a characteristic region precision threshold value of the first die clamping characteristic region, adding a non-coincident coordinate distribution region of the first die clamping characteristic region of the first die clamping slice into the die clamping defect region;

and when the maximum size deviation is smaller than or equal to a characteristic region precision threshold value of the first die clamping characteristic region, performing defect detection qualification identification on the first die clamping characteristic region of the first die clamping section.

5. The method of claim 4, wherein traversing the multi-set mold clamping feature region standard coordinates for precision calibration generates a plurality of feature region precision thresholds, wherein the plurality of feature region precision thresholds are in one-to-one correspondence with the mold clamping feature region standard coordinates, comprising:

traversing the standard coordinates of the multi-combination die characteristic area to obtain a first characteristic area element type and a second characteristic area element type until an Nth characteristic area element type;

Traversing the first characteristic area element type and the second characteristic area element type until the Nth characteristic area element type is subjected to punching control precision analysis, and generating a first characteristic area punching precision set, and a second characteristic area punching precision set until the Nth characteristic area punching precision set;

traversing the first characteristic region stamping precision set, respectively extracting precision maximum values from the second characteristic region stamping precision set to the Nth characteristic region stamping precision set, and adding the precision maximum values into the characteristic region precision threshold values.

6. The method of claim 5, wherein traversing the first feature region element type, the second feature region element type, and so on until the nth feature region element type performs a punch control precision resolution, generating a first feature region punch precision set, and a second feature region punch precision set, and so on until the nth feature region punch precision set, comprises:

searching a plurality of first characteristic area stamping control record data by taking the first characteristic area element type as an area constraint condition, wherein the plurality of first characteristic area stamping control record data have a plurality of stamping control precision record values;

Clustering and grouping the stamping control precision record values according to the precision consistency deviation to generate a plurality of groups of stamping control precision record values;

removing groups with the number of stamping control precision recorded values smaller than or equal to the threshold value of the number of recorded values from the plurality of groups of stamping control precision recorded values, and generating a stamping precision set of the first characteristic region;

and traversing the second characteristic area element type until the Nth characteristic area element type is subjected to stamping control precision analysis, and generating a second characteristic area stamping precision set until the Nth characteristic area stamping precision set.

7. The method of claim 1, wherein performing defect entity detection based on the mold-closed defect region and the mold-open defect region comprises:

extracting a die closing defect slice sequence number set and a die opening defect slice sequence number set of the first characteristic region;

extracting a first slice clamping defect area and a second slice clamping defect area of adjacent serial number slices of the clamping defect slice serial number set;

counting the number of slice plane coincident coordinates of the first slice die-closing defect area and the second slice die-closing defect area, calculating the ratio of the number of slice plane coincident coordinates to the total number of slice plane coordinates of the defect area, and setting the ratio as a defect area azimuth coincident coefficient, wherein the total number of slice plane coordinates of the defect area is equal to the sum of the number of slice plane coincident coordinates and the number of slice plane non-coincident coordinates;

When the azimuth consistency coefficient of the defect area is larger than or equal to an azimuth consistency coefficient threshold value, communicating the first slice mold-closing defect area and the second slice mold-closing defect area, and growing the mold-closing defect area to generate a mold-closing defect area expansion result;

performing defect region growth according to the open-mould defect slice sequence number set to generate an open-mould defect region expansion result;

and detecting a defect entity according to the mold closing defect area expansion result and the mold opening defect area expansion result.

8. A defect detection system for a composite stamping die, characterized by the steps for performing the method in the defect detection method for a composite stamping die as claimed in any one of claims 1 to 7, comprising:

the three-dimensional simulation model acquisition module is used for acquiring a three-dimensional simulation model of the composite stamping die, wherein the three-dimensional simulation model comprises a die opening three-dimensional model and a die closing three-dimensional model;

the standard three-dimensional model matching module is used for matching a standard die opening three-dimensional model with a standard die closing three-dimensional model from a standard three-dimensional model library according to the model of the composite stamping die;

The model mold opening standard slice generation module is used for slicing the standard mold opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model mold opening standard slices, wherein the plurality of model mold opening standard slices are provided with a plurality of groups of mold opening characteristic region standard coordinates;

the model die assembly standard slice generation module is used for slicing the standard die assembly three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die assembly standard slices, wherein the plurality of model die assembly standard slices are provided with a plurality of groups of die assembly characteristic region standard coordinates;

the die-opening and die-cutting sheet generating module is used for slicing the die-opening three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of die-opening and die-cutting sheets, wherein the die-opening and die-cutting sheets are provided with a plurality of groups of die-opening characteristic region coordinates;

the model die-closing slice generation module is used for slicing the die-closing three-dimensional model from the top surface of the upper die base to the bottom surface of the lower die base according to a first preset interval to generate a plurality of model die-closing slices, wherein the plurality of model die-closing slices are provided with a plurality of groups of die-closing characteristic region coordinates;

The defect entity detection module is used for comparing the coordinates of the multi-group die clamping characteristic areas with the standard coordinates of the multi-group die clamping characteristic areas to generate die clamping defect areas, comparing the coordinates of the multi-group die opening characteristic areas with the standard coordinates of the multi-group die opening characteristic areas to generate die opening defect areas, and detecting the defect entity based on the die clamping defect areas and the die opening defect areas.