CN116423003A - Tin soldering intelligent evaluation method and system based on data mining - Google Patents
Tin soldering intelligent evaluation method and system based on data mining Download PDFInfo
- Publication number
- CN116423003A CN116423003A CN202310696553.4A CN202310696553A CN116423003A CN 116423003 A CN116423003 A CN 116423003A CN 202310696553 A CN202310696553 A CN 202310696553A CN 116423003 A CN116423003 A CN 116423003A
- Authority
- CN
- China
- Prior art keywords
- welding
- evaluation
- groups
- control parameter
- soldering
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 124
- 238000005476 soldering Methods 0.000 title claims abstract description 119
- 238000007418 data mining Methods 0.000 title claims abstract description 53
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 title claims abstract description 38
- 238000003466 welding Methods 0.000 claims abstract description 384
- 238000001514 detection method Methods 0.000 claims abstract description 135
- 238000013441 quality evaluation Methods 0.000 claims abstract description 70
- 238000000034 method Methods 0.000 claims abstract description 42
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000005457 optimization Methods 0.000 claims abstract description 27
- 229910052742 iron Inorganic materials 0.000 claims abstract description 20
- 238000007689 inspection Methods 0.000 claims abstract description 10
- 238000009826 distribution Methods 0.000 claims description 37
- 238000012216 screening Methods 0.000 claims description 36
- 230000007547 defect Effects 0.000 claims description 28
- 238000009736 wetting Methods 0.000 claims description 28
- 238000010606 normalization Methods 0.000 claims description 27
- 230000011218 segmentation Effects 0.000 claims description 23
- 238000005286 illumination Methods 0.000 claims description 19
- 229910000679 solder Inorganic materials 0.000 claims description 18
- 239000006185 dispersion Substances 0.000 claims description 14
- 238000007417 hierarchical cluster analysis Methods 0.000 claims description 10
- 238000005065 mining Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000001303 quality assessment method Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 15
- 238000005070 sampling Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 13
- 238000004891 communication Methods 0.000 description 8
- 238000010276 construction Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000003706 image smoothing Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K3/00—Tools, devices, or special appurtenances for soldering, e.g. brazing, or unsoldering, not specially adapted for particular methods
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K3/00—Tools, devices, or special appurtenances for soldering, e.g. brazing, or unsoldering, not specially adapted for particular methods
- B23K3/08—Auxiliary devices therefor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K31/00—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups
- B23K31/12—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups relating to investigating the properties, e.g. the weldability, of materials
- B23K31/125—Weld quality monitoring
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Quality & Reliability (AREA)
- General Factory Administration (AREA)
Abstract
The invention relates to the technical field of data processing, and provides a tin soldering intelligent evaluation method and system based on data mining, wherein the method comprises the following steps: after a workpiece is conveyed to a preset position, acquiring image information of the workpiece, dividing the image information, and acquiring characteristic information of a bonding pad; and the data mining obtains welding control parameters and detection record information, evaluates and generates a welding quality evaluation result, if the welding quality evaluation result does not meet the evaluation index combination expectations, the optimization result of the welding control parameters is optimized and generated and is sent to a multi-axis industrial robot and a soldering iron control module, so that the technical problems that the quality detection of a welding product is limited after production, meanwhile, the cost of determining the optimal parameter through sampling inspection is higher and the efficiency is lower after production are solved, the welding control is evaluated before the quality detection of the welding product is set before production, the quality of the welding product is guaranteed, the data mining is performed, the welding control parameter is optimized, the optimal parameter is determined in advance before production, the cost of determining the optimal parameter is reduced, and the technical effect of improving the efficiency is achieved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a tin soldering intelligent evaluation method and system based on data mining.
Background
The soldering evaluation is used for the soldering process, and various links in the soldering process are evaluated and monitored by multiple fingers so as to ensure that the quality of products meets the requirements, and meanwhile, the existing problems can be identified and corrected, the enterprises can be helped to find the problems in time and improve the problems, the defective rate and customer complaint rate are reduced, and the customer satisfaction degree is improved.
In general, sampling products are selected from soldering products in production according to a certain proportion for sampling inspection, so that the quality of the products is ensured and meets standard requirements, but the cost for determining optimal parameters through sampling inspection after production is higher, the efficiency is lower, the method is not suggested to be used for actual production and processing, and based on the method, the welding control is evaluated in advance, so that the quality of the soldering products is ensured.
In summary, the quality detection of the soldered product in the prior art is limited to the post-production process, and meanwhile, the cost of determining the optimal parameters by sampling inspection after the production process is high and the efficiency is low.
Disclosure of Invention
The application provides a data mining-based intelligent tin soldering evaluation method and system, which aim to solve the technical problems that quality detection of tin soldering products in the prior art is limited after production, and meanwhile, the cost of determining optimal parameters through sampling inspection after production is high and the efficiency is low.
In view of the above problems, the embodiments of the present application provide a method and a system for intelligent evaluation of soldering based on data mining.
In a first aspect of the disclosure, a method for intelligently evaluating soldering based on data mining is provided, wherein the method is applied to an automatic welding device, the automatic welding device comprises a vision module, an illumination module, a multi-axis industrial robot and a soldering iron control module, and the method comprises: when a workpiece to be welded is conveyed to a preset position, acquiring workpiece image information through a vision module and an illumination module for segmentation, and acquiring pad characteristic information, wherein the pad characteristic information comprises welding spot positioning characteristics and welding spot size characteristics; performing data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to acquire M groups of welding control parameters and M groups of detection record information; traversing the M groups of detection record information according to the soldering quality evaluation indexes to evaluate the M groups of welding control parameters, and generating M soldering quality evaluation results; and when all the M soldering quality evaluation results do not meet the evaluation index combination expectations, optimizing based on the M groups of welding control parameters, generating a welding control parameter optimization result, and sending the welding control parameter optimization result to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimization result is a control parameter with qualified soldering quality evaluation.
In another aspect of the disclosure, there is provided a data mining-based intelligent evaluation system for soldering, wherein the system comprises: the welding disc characteristic information acquisition module is used for acquiring image information of the workpiece to be welded to be segmented through the vision module and the illumination module when the workpiece to be welded is conveyed to a preset position, and acquiring welding disc characteristic information, wherein the welding disc characteristic information comprises welding spot positioning characteristics and welding spot size characteristics; the data mining module is used for carrying out data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to obtain M groups of welding control parameters and M groups of detection record information; the control parameter evaluation module is used for evaluating the M groups of welding control parameters according to the M groups of detection record information by traversing the soldering quality evaluation indexes to generate M soldering quality evaluation results; and the optimizing result sending module is used for optimizing based on the M groups of welding control parameters when all the M welding quality evaluation results do not meet the evaluation index combination expectations, generating a welding control parameter optimizing result and sending the welding control parameter optimizing result to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimizing result is a control parameter qualified in welding quality evaluation.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
because the method adopts the steps that when the workpiece to be welded is conveyed to a preset position, workpiece image information is collected and divided, and pad characteristic information is obtained; according to the model of the workpiece to be welded, the welding spot positioning characteristic and the welding spot size characteristic, data mining is carried out, M groups of welding control parameters and M groups of detection record information are obtained, and in combination with the welding quality evaluation index traversal evaluation, M welding quality evaluation results are generated, if the evaluation index combination expectations are not met, the optimization results of the welding control parameters are optimized and generated and sent to the multi-axis industrial robot and the soldering iron control module, the welding control is evaluated before the quality detection of the welding products is set in production, the quality of the welding products is guaranteed, the data mining is carried out, the welding control parameters are optimized, the optimal parameters are determined in advance before the production, the cost for determining the optimal parameters is reduced, and the efficiency is improved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic diagram of a possible flow of a method for intelligent evaluation of soldering based on data mining according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a possible addition of M tin soldering quality evaluation results in a tin soldering intelligent evaluation method based on data mining according to an embodiment of the present application;
fig. 3 is a schematic flow chart of a possible process of generating a first joint evaluation rule in a data mining-based intelligent tin soldering evaluation method according to an embodiment of the present application;
fig. 4 is a schematic diagram of a possible structure of a data mining-based intelligent evaluation system for soldering according to an embodiment of the present application.
Reference numerals illustrate: the device comprises a pad characteristic information acquisition module 100, a data mining module 200, a control parameter evaluation module 300 and an optimization result transmission module 400.
Detailed Description
The embodiment of the application provides a data mining-based intelligent tin soldering evaluation method and system, which solve the technical problems that the quality detection of a tin soldering product is limited after production, and meanwhile, the cost of determining the optimal parameter through sampling inspection after production is higher and the efficiency is lower.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a method for intelligently evaluating soldering based on data mining, where the method is applied to an automatic soldering device, the automatic soldering device includes a vision module, an illumination module, a multi-axis industrial robot, and a soldering iron control module, and the method includes:
s10: when a workpiece to be welded is conveyed to a preset position, acquiring workpiece image information through a vision module and an illumination module for segmentation, and acquiring pad characteristic information, wherein the pad characteristic information comprises welding spot positioning characteristics and welding spot size characteristics;
step S10 includes the steps of:
s11: converting the workpiece image information into a digital image for image preprocessing to generate a first digital image;
s12: threshold segmentation is carried out on the first digital image according to a preset gray threshold value, and a plurality of positioning round hole coordinates are obtained;
s13: after the workpiece to be welded is positioned according to the positioning round hole coordinates, a second digital image is acquired through the vision module and the illumination module;
S14: performing semantic segmentation on the second digital image to generate the solder joint positioning feature and the solder joint size feature to be added into the solder joint feature information, wherein the performing semantic segmentation on the second digital image comprises,
s15: and acquiring a welding workpiece image set to perform welding spot boundary identification, generating a welding workpiece image identification data set, and performing semantic segmentation on the second digital image based on a u-net topological network training welding spot identification model.
Specifically, the automatic welding device comprises a vision module, an illumination module, a multi-axis industrial robot and a soldering iron control module, wherein the automatic welding device is in communication connection with the vision module, the automatic welding device is in communication connection with the illumination module, the automatic welding device is in communication connection with the multi-axis industrial robot and the automatic welding device is in communication connection with the soldering iron control module simply through signal transmission interaction, and a communication network is formed among the automatic welding device and the vision module, the automatic welding device is in communication connection with the illumination module, the automatic welding device is in communication connection with the multi-axis industrial robot and the automatic welding device is in communication connection with the soldering iron control module, so that a hardware basis is provided for intelligent evaluation of soldering;
When a workpiece to be welded is conveyed to a preset position, acquiring image information of the workpiece to be welded through a vision module and an illumination module, and dividing the image information to acquire characteristic information of a welding pad, wherein the characteristic information of the welding pad comprises welding spot positioning characteristics and welding spot size characteristics, the preset position is a soldering operation table, the workpiece to be welded is subjected to soldering preparation, the soldering preparation comprises a series of surface processing such as polishing, when the workpiece to be welded is conveyed to the preset position, an image acquisition device is started, an image of the workpiece to be welded is acquired by using the image acquisition device, and the acquired image is the image information of the workpiece;
the digital image is an image expressed in a two-dimensional digital group form, the workpiece image information is subjected to digital processing, the workpiece image information is converted into a digital image and subjected to image preprocessing, a first digital image is generated, the image preprocessing at least comprises one of image enhancement processing, histogram processing, image smoothing processing and sharpening processing, and if edges and contours of the image are not clear, the image enhancement processing can be performed; if noise generated by interference can be subjected to image smoothing processing, corresponding preprocessing operation is needed to be performed on the contrast image;
The method comprises the steps that a preset gray threshold is used as a preset parameter index, a screw hole and a screw are arranged on a fixed point of a welding plate, the gray value on the fixed point of the welding plate is far greater than that of other positions, the preset gray threshold is set based on the preset gray threshold, the preset gray threshold is used as a segmentation constraint condition, the first digital image is subjected to threshold segmentation to obtain a plurality of positioning round hole coordinates, the positioning round hole coordinates are coordinates of the fixed point of the welding plate, and in general, in order to ensure that workpieces to be welded can be stably fixed on the welding plate, the number of the positioning round hole coordinates is not less than 3;
after the workpiece to be welded is positioned according to the positioning round hole coordinates, the workpiece to be welded can be stably fixed on a bonding pad, the vision module is a vision image acquisition device, the illumination module comprises a plurality of self-adjusting luminous sources for providing support for the vision module, and the image acquisition is carried out on the workpiece to be welded through the mutual matching between the vision module and the illumination module to acquire a second digital image;
performing semantic segmentation on the second digital image to generate the welding spot positioning feature and the welding spot size feature and adding the welding spot positioning feature and the welding spot size feature into the welding spot feature information, wherein the performing semantic segmentation on the second digital image comprises acquiring an image of a welded welding workpiece by using the vision module to acquire a welding workpiece image set, wherein the welding workpiece image set corresponds to the welded welding workpiece one by one;
And carrying out welding spot boundary identification on the welding workpiece image set to generate a welding workpiece image identification data set, wherein the welding spot boundary identification comprises welding spot positioning identifications and welding spot size identifications of welded welding workpieces, the welding workpiece image identification data set comprises welding spot positioning features and welding spot size features of a plurality of welded welding workpieces, the welding workpiece image set and the welding workpiece image identification data set are in one-to-one correspondence based on a u-net topological network, the welding workpiece image set and the welding workpiece image identification data set are used as training data, training is carried out in the u-net topological network according to the corresponding grouping between the welding workpiece image set and the welding workpiece image identification data set, after model output tends to be stable (model stability: the welding spot characteristics output are consistent with the welding spot positioning features and the welding spot size features of the welded welding workpieces), a welding spot identification model is obtained through training, and the second digital image is subjected to semantic segmentation by using the training-completed welding spot identification model to provide model support for welding spot positioning.
S20: performing data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to acquire M groups of welding control parameters and M groups of detection record information;
Step S20 includes the steps of:
s21: taking the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics as scene constraint factors, and taking welding control parameters and detection record information as mining target factors to acquire welding big data, so as to generate initial welding control parameter record data and initial detection record information;
s22: setting a plurality of control parameter deviation thresholds through the welding control parameters;
s23: performing hierarchical clustering analysis on the initial welding control parameter record data according to a first control parameter deviation threshold value of the control parameter deviation threshold values to generate a first clustering result;
s24: traversing the ith-1 clustering result according to the ith control parameter deviation threshold value of the control parameter deviation threshold values to perform hierarchical clustering analysis to generate an ith clustering result;
s25: when i is equal to the number of welding control parameter types, acquiring the intra-class support degree of the i clustering result;
s26: setting the initial welding control parameter record data with the in-class support degree larger than or equal to a support degree threshold value and the corresponding initial detection record information as the M groups of welding control parameters and the M groups of detection record information;
S27: wherein the welding control parameters comprise tin feeding quantity, tin feeding speed, pause time and welding temperature; the support degree in the class characterizes the clustering frequency parameter of the control parameter record data; the detection record information comprises appearance detection record data and electrical property detection record data.
Specifically, data mining is performed according to a type of a workpiece to be welded, the welding spot positioning characteristic and the welding spot size characteristic, and M groups of welding control parameters and M groups of detection record information are obtained, wherein the type of the workpiece to be welded comprises a type of the workpiece and corresponding codes, the workpiece to be welded can be uniquely determined, the detection record information comprises appearance detection record data and electrical property detection record data, the welding control parameters comprise tin feeding quantity, tin feeding speed, pause time and welding temperature, the type of the workpiece to be welded, the welding spot positioning characteristic and the welding spot size characteristic are set as scene constraint factors, the welding control parameters and the detection record information are used as mining retrieval contents, mining target factors are set, welding big data acquisition is performed by the mining target factors, initial welding control parameter record data and initial detection record information are acquired, the data type of the initial welding control parameter record data is consistent with the data type of the welding control parameters, and the data type of the initial detection record information is consistent with the data type of the detection record information;
Traversing the welding control parameters, comparing the maximum value and the minimum value in each parameter mark in the welding control parameters, and correspondingly setting a plurality of control parameter deviation thresholds, wherein the welding control parameters comprise, but are not limited to, tin feeding quantity, tin feeding speed, pause time and welding temperature, and the corresponding control parameter deviation thresholds at least comprise tin feeding quantity thresholds, tin feeding speed thresholds, pause time thresholds and welding temperature thresholds;
the first control parameter deviation threshold is any one of the control parameter deviation thresholds, hierarchical clustering analysis is performed on the initial welding control parameter record data according to the first control parameter deviation threshold of the control parameter deviation thresholds, namely, the median value of the first control parameter deviation threshold is simply selected as a reference point, the initial welding control parameter record data is subjected to bottom-up condensation hierarchical clustering, and iteration is performed until the welding control parameter distribution in the initial welding control parameter record data is not changed any more, and a first clustering result is generated; repeating the steps according to the first clustering result, traversing the ith-1 clustering result according to the ith control parameter deviation threshold value of the control parameter deviation threshold values, performing hierarchical clustering analysis, and generating an ith clustering result;
The intra-class support represents the clustering frequency parameter of the control parameter record data, and if the welding control parameters are tin feeding quantity, tin feeding speed, pause time and welding temperature, i.e. when i=4, i is equal to the number of welding control parameter types, the intra-class support of the ith clustering result is obtained; the detection record information comprises appearance detection record data and electrical property detection record data; setting the initial welding control parameter record data with the in-class support degree larger than or equal to a support degree threshold value and the corresponding initial detection record information as the M groups of welding control parameters and the M groups of detection record information, and providing a data basis for welding control parameter optimization through data mining.
Step S26 includes the steps of:
s261: carrying out serialization identification on the initial welding control parameter record data with the support degree in the class being greater than or equal to the support degree threshold value according to the support degree in the class from large to small, and generating a sequence identification result;
s262: and screening the M groups of welding control parameters and the M groups of detection record information corresponding to the M groups of welding control parameters from the beginning to the end according to the sequence identification result, wherein M is more than or equal to 20, M is a positive integer, and the data mining target quantity is represented.
Specifically, the initial welding control parameter record data with the Intra-class support degree greater than or equal to a support degree threshold and the corresponding initial detection record information are set as the M groups of welding control parameters and the M groups of detection record information, including that the support degree threshold is set by a person skilled in the art, the Intra-class support degree (ICS) is an index for evaluating the clustering quality, and is mainly used for evaluating the similarity and compactness between information in the same class, the Intra-class support degree and the support degree threshold are compared, if the Intra-class support degree is greater than or equal to the support degree threshold, the initial welding control parameter record data with the Intra-class support degree greater than or equal to the support degree threshold is subjected to serialization identification according to the Intra-class support degree from large to small, and after the serialization identification is completed until the rest Intra-class support degree is less than the support degree threshold, a sequence identification result is obtained, and the Intra-class support degree in the sequence identification result is greater than or equal to the initial welding control parameter record data with the support degree greater than or equal to the support degree threshold is not less than 100; and screening M groups of welding control parameters and corresponding M groups of detection record information which are ranked at the front from the beginning to the end according to the sequence identification result, wherein M is more than or equal to 20, M is a positive integer and used for representing the data mining target quantity, and providing support for ensuring that the data quantity of data mining meets the requirement.
S30: traversing the M groups of detection record information according to the soldering quality evaluation indexes to evaluate the M groups of welding control parameters, and generating M soldering quality evaluation results;
as shown in fig. 2, step S30 includes the steps of:
s31: the tin soldering quality evaluation indexes comprise appearance detection indexes and electrical performance detection indexes, wherein the appearance detection indexes comprise wetting angle parameters, welding smoothness, welding spot thickness, welding spot size and welding spot defect parameters, and the electrical performance detection indexes comprise conductivity, insulativity and component reject ratio;
s32: constructing a first-level screening system and a second-level evaluation system according to the appearance detection index and the electrical performance detection index;
s33: when any one of the M groups of detection record information meets the primary screening system, carrying out joint evaluation according to the secondary evaluation system, and adding the M tin soldering quality evaluation results;
s34: and when any one of the M groups of detection record information does not meet the first-stage screening system, setting the soldering quality evaluation result as the lowest score, and adding the M soldering quality evaluation results.
Specifically, the M groups of welding control parameters are evaluated according to the welding quality evaluation indexes by traversing the M groups of detection record information, and M welding quality evaluation results are generated, wherein the welding quality evaluation indexes comprise appearance detection indexes and electrical performance detection indexes, the appearance detection indexes comprise wetting angle parameters, welding smoothness, welding spot thickness, welding spot size and welding spot defect parameters, and the electrical performance detection indexes comprise conductivity, insulativity and component reject ratio; the first-level screening system can be a screening system of the appearance of the workpiece corresponding to the appearance detection index, the second-level screening system can be a screening system of the electrical performance of the workpiece corresponding to the electrical performance detection index, and a first-level screening system and a second-level evaluation system are respectively constructed according to the appearance detection index and the electrical performance detection index;
Comparing the M groups of detection record information with appearance detection indexes, if the appearance of the workpiece is screened to pass, namely any one of the M groups of detection record information meets the first-level screening system, and when any one of the M groups of detection record information meets the first-level screening system, comparing the M groups of detection record information with the electric performance detection indexes, carrying out joint evaluation according to the second-level evaluation system, and adding the M tin soldering quality evaluation results, wherein the evaluation indexes corresponding to the joint evaluation comprise electric performance detection indexes such as conductivity, insulativity and component reject ratio, and the tin soldering quality evaluation results comprise a conductivity quality evaluation result, an insulativity quality evaluation result and a component reject ratio quality evaluation result;
and comparing the M groups of detection record information with appearance detection indexes, if one or more than one item of workpiece appearance screening passes, namely any one of the M groups of detection record information does not meet the primary screening system, and when any one of the M groups of detection record information does not meet the primary screening system, directly setting a soldering quality evaluation result as the lowest score, and adding the M soldering quality evaluation results to provide a reference for soldering quality evaluation.
Step S32 includes the steps of:
s321: setting a detection index dispersion threshold value by traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter, the conductivity, the insulativity and the component reject ratio;
s322: traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter for weight distribution, and obtaining a first weight distribution result;
s323: traversing the conductivity, the insulativity and the component reject ratio to perform weight distribution, and obtaining a second weight distribution result;
s324: setting a first joint evaluation rule for the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter according to the first weight distribution result;
s325: setting a second joint evaluation rule for the conductivity, the insulativity and the component reject ratio according to the second weight distribution result;
s326: constructing the first-level screening system according to the detection index dispersion threshold value;
s327: and constructing the second-level evaluation system according to the first joint evaluation rule and the second joint evaluation rule.
Specifically, a first-level screening system and a second-level evaluation system are constructed according to the appearance detection indexes and the electrical performance detection indexes, wherein the first-level screening system and the second-level evaluation system comprise that the minimum value and the maximum value of wetting angle parameters, welding smoothness, welding spot thickness, welding spot size and welding spot defect parameters, and conduction degree, insulativity and component reject ratio in the electrical performance detection indexes are taken as traversal directions, M groups of detection record information are taken as traversal contents, and detection index dispersion thresholds are set according to the minimum value and the maximum value of the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameters, the conduction degree, the insulativity and the component reject ratio;
traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter by using an analytic hierarchy process to carry out weight distribution, and obtaining a first weight distribution result: the analytic hierarchy process is a subjective weighting process, a structure of a hierarchical hierarchy of the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter and the detection index dispersion threshold is constructed through the analytic hierarchy process, matrix transformation is carried out through structural features of the hierarchical hierarchy of the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter and the detection index dispersion threshold, a judged matrix is generated, and weights corresponding to the hierarchical hierarchy of the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter and the detection index dispersion threshold are calculated to obtain a first weight distribution result; comparing the analysis step of the first weight distribution result, traversing the conductivity, the insulativity and the component reject ratio by using an analytic hierarchy process to carry out weight distribution, and obtaining a second weight distribution result;
Carrying out weighted evaluation on the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameters according to the first weight distribution result, and setting a first joint evaluation rule; setting a second joint evaluation rule for the conductivity, the insulativity and the component reject ratio according to the second weight distribution result by comparing with the first joint evaluation rule;
in order, the detection index dispersion threshold value is set as a screening condition, the first-stage screening system is constructed according to the detection index dispersion threshold value, related data cannot be deleted if the detection index dispersion threshold value is not met, and after all the detection index dispersion threshold value is met, the second-stage evaluation system is constructed according to the first joint evaluation rule and the second joint evaluation rule, wherein the first joint evaluation rule is used for restricting appearance detection, the second joint evaluation rule is used for restricting electrical performance detection, and support is provided for appearance detection and electrical performance detection in tin soldering quality evaluation.
As shown in fig. 3, step S324 includes the steps of:
s324-1: a normalization processing channel is constructed, and the normalization processing layer is used for carrying out normalization processing on the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameters;
S324-2: a weighted evaluation channel is constructed, and the weighted evaluation channel is connected in series with the normalization processing channel and is used for carrying out weighted evaluation on the output data of the normalization processing channel according to the first weight distribution result;
s324-3: and generating the first joint evaluation rule according to the normalization processing channel and the weighted evaluation channel.
Specifically, a first joint evaluation rule is set for the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter according to the first weight distribution result, and the method comprises the steps of constructing a normalization processing channel by using a variation coefficient method, wherein the variation coefficient method is an objective weighting method, and the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter are mapped into a section (0, 1) by directly using information contained in the wetting angle parameter, the welding spot thickness, the welding spot size and the welding spot defect parameter, and the normalization processing layer is used for carrying out normalization processing on the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the defect parameter;
The method comprises the steps of carrying out weight distribution by using an analytic hierarchy process to obtain a first weight distribution result, wherein the analytic hierarchy process is a subjective weighting process, a comprehensive variation coefficient process and the analytic hierarchy process carry out combined weighting, the combined weighting is carried out by combining subjective and objective weighting results, a weighted evaluation channel is constructed, and after the weighted evaluation channel is connected with the normalization processing channel in series, the weighted evaluation channel is used for carrying out weighted evaluation on output data of the normalization processing channel according to the first weight distribution result;
and setting the obtained serial connection of the normalization processing channel and the weighted evaluation channel as a first joint evaluation rule, determining the first joint evaluation rule, and carrying out weighted combination on different indexes or characteristics by utilizing combination weights to eliminate noise and deviation of individual indexes or characteristics and enhance the stability and generalization capability of a screening system of the appearance of the workpiece.
S40: and when all the M soldering quality evaluation results do not meet the evaluation index combination expectations, optimizing based on the M groups of welding control parameters, generating a welding control parameter optimization result, and sending the welding control parameter optimization result to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimization result is a control parameter with qualified soldering quality evaluation.
Step S40 includes the steps of:
s41: when all the M tin soldering quality evaluation results do not meet the evaluation index combined expectation, carrying out secondary data mining according to the type of the workpiece to be soldered, the welding point positioning characteristics and the welding point size characteristics to obtain N groups of soldering control parameters, wherein the N groups of soldering control parameters are different from the M groups of soldering control parameters, and N is more than or equal to 2M;
s42: when any one of N soldering quality evaluation results of the N groups of soldering control parameters meets the evaluation index combination expectation, the soldering control parameter optimization result is set to be sent to the multi-axis industrial robot and the soldering iron control module.
In particular, when all of the M soldering quality evaluation results do not meet the evaluation index combined expectation, optimizing based on the M groups of welding control parameters to generate welding control parameter optimization results, and sending the welding control parameter optimization results to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimization results are control parameters qualified in soldering quality evaluation, the evaluation index combined expectation is an index which is custom set by relevant technicians in the field, the M soldering quality evaluation results and the evaluation index combined expectation are compared, when all of the M soldering quality evaluation results do not meet the evaluation index combined expectation, namely, samples which meet the evaluation index combined expectation do not exist in M groups of welding control parameters and M groups of detection record information which indicate data mining, the data range is further expanded, secondary data mining is carried out according to the model of the workpiece to be welded, the welding spot positioning characteristic and the welding spot size characteristic on the basis of the M groups of welding control parameters and the M groups of detection record information, N groups of welding control parameters are obtained, the data mining is consistent with the operation steps of the secondary data mining, the N groups of welding control parameters are different from the M groups of welding control parameters, N is more than or equal to 2M, and likewise, when the N tin soldering quality assessment results corresponding to the N groups of welding control parameters do not meet the assessment index joint expectations, three times of data mining are still needed until the assessment index joint expectations are met, and likewise, N is a positive integer;
Comparing the M soldering quality evaluation results with the evaluation index combination expectation, and when any one of the N soldering quality evaluation results of the N groups of soldering control parameters meets the evaluation index combination expectation, namely at least one sample meeting the evaluation index combination expectation exists in the M groups of soldering control parameters and the M groups of detection record information which indicate data mining, displaying and marking the sample meeting the evaluation index combination expectation, setting the soldering control parameters with the display marks in the M groups of soldering control parameters as the soldering control parameter optimization result, and sending the soldering control parameters to the multi-axis industrial robot and the soldering iron control module, thereby providing support for automatic optimization of soldering control parameters.
In summary, the intelligent tin soldering evaluation method and system based on data mining provided by the embodiment of the application have the following technical effects:
1. because the method adopts the steps that when the workpiece to be welded is conveyed to a preset position, workpiece image information is collected and divided, and pad characteristic information is obtained; according to the model of a workpiece to be welded, welding spot positioning characteristics and welding spot size characteristics, data mining is carried out, M groups of welding control parameters and M groups of detection record information are obtained, and in combination with welding quality evaluation indexes, traversing evaluation is carried out, M welding quality evaluation results are generated, if the evaluation indexes are not met, combined expectations are not met, the optimization results of the welding control parameters are optimized and generated, and the optimization results are sent to a multi-axis industrial robot and a soldering iron control module.
2. Due to the adoption of the construction normalization processing channel; and constructing a weighted evaluation channel, generating a first joint evaluation rule by combining the normalization processing channel, and carrying out weighted combination on different indexes or characteristics by utilizing the combination weighting, so that noise and deviation of individual indexes or characteristics are eliminated, and the stability and generalization capability of a screening system of the appearance of the workpiece are enhanced.
Example two
Based on the same inventive concept as the data mining-based intelligent evaluation method in the foregoing embodiment, as shown in fig. 4, an embodiment of the present application provides an intelligent evaluation system for soldering based on data mining, where the system includes:
the welding disc characteristic information acquisition module 100 is used for acquiring workpiece image information through the vision module and the illumination module to divide the workpiece image information when the workpiece to be welded is conveyed to a preset position, so as to acquire welding disc characteristic information, wherein the welding disc characteristic information comprises welding spot positioning characteristics and welding spot size characteristics;
the data mining module 200 is used for carrying out data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to obtain M groups of welding control parameters and M groups of detection record information;
the control parameter evaluation module 300 is configured to evaluate the M groups of welding control parameters according to the M groups of detection record information by traversing the soldering quality evaluation index, and generate M soldering quality evaluation results;
And the optimizing result sending module 400 is configured to optimize based on the M groups of welding control parameters when none of the M welding quality evaluation results meets the evaluation index combination expectation, and generate a welding control parameter optimizing result, and send the welding control parameter optimizing result to the multi-axis industrial robot and the soldering iron control module, where the welding control parameter optimizing result is a control parameter qualified in welding quality evaluation.
Further, the system includes:
the first digital image acquisition module is used for converting the workpiece image information into a digital image for image preprocessing to generate a first digital image;
the threshold segmentation module is used for carrying out threshold segmentation on the first digital image according to a preset gray threshold value to obtain a plurality of positioning round hole coordinates;
the second digital image acquisition module is used for acquiring a second digital image through the vision module and the illumination module after the workpiece to be welded is positioned according to the coordinates of the positioning round holes;
a pad feature information generating module, configured to perform semantic segmentation on the second digital image, generate the solder joint positioning feature and the solder joint size feature, and add the solder joint positioning feature and the solder joint size feature to the pad feature information, where the performing semantic segmentation on the second digital image includes,
The semantic segmentation module is used for collecting a welding workpiece image set to carry out welding spot boundary identification, generating a welding workpiece image identification data set, and carrying out semantic segmentation on the second digital image based on a u-net topological network training welding spot identification model.
Further, the system includes:
the welding data acquisition module is used for acquiring welding big data by taking the model of the workpiece to be welded, the welding spot positioning characteristic and the welding spot size characteristic as scene constraint factors and taking welding control parameters and detection record information as mining target factors to generate initial welding control parameter record data and initial detection record information;
the control parameter deviation threshold setting module is used for traversing the welding control parameters to set a plurality of control parameter deviation thresholds;
the first hierarchical clustering analysis module is used for performing hierarchical clustering analysis on the initial welding control parameter record data according to a first control parameter deviation threshold value of the control parameter deviation threshold values to generate a first clustering result;
the ith hierarchical clustering analysis module is used for traversing the ith-1 clustering result according to the ith control parameter deviation threshold value of the control parameter deviation threshold values to perform hierarchical clustering analysis and generate an ith clustering result;
The intra-class support degree acquisition module is used for acquiring the intra-class support degree of the i-th clustering result when the i is equal to the number of welding control parameter types;
the welding control parameter and detection record setting module is used for setting the initial welding control parameter record data with the in-class support degree larger than or equal to a support degree threshold value and the corresponding initial detection record information into the M groups of welding control parameters and M groups of detection record information;
the welding control parameters comprise tin feeding quantity, tin feeding speed, pause time and welding temperature; the support degree in the class characterizes the clustering frequency parameter of the control parameter record data; the detection record information comprises appearance detection record data and electrical property detection record data.
Further, the system includes:
the serialization identification module is used for carrying out serialization identification on the initial welding control parameter record data with the in-class support degree greater than or equal to the support degree threshold value according to the in-class support degree, and generating a sequence identification result;
and the data screening module is used for screening the M groups of welding control parameters and the M groups of detection record information corresponding to the M groups of welding control parameters from the beginning to the end according to the sequence identification result, wherein M is more than or equal to 20, M is a positive integer, and the data mining target quantity is represented.
Further, the system includes:
the soldering quality evaluation index, the appearance detection index and the electrical property detection index determining module is used for the soldering quality evaluation index comprising the appearance detection index and the electrical property detection index, the appearance detection indexes comprise wetting angle parameters, welding smoothness, welding spot thickness, welding spot size and welding spot defect parameters, and the electrical performance detection indexes comprise conductivity, insulativity and component reject ratio;
1. the second-level evaluation system construction module is used for constructing a first-level screening system and a second-level evaluation system according to the appearance detection index and the electrical performance detection index;
the joint evaluation module is used for performing joint evaluation according to the second-level evaluation system when any one of the M groups of detection record information meets the first-level screening system, and adding the M tin soldering quality evaluation results;
and the soldering quality evaluation result adding module is used for setting the soldering quality evaluation result as the lowest score when any one of the M groups of detection record information does not meet the primary screening system, and adding the M soldering quality evaluation results.
Further, the system includes:
the index traversing module is used for traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter, the conductivity, the insulativity and the component reject ratio to set a detection index dispersion threshold;
The first weight distribution result acquisition module is used for traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter to carry out weight distribution, so as to acquire a first weight distribution result;
the second weight distribution result acquisition module is used for traversing the conductivity, the insulativity and the component reject ratio to perform weight distribution and acquiring a second weight distribution result;
the first joint evaluation rule setting module is used for setting a first joint evaluation rule for the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameters according to the first weight distribution result;
a second joint evaluation rule setting module, configured to set a second joint evaluation rule for the conductivity, the insulation, and the component reject ratio according to the second weight distribution result;
the first-level screening system construction module is used for constructing the first-level screening system according to the detection index dispersion threshold value;
and the second-level screening system construction module is used for constructing the second-level evaluation system according to the first joint evaluation rule and the second joint evaluation rule.
Further, the system includes:
the normalization processing module is used for constructing a normalization processing channel, and the normalization processing layer is used for carrying out normalization processing on the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameters;
the weighting evaluation module is used for constructing a weighting evaluation channel which is connected in series with the normalization processing channel and then used for carrying out weighting evaluation on the output data of the normalization processing channel according to the first weight distribution result;
and the first joint evaluation rule generation module is used for generating the first joint evaluation rule according to the normalization processing channel and the weighted evaluation channel.
Further, the system includes:
the secondary data mining module is used for carrying out secondary data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics when all the M tin soldering quality evaluation results do not meet the evaluation index joint expectations, and obtaining N groups of welding control parameters, wherein the N groups of welding control parameters are different from the M groups of welding control parameters, and N is more than or equal to 2M;
and the welding control parameter optimization result sending module is used for setting the welding control parameter optimization result to be sent to the multi-axis industrial robot and the soldering iron control module when any one of N soldering quality evaluation results of the N groups of welding control parameters meets the evaluation index combination expectation.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any of the methods to implement embodiments of the present application, without unnecessary limitations.
Further, the first or second may represent not only the order relationship but also a specific concept. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (9)
1. The utility model provides a soldering intelligent evaluation method based on data mining which characterized in that is applied to automatic welding device, automatic welding device includes vision module, illumination module, multiaxis industrial robot and flatiron control module, includes:
when a workpiece to be welded is conveyed to a preset position, acquiring workpiece image information through a vision module and an illumination module for segmentation, and acquiring pad characteristic information, wherein the pad characteristic information comprises welding spot positioning characteristics and welding spot size characteristics;
Performing data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to acquire M groups of welding control parameters and M groups of detection record information;
traversing the M groups of detection record information according to the soldering quality evaluation indexes to evaluate the M groups of welding control parameters, and generating M soldering quality evaluation results;
and when all the M soldering quality evaluation results do not meet the evaluation index combination expectations, optimizing based on the M groups of welding control parameters, generating a welding control parameter optimization result, and sending the welding control parameter optimization result to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimization result is a control parameter with qualified soldering quality evaluation.
2. The method of claim 1, wherein when the workpiece to be welded is transported to a preset position, the workpiece image information is collected by a vision module and an illumination module for segmentation, and pad feature information is obtained, wherein the pad feature information includes a pad positioning feature and a pad size feature, and the method comprises:
converting the workpiece image information into a digital image for image preprocessing to generate a first digital image;
threshold segmentation is carried out on the first digital image according to a preset gray threshold value, and a plurality of positioning round hole coordinates are obtained;
After the workpiece to be welded is positioned according to the positioning round hole coordinates, a second digital image is acquired through the vision module and the illumination module;
performing semantic segmentation on the second digital image to generate the solder joint positioning feature and the solder joint size feature to be added into the solder joint feature information, wherein the performing semantic segmentation on the second digital image comprises,
and acquiring a welding workpiece image set to perform welding spot boundary identification, generating a welding workpiece image identification data set, and performing semantic segmentation on the second digital image based on a u-net topological network training welding spot identification model.
3. The method of claim 1, wherein performing data mining based on the model of the workpiece to be welded, the weld spot positioning feature, and the weld spot size feature to obtain M sets of welding control parameters and M sets of inspection log information, comprising:
taking the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics as scene constraint factors, and taking welding control parameters and detection record information as mining target factors to acquire welding big data, so as to generate initial welding control parameter record data and initial detection record information;
Setting a plurality of control parameter deviation thresholds through the welding control parameters;
performing hierarchical clustering analysis on the initial welding control parameter record data according to a first control parameter deviation threshold value of the control parameter deviation threshold values to generate a first clustering result;
traversing the ith-1 clustering result according to the ith control parameter deviation threshold value of the control parameter deviation threshold values to perform hierarchical clustering analysis to generate an ith clustering result;
when i is equal to the number of welding control parameter types, acquiring the intra-class support degree of the i clustering result;
setting the initial welding control parameter record data with the in-class support degree larger than or equal to a support degree threshold value and the corresponding initial detection record information as the M groups of welding control parameters and the M groups of detection record information;
wherein the welding control parameters comprise tin feeding quantity, tin feeding speed, pause time and welding temperature; the support degree in the class characterizes the clustering frequency parameter of the control parameter record data; the detection record information comprises appearance detection record data and electrical property detection record data.
4. The method of claim 3, wherein setting the initial welding control parameter record data and the corresponding initial inspection record information for which the in-class support is greater than or equal to a support threshold as the M sets of welding control parameters and the M sets of inspection record information comprises:
Carrying out serialization identification on the initial welding control parameter record data with the support degree in the class being greater than or equal to the support degree threshold value according to the support degree in the class from large to small, and generating a sequence identification result;
and screening the M groups of welding control parameters and the M groups of detection record information corresponding to the M groups of welding control parameters from the beginning to the end according to the sequence identification result, wherein M is more than or equal to 20, M is a positive integer, and the data mining target quantity is represented.
5. The method of claim 1, wherein evaluating the M sets of welding control parameters according to the M sets of inspection log information to generate M solder quality evaluation results comprises:
the tin soldering quality evaluation indexes comprise appearance detection indexes and electrical performance detection indexes, wherein the appearance detection indexes comprise wetting angle parameters, welding smoothness, welding spot thickness, welding spot size and welding spot defect parameters, and the electrical performance detection indexes comprise conductivity, insulativity and component reject ratio;
constructing a first-level screening system and a second-level evaluation system according to the appearance detection index and the electrical performance detection index;
when any one of the M groups of detection record information meets the primary screening system, carrying out joint evaluation according to the secondary evaluation system, and adding the M tin soldering quality evaluation results;
And when any one of the M groups of detection record information does not meet the first-stage screening system, setting the soldering quality evaluation result as the lowest score, and adding the M soldering quality evaluation results.
6. The method of claim 5, wherein constructing a primary screening system and a secondary evaluation system based on the appearance detection indicator and the electrical performance detection indicator comprises:
setting a detection index dispersion threshold value by traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size, the welding spot defect parameter, the conductivity, the insulativity and the component reject ratio;
traversing the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter for weight distribution, and obtaining a first weight distribution result;
traversing the conductivity, the insulativity and the component reject ratio to perform weight distribution, and obtaining a second weight distribution result;
setting a first joint evaluation rule for the wetting angle parameter, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameter according to the first weight distribution result;
Setting a second joint evaluation rule for the conductivity, the insulativity and the component reject ratio according to the second weight distribution result;
constructing the first-level screening system according to the detection index dispersion threshold value;
and constructing the second-level evaluation system according to the first joint evaluation rule and the second joint evaluation rule.
7. The method of claim 6, wherein setting a first joint evaluation rule for the wetting angle parameter, the weld smoothness, the weld spot thickness, the weld spot size, and the weld spot defect parameter based on the first weight distribution result comprises:
a normalization processing channel is constructed, and the normalization processing layer is used for carrying out normalization processing on the wetting angle parameters, the welding smoothness, the welding spot thickness, the welding spot size and the welding spot defect parameters;
a weighted evaluation channel is constructed, and the weighted evaluation channel is connected in series with the normalization processing channel and is used for carrying out weighted evaluation on the output data of the normalization processing channel according to the first weight distribution result;
and generating the first joint evaluation rule according to the normalization processing channel and the weighted evaluation channel.
8. The method of claim 1, wherein when none of the M solder quality assessment results meets an evaluation index joint expectation, optimizing based on the M sets of solder control parameters, generating a solder control parameter optimization result, and sending the solder control parameter optimization result to the multi-axis industrial robot and the soldering iron control module, wherein the solder control parameter optimization result is a control parameter that is qualified for solder quality assessment, comprising:
when all the M tin soldering quality evaluation results do not meet the evaluation index combined expectation, carrying out secondary data mining according to the type of the workpiece to be soldered, the welding point positioning characteristics and the welding point size characteristics to obtain N groups of soldering control parameters, wherein the N groups of soldering control parameters are different from the M groups of soldering control parameters, and N is more than or equal to 2M;
when any one of N soldering quality evaluation results of the N groups of soldering control parameters meets the evaluation index combination expectation, the soldering control parameter optimization result is set to be sent to the multi-axis industrial robot and the soldering iron control module.
9. A data mining based intelligent evaluation system for solder, which is configured to implement the data mining based intelligent evaluation method of any one of claims 1-8, comprising:
The welding disc characteristic information acquisition module is used for acquiring image information of the workpiece to be welded to be segmented through the vision module and the illumination module when the workpiece to be welded is conveyed to a preset position, and acquiring welding disc characteristic information, wherein the welding disc characteristic information comprises welding spot positioning characteristics and welding spot size characteristics;
the data mining module is used for carrying out data mining according to the model of the workpiece to be welded, the welding spot positioning characteristics and the welding spot size characteristics to obtain M groups of welding control parameters and M groups of detection record information;
the control parameter evaluation module is used for evaluating the M groups of welding control parameters according to the M groups of detection record information by traversing the soldering quality evaluation indexes to generate M soldering quality evaluation results;
and the optimizing result sending module is used for optimizing based on the M groups of welding control parameters when all the M welding quality evaluation results do not meet the evaluation index combination expectations, generating a welding control parameter optimizing result and sending the welding control parameter optimizing result to the multi-axis industrial robot and the soldering iron control module, wherein the welding control parameter optimizing result is a control parameter qualified in welding quality evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310696553.4A CN116423003B (en) | 2023-06-13 | 2023-06-13 | Tin soldering intelligent evaluation method and system based on data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310696553.4A CN116423003B (en) | 2023-06-13 | 2023-06-13 | Tin soldering intelligent evaluation method and system based on data mining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116423003A true CN116423003A (en) | 2023-07-14 |
| CN116423003B CN116423003B (en) | 2023-10-31 |
Family
ID=87080092
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310696553.4A Active CN116423003B (en) | 2023-06-13 | 2023-06-13 | Tin soldering intelligent evaluation method and system based on data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116423003B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116680657A (en) * | 2023-08-01 | 2023-09-01 | 张家港禾福新材料科技有限公司 | Processing control method and system for functional plastic master batch |
| CN116822098A (en) * | 2023-08-18 | 2023-09-29 | 广州誉鑫精密部件有限公司 | Intelligent monitoring control method and system for welding |
| CN118046132A (en) * | 2024-04-16 | 2024-05-17 | 嘉兴云切在线科技有限公司 | Groove precision detection method and system applied to cloud cutting platform |
| CN118081218A (en) * | 2024-04-23 | 2024-05-28 | 南通南铭电子有限公司 | Automatic welding control method and system for capacitor pins |
| CN118218865A (en) * | 2024-05-24 | 2024-06-21 | 四川中宝科技有限公司 | Welding spot cracking early warning system and method applied to intelligent bonding copper wire equipment |
| CN118380349A (en) * | 2024-06-21 | 2024-07-23 | 南通华隆微电子股份有限公司 | Method for optimizing packaging process of semiconductor chip |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112419274A (en) * | 2020-11-24 | 2021-02-26 | 英业达(重庆)有限公司 | Solder paste detection method, system, electronic device and medium |
| CN113506243A (en) * | 2021-06-04 | 2021-10-15 | 联合汽车电子有限公司 | PCB welding defect detection method and device and storage medium |
| CN113724216A (en) * | 2021-08-24 | 2021-11-30 | 广东工业大学 | Method and system for detecting wave crest welding spot defects |
| CN115375889A (en) * | 2022-08-10 | 2022-11-22 | 深圳市大德激光技术有限公司 | Processing method and system for identifying positioning pad through fusion of multiple illuminations |
| CN115922021A (en) * | 2021-08-12 | 2023-04-07 | 台达电子工业股份有限公司 | Automatic tin soldering system and automatic tin soldering method |
-
2023
- 2023-06-13 CN CN202310696553.4A patent/CN116423003B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112419274A (en) * | 2020-11-24 | 2021-02-26 | 英业达(重庆)有限公司 | Solder paste detection method, system, electronic device and medium |
| CN113506243A (en) * | 2021-06-04 | 2021-10-15 | 联合汽车电子有限公司 | PCB welding defect detection method and device and storage medium |
| CN115922021A (en) * | 2021-08-12 | 2023-04-07 | 台达电子工业股份有限公司 | Automatic tin soldering system and automatic tin soldering method |
| CN113724216A (en) * | 2021-08-24 | 2021-11-30 | 广东工业大学 | Method and system for detecting wave crest welding spot defects |
| CN115375889A (en) * | 2022-08-10 | 2022-11-22 | 深圳市大德激光技术有限公司 | Processing method and system for identifying positioning pad through fusion of multiple illuminations |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116680657A (en) * | 2023-08-01 | 2023-09-01 | 张家港禾福新材料科技有限公司 | Processing control method and system for functional plastic master batch |
| CN116680657B (en) * | 2023-08-01 | 2023-09-26 | 张家港禾福新材料科技有限公司 | Processing control method and system for functional plastic master batch |
| CN116822098A (en) * | 2023-08-18 | 2023-09-29 | 广州誉鑫精密部件有限公司 | Intelligent monitoring control method and system for welding |
| CN116822098B (en) * | 2023-08-18 | 2024-01-16 | 广州誉鑫精密部件有限公司 | Intelligent monitoring control method and system for welding |
| CN118046132A (en) * | 2024-04-16 | 2024-05-17 | 嘉兴云切在线科技有限公司 | Groove precision detection method and system applied to cloud cutting platform |
| CN118081218A (en) * | 2024-04-23 | 2024-05-28 | 南通南铭电子有限公司 | Automatic welding control method and system for capacitor pins |
| CN118081218B (en) * | 2024-04-23 | 2024-07-05 | 南通南铭电子有限公司 | Automatic welding control method and system for capacitor pins |
| CN118218865A (en) * | 2024-05-24 | 2024-06-21 | 四川中宝科技有限公司 | Welding spot cracking early warning system and method applied to intelligent bonding copper wire equipment |
| CN118380349A (en) * | 2024-06-21 | 2024-07-23 | 南通华隆微电子股份有限公司 | Method for optimizing packaging process of semiconductor chip |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116423003B (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116423003B (en) | Tin soldering intelligent evaluation method and system based on data mining | |
| CN111047225B (en) | SMT surface mounting component welding spot quality evaluation method based on edge side model processing | |
| CN111145175B (en) | SMT welding spot defect detection method based on iForest model verification | |
| CN101915769B (en) | Automatic optical inspection method for printed circuit board comprising resistance element | |
| CN101661004B (en) | Visible detection method of welding quality of circuit board based on support vector machine | |
| JP2011158373A (en) | Method for creation of teacher data for use in automatic defect classification, and method and apparatus for automatic defect classification | |
| CN114066848B (en) | FPCA appearance defect visual detection system | |
| CN105334224B (en) | Automate quality testing cloud platform | |
| CN113222913B (en) | Circuit board defect detection positioning method, device and storage medium | |
| CN116860720A (en) | Multi-source heterogeneous data model modeling system oriented to big data analysis | |
| CN114170138B (en) | Method for establishing unsupervised industrial image anomaly detection model, detection method and system | |
| CN115294009A (en) | Method and equipment for detecting welding defects of battery tabs based on machine learning and storage medium | |
| CN117422681A (en) | PCB surface defect detection method based on improved YOLOv8 algorithm | |
| CN109829887B (en) | Image quality evaluation method based on deep neural network | |
| CN109376770B (en) | Grid area identification method and device applied to negative film inspection machine | |
| CN116030013A (en) | Multi-target multi-category damage defect detection method, system and computer storage medium | |
| CN105291104B (en) | Robot remote control method based on cloud platform | |
| CN111382191A (en) | Machine learning identification method based on deep learning | |
| CN112860531A (en) | Block chain wide consensus performance evaluation method based on deep heterogeneous graph neural network | |
| CN111709602A (en) | Reliability evaluation method in ubiquitous power Internet of things system | |
| CN105611556B (en) | A kind of wireless access point device performance superiority and inferiority judgment method and device | |
| CN117196399B (en) | Customer service center operation supervision optimization system based on data analysis | |
| Li et al. | Apple Detection and Yield Estimation Based on YOLOv5 | |
| CN113706452B (en) | Automatic charging pile detection system and detection method based on image recognition | |
| CN117333592B (en) | AI digital population type animation drawing system based on big data fusion training model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |