CN110298408B - Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing - Google Patents
Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing Download PDFInfo
- Publication number
- CN110298408B CN110298408B CN201910595678.1A CN201910595678A CN110298408B CN 110298408 B CN110298408 B CN 110298408B CN 201910595678 A CN201910595678 A CN 201910595678A CN 110298408 B CN110298408 B CN 110298408B
- Authority
- CN
- China
- Prior art keywords
- hub
- data
- model
- identified
- weight data
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 184
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 title claims abstract description 36
- 238000005303 weighing Methods 0.000 title claims abstract description 29
- 238000013499 data model Methods 0.000 claims abstract description 30
- 238000012795 verification Methods 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims description 106
- 239000002994 raw material Substances 0.000 claims description 31
- 238000013507 mapping Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 12
- 239000000463 material Substances 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 230000006872 improvement Effects 0.000 description 10
- 238000003754 machining Methods 0.000 description 5
- 239000011248 coating agent Substances 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 238000003723 Smelting Methods 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 238000004512 die casting Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 229910000838 Al alloy Inorganic materials 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013501 data transformation Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
-
- 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)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention belongs to the technical field of manufacturing and informatization of the hub industry, and particularly relates to a real-time identification method for the hub wheel type of a production processing unit based on intelligent weighing. The invention provides a novel real-time identification method for a hub wheel type of a production and processing unit based on intelligent weighing, which is characterized in that a model identification data model is established by intelligently learning the hub model and corresponding multidimensional characteristic data of 2 or more procedures, including at least theoretical weight data and error data, of the hub model, relevant wheel types are identified through the model identification data model based on the multidimensional characteristic data of the last procedure, including at least actual weight data, the determined hub model is output after the model passes verification completely, and the previous procedures can be traced back in sequence for identification verification after the model passes verification partially, so that the purpose of quickly, accurately and stably identifying the wheel types is achieved.
Description
Technical Field
The invention belongs to the technical field of manufacturing and informatization of a hub industry, and particularly relates to a method for identifying a hub wheel type of a production processing unit in real time based on intelligent weighing.
Background
With the rapid increase of the automobile hub yield in recent years, more and more problems are exposed when various hubs on a production line are identified in an image identification mode. For example, the image recognition is subject to a bottleneck in speed, the special wheel type cannot be recognized, and the like, which all reduce the recognition rate and influence the processing of the aluminum alloy wheel hub. The problem is highlighted in the following four aspects:
firstly, wheel hub production processing unit wheel type image identification speed and limitation problem. In a production processing unit, wheel type identification is a key link of intelligent manufacturing, a large amount of picture data is required to be acquired by only adopting image identification, identification is slow, and the requirement of production rhythm cannot be met;
and secondly, some special wheel type images cannot be effectively identified. For example, the appearance of the wheel type sold overseas is the same as that of the wheel type sold domestically, but the weight is different, so that the problem cannot be solved by image recognition;
and thirdly, the problem of tracing the whole process of the production and processing unit cannot be solved. Because the whole chain tracing of smelting, die casting, machining, coating and packaging is carried out in the hub production process, the whole process tracing from aluminum ingots and aluminum bars to finished product off-line has shape change, ultrahigh temperature and the like, and more attention is paid to the whole process tracing.
Fourthly, the popularization rate of the image recognition system is low. Because few domestic manufacturers exist, the products are monopolized abroad all the time, but the foreign products are expensive in price, the use instruction is obscure and is difficult to understand, and the maintenance is very difficult; meanwhile, due to the regional difference, the technical support of foreign products is not in place, the maintenance cost is high, and the general enterprises cannot afford the support.
Disclosure of Invention
In order to solve the technical problem, the invention provides a novel method for identifying the wheel type of a hub of a production and processing unit in real time based on intelligent weighing.
The specific technical scheme of the invention is as follows:
the invention provides a real-time identification method for a hub wheel type of a production and processing unit based on intelligent weighing, which comprises the following steps:
s1: data acquisition:
acquiring the model number of the wheel hub and multidimensional characteristic data, which at least comprises theoretical weight data and error data, of 2 or more procedures corresponding to the model number of the wheel hub, and constructing 2 or more than 2 training sets;
s2: data processing:
reducing the dimension of each training set through at least one dimension reduction algorithm;
learning the training set after dimensionality reduction through at least one learning algorithm, and establishing 2 or more than 2 model identification data models;
s3: data identification:
when the multi-dimensional characteristic data at least comprising actual weight data of the last procedure is input into a model identification data model corresponding to the last procedure and passes verification completely, outputting the determined model of the hub; when part of the wheel hub passes the verification, returning to the previous procedure adjacent to the last procedure, inputting the multi-dimensional characteristic data at least comprising the actual weight data of the previous procedure into a model identification data model corresponding to the previous procedure, and outputting the determined wheel hub model after the part of the wheel hub passes the verification; and when the other part of the hub model data is not verified, returning to a previous process adjacent to the previous process, and circularly inputting multi-dimensional characteristic data at least comprising actual weight data of the corresponding process into the model identification data model corresponding to the corresponding process until the determined hub model is found.
In a further improvement, step S3 includes the following steps:
s31: when the model identification data model corresponding to the last process Yn is input by the multi-dimensional characteristic data at least comprising the actual weight data of the last process Yn, the model and the theoretical weight data corresponding to the model are preliminarily identified in the error data range of the theoretical weight data of the hub offline of the last process Yn;
s32: judging whether intersection exists between the theoretical weight data preliminarily identified by the last working procedure Yn and the actual weight data, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S33;
s33: returning to the previous procedure Yn-a, inputting multidimensional characteristic data at least comprising actual weight data of the previous procedure Yn-a into a model identification data model corresponding to the previous procedure Yn-a, and preliminarily identifying the model and theoretical weight data corresponding to the model, wherein n represents the number of procedures, a represents the number of cycles, n and a are positive integers, and a is smaller than n;
s34: judging whether intersection exists between the theoretical weight data preliminarily identified in the previous process Yn-a and the actual weight data, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S35;
s35: and judging whether the theoretical weight data preliminarily identified in the previous process Yn-a and the theoretical weight data preliminarily identified in the subsequent process have intersection, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S33.
In a further improvement, the multidimensional data characteristics in step S1 include weight data and error data of the off-line hub in each process, raw materials for manufacturing the target hub, total weight data of the raw materials, and proportion of each material, consumption of the raw materials in the related process, and usage of the related mold; the dimensional data characteristics of the step S3 comprise weight data of offline hubs in each process in the actual production process, raw materials for manufacturing target hubs, total weight data of the raw materials, the proportion of each material, raw material consumption conditions of related processes and use conditions of related dies; the consumption condition of the raw material includes the remaining weight of the material in the corresponding process, and the use condition of the relevant mold includes the thickness of the mold.
In a further improvement, in step S35, when a is equal to 1, it is determined whether there is an intersection between the theoretical weight data preliminarily identified in the previous process Yn-1 and the theoretical weight data preliminarily identified in the process Yn; when a is 2, judging whether the theoretical weight data preliminarily identified by the previous process Yn-2 and the theoretical weight data preliminarily identified by the process Yn and the process Yn-1 have intersection or not; when a is n-1, it is determined whether there is an intersection between the theoretical weight data preliminarily identified in the previous process Y1 and the theoretical weight data preliminarily identified in the process Y2, Y3..
In a further improvement, the identification method further comprises the following steps:
s4: judging whether the number of the target hub-related wheel types identified in the step S3 is greater than 1, if so, performing a step S5, otherwise, identifying the identified wheel type as the determined hub type;
s5: acquiring an image of the off-line hub in the last process, and performing feature extraction on the image to obtain feature data to be detected;
s6: the wheel type is identified through a weighing identification data model based on the characteristic data to be detected, the model of the target hub is obtained through data intersection generated by the wheel type and the relevant wheel type identified in the step S3, and the multi-dimensional characteristic data further comprises image information of the line hub in the last process.
In a further improvement, in step S6, a set B1 ═ B11, B12, …, and B1n is formed by summarizing the wheel types identified by the feature data to be detected, a set B2 ═ B21, B22, …, and B2n are formed by summarizing the relevant wheel types identified in step S3, and an intersection B1 ∞ B2 of the two sets is calculated, and elements in the obtained set are the model of the target hub.
In a further improvement, step S5 includes the following steps:
s51: collecting an overlook image of the offline hub in the last process, and summarizing and extracting the outer size, spoke width duty ratio and spoke number of the hub from the overlook image;
s52: collecting strip-shaped bottom images of the hub, splicing the images, and splicing a complete bottom image of the hub;
s53: extracting the size parameters of the central hole of the hub from the bottom view of the hub;
s54: and (3) dividing an image of the hub assembling surface from the bottom image of the hub, and measuring the offset parameter from the hub assembling surface to the peripheral plane at the bottom of the hub by using a non-contact distance measuring instrument.
In a further improvement, in step S6, the image information of the off-line hub in the last process includes an outer dimension of the hub, a spoke width duty ratio, a spoke number, a dimension parameter of the center hole, and a deviation distance parameter from the hub assembling surface to the outer peripheral plane of the hub bottom, and in the database, a fourth mapping is respectively established between the hub model and each parameter in the image information of the off-line hub in the last process.
In a further improvement, step S1 includes the following steps:
s11: collecting model and corresponding multi-dimensional feature data, removing noise and irrelevant data in the multi-dimensional feature data, and performing data aggregation processing on the multi-dimensional feature data;
s12: converting the processed multi-dimensional feature data into a data set based on a data reduction technology;
s13: and establishing mapping between the hub model and the corresponding multi-dimensional characteristic data which is converted into a data set.
In a further improvement, step S13 includes the following steps:
s131: respectively establishing a first mapping between the model number of the hub and a theoretical weight set of the hub which is offline in each corresponding process, wherein the theoretical weight set is a set C belonging to (m-k, m + k) of theoretical weight m and error data +/-k;
s132: establishing a second mapping between the theoretical weight of the off-line hub in each process and the corresponding process name;
s133: and establishing a third mapping between each process name and the raw material of the corresponding target hub, the total weight of the raw material, the proportion of each material, the raw material consumption condition of the related process or the use condition of the related mold.
The invention has the following beneficial effects:
the invention provides a novel real-time identification method for a hub wheel type of a production and processing unit based on intelligent weighing, which is characterized in that a model identification data model is established by intelligently learning the hub model and corresponding multidimensional characteristic data of 2 or more procedures, including at least theoretical weight data and error data, of the hub model, relevant wheel types are identified through the model identification data model based on the multidimensional characteristic data of the last procedure, including at least actual weight data, the determined hub model is output after the model passes verification completely, and the previous procedures can be traced back in sequence for identification verification after the model passes verification partially, so that the purpose of quickly, accurately and stably identifying the wheel types is achieved.
Drawings
FIG. 1 is a flowchart of a method for identifying wheel hub type of a production and processing unit in real time based on intelligent weighing in embodiment 1;
FIG. 2 is a flowchart of step S3 in example 2;
FIG. 3 is a flowchart of a method for identifying wheel hub type of a production and processing unit in real time based on intelligent weighing in embodiment 3;
FIG. 4 is a flowchart of step S5 in example 4;
FIG. 5 is a flowchart showing step S1 of embodiment 5;
FIG. 6 is a flowchart of step S13 in example 5.
Detailed Description
The invention is further described with reference to the following figures and examples, which are provided only for illustrating the inventive content of the present invention and are not intended to limit the scope of the invention.
The steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flow diagrams, in some cases, the steps described may be performed in an order different than here.
Example 1
The embodiment 1 of the invention provides a method for identifying a hub wheel type of a production and processing unit in real time based on intelligent weighing, which comprises the following steps of:
s1: data acquisition:
acquiring the model number of the wheel hub and multidimensional characteristic data, which at least comprises theoretical weight data and error data, of 2 or more procedures corresponding to the model number of the wheel hub, and constructing 2 or more than 2 training sets;
s2: data processing:
performing dimension reduction on each training set through at least one dimension reduction algorithm, wherein the dimension reduction algorithm comprises but is not limited to PCA theory;
learning the training set after dimensionality reduction through at least one learning algorithm, and establishing 2 or more than 2 model identification data models, wherein the learning algorithm comprises but is not limited to an artificial intelligence deep learning technology;
s3: data identification:
when the multi-dimensional characteristic data at least comprising actual weight data of the last procedure is input into a model identification data model corresponding to the last procedure and passes verification completely, outputting the determined model of the hub; when part of the wheel hub passes the verification, returning to the previous procedure adjacent to the last procedure, inputting the multi-dimensional characteristic data at least comprising the actual weight data of the previous procedure into a model identification data model corresponding to the previous procedure, and outputting the determined wheel hub model after the part of the wheel hub passes the verification; and when the other part of the hub model data is not verified, returning to a previous process adjacent to the previous process, and circularly inputting multi-dimensional characteristic data at least comprising actual weight data of the corresponding process into the model identification data model corresponding to the corresponding process until the determined hub model is found.
The invention provides a novel real-time identification method for a hub wheel type of a production and processing unit based on intelligent weighing, which is characterized in that a model identification data model is established by intelligently learning the hub model and corresponding multidimensional characteristic data of 2 or more procedures, including at least theoretical weight data and error data, of the hub model, relevant wheel types are identified through the model identification data model based on the multidimensional characteristic data of the last procedure, including at least actual weight data, the determined hub model is output after the model passes verification completely, and the previous procedures can be traced back in sequence for identification verification after the model passes verification partially, so that the purpose of quickly, accurately and stably identifying the wheel types is achieved.
The existing hub models and the corresponding multidimensional characteristic data in this embodiment are models of all existing hubs and various corresponding characteristic parameters when corresponding hubs are produced.
Example 2
A method for identifying a hub wheel type of a production and processing unit in real time based on intelligent weighing, which is different from embodiment 1 in that, as shown in fig. 2, step S3 includes the following steps:
s31: when the model identification data model corresponding to the last process Yn is input by the multi-dimensional characteristic data at least comprising the actual weight data of the last process Yn, the model and the theoretical weight data corresponding to the model are preliminarily identified in the error data range of the theoretical weight data of the hub offline of the last process Yn;
s32: judging whether intersection exists between the theoretical weight data preliminarily identified by the last procedure Yn and the actual weight data, if not, determining that the model corresponding to the identified theoretical weight data is the determined hub model, and if so, performing step S33;
s33: returning to the previous procedure Yn-a, inputting multidimensional characteristic data at least comprising actual weight data of the previous procedure Yn-a into a model identification data model corresponding to the previous procedure Yn-a, and preliminarily identifying the model and theoretical weight data corresponding to the model, wherein n represents the number of procedures, a represents the number of cycles, n and a are positive integers, and a is smaller than n;
s34: judging whether intersection exists between the theoretical weight data preliminarily identified in the previous process Yn-a and the actual weight data, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S35;
s35: and judging whether intersection exists between the theoretical weight data preliminarily identified in the previous process Yn-a and the theoretical weight data preliminarily identified in the subsequent process, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S33.
In this embodiment, the multidimensional data characteristics described in step S1 include weight data and error data of the off-line hub in each process, raw materials for manufacturing the target hub, total weight data of the raw materials, and proportions of the raw materials, raw material consumption in the related process, and usage of the related mold; the dimensional data characteristics of the step S3 comprise weight data of offline hubs in each process in the actual production process, raw materials for manufacturing target hubs, total weight data of the raw materials, the proportion of each material, raw material consumption conditions of related processes and use conditions of related dies; the consumption condition of the raw material includes the remaining weight of the material in the corresponding process, and the use condition of the relevant mold includes the thickness of the mold.
In this embodiment, when a is equal to 1 in step S35, it is determined whether there is an intersection between the theoretical weight data preliminarily identified in the previous process Yn-1 and the theoretical weight data preliminarily identified in the process Yn; when a is 2, judging whether the theoretical weight data preliminarily identified by the previous process Yn-2 and the theoretical weight data preliminarily identified by the process Yn and the process Yn-1 have intersection or not; when a is n-1, it is determined whether there is an intersection between the theoretical weight data preliminarily identified in the previous process Y1 and the theoretical weight data preliminarily identified in the process Y2, Y3..
In the embodiment, the hub model is identified by weight through the method; the production of the wheel hub generally comprises 5 working procedures of smelting, die casting, heat treatment, machining and coating, when the coating is off-line, multi-dimensional characteristic data at least comprising actual weight data of the wheel hub are collected, the obtained actual weight of the wheel hub can be accurately matched with theoretical weight provided by a reference model within an error allowable range, and therefore the off-line wheel type can be correctly identified; if the actual weight of the coated offline hub is intersected with the identified theoretical weight data of the corresponding model, the five processes are traced in sequence, firstly, the machining process is carried out, multidimensional characteristic data at least comprising the actual weight data of the machining process are input into a corresponding model identification data model, the model and the corresponding theoretical weight data are output, if the actual weight of the coated offline hub is intersected with the identified theoretical weight data of the corresponding model, whether intersection exists between the theoretical weight data of the corresponding model identified by the coated offline hub and the theoretical weight data of the corresponding model identified by the machining process is judged, and the process is repeated until the determined hub model is identified.
Example 3
A method for identifying a hub wheel type of a production and processing unit in real time based on intelligent weighing, which is different from embodiment 1, and as shown in fig. 3, the method further comprises the following steps:
s4: judging whether the number of the target hub-related wheel types identified in the step S3 is greater than 1, if so, performing a step S5, otherwise, identifying the identified wheel type as the determined hub type;
s5: acquiring an image of the off-line hub in the last process, and performing feature extraction on the image to obtain feature data to be detected;
s6: the wheel type is identified through a weighing identification data model based on the characteristic data to be detected, the model of the target hub is obtained through data intersection generated by the wheel type and the relevant wheel type identified in the step S3, and the multi-dimensional characteristic data further comprises image information of the line hub in the last process.
In step S6 of this embodiment, a set B1 ═ B11, B12, …, B1n is formed by summarizing the wheel types identified by the feature data to be detected, a set B2 ═ B21, B22, …, B2n is formed by summarizing the relevant wheel types identified in step S3, and an intersection B1 ∞ B2 of the two sets is calculated, where elements in the obtained set are the model of the target hub.
In this embodiment, the image information of the off-line hub in the last process in step S6 includes the outer dimension of the hub, the duty ratio of the spoke width, the number of spokes, the dimensional parameter of the center hole, and the offset parameter from the hub mounting surface to the outer peripheral plane of the hub bottom, and fourth mappings are respectively established between the hub model and each parameter in the image information of the off-line hub in the last process in the database.
In the embodiment, the hub model is further identified by the method; when the wheel types identified by the weights are multiple, images of the coated off-line wheel hub need to be acquired, the wheel types are identified through a model identification data model based on the characteristic data to be detected of the images, if the identified wheel types are B1 ═ B1, B2, B3 and B4, the wheel types identified by the weights in the step S4 are B2 ═ B1 and B5, then intersection is calculated, B1 ═ B2 ═ B1, and the model of the target wheel hub is B1.
According to the invention, the identification method with weight identification as the main part and image identification as the auxiliary part is adopted, so that the identification rate and the identification degree of the wheel type of the hub can be greatly improved, and the popularization rate is higher.
Example 4
A method for identifying a hub wheel type of a production and processing unit in real time based on intelligent weighing, which is different from embodiment 3 in that, as shown in fig. 4, step S5 in this embodiment includes the following steps:
s51: collecting an overlook image of the offline hub in the last process, and summarizing and extracting the outer size, spoke width duty ratio and spoke number of the hub from the overlook image;
s52: collecting strip-shaped bottom images of the hub, splicing the images, and splicing a complete bottom image of the hub;
s53: extracting the size parameters of the central hole of the hub from the bottom view of the hub;
s54: and (3) dividing an image of the hub assembling surface from the bottom image of the hub, and measuring the offset parameter from the hub assembling surface to the peripheral plane at the bottom of the hub by using a non-contact distance measuring instrument.
In the embodiment, the extraction of the feature data in the image recognition is further limited, the extraction of the feature data is greatly reduced, the image recognition method is simplified, and the recognition degree is greatly improved by using the recognition method combining the weight and the image.
Example 5
A method for identifying a hub wheel type of a production and processing unit in real time based on intelligent weighing, which is different from embodiment 2 in that, as shown in fig. 5, step S1 in this embodiment includes the following steps:
s11: collecting model and corresponding multi-dimensional feature data, removing noise and irrelevant data in the multi-dimensional feature data, and performing data aggregation processing on the multi-dimensional feature data;
s12: converting the processed multi-dimensional feature data into a data set based on a data reduction technology;
s13: and establishing mapping between the hub model and the corresponding multi-dimensional characteristic data which is converted into a data set.
As shown in fig. 6, step S13 in this embodiment includes the following steps:
s131: respectively establishing a first mapping between the model number of the hub and a theoretical weight set of the hub which is offline in each corresponding process, wherein the theoretical weight set is a set C belonging to (m-k, m + k) of theoretical weight m and error data +/-k;
s132: establishing a second mapping between the theoretical weight of the off-line hub in each process and the corresponding process name;
s133: and establishing a third mapping between the names of the processes and the raw materials of the corresponding target hub, the total weight of the raw materials, the proportion of the materials, the consumption condition of the raw materials of the related processes or the use condition of the related die.
The data acquired by the training set needs to be preprocessed by data cleaning, data integration, data transformation, data reduction and the like, and then a mapping relation is established, so that the data storage quality can be improved, and the model identification data model can be conveniently learned and trained; the preprocessed data are classified to establish a mapping relation, so that the relevant wheel types can be identified quickly and accurately; because the actual weight has errors, the training set needs to collect theoretical weight data and corresponding error data in each process.
Since the method description of the invention is implemented in a computer system. The computer system may be provided in a processor of a server or a client, for example. For example, the methods described herein may be implemented as software executable with control logic that is executed by a CPU in a server. The functionality described herein may be implemented as a set of program instructions stored in a non-transitory tangible computer readable medium. When implemented in this manner, the computer program comprises a set of instructions which, when executed by a computer, cause the computer to perform a method capable of carrying out the functions described above. Programmable logic may be installed temporarily or permanently in a non-transitory tangible computer-readable medium, such as a read-only memory chip, a computer memory, a disk, or other storage medium. In addition to being implemented in software, the logic described herein may be embodied using a discrete component, an integrated circuit, programmable logic used in conjunction with a programmable logic device such as a Field Programmable Gate Array (FPGA) or microprocessor, or any other device including any combination thereof. All such implementations are intended to fall within the scope of the present invention.
The above-mentioned embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solution of the present invention by those skilled in the art should fall within the protection scope defined by the claims of the present invention without departing from the spirit of the present invention.
Claims (10)
1. A real-time identification method for hub wheel type of a production and processing unit based on intelligent weighing is characterized by comprising the following steps:
s1: data acquisition:
acquiring the model number of the wheel hub and multidimensional characteristic data, which at least comprises theoretical weight data and error data, of 2 or more procedures corresponding to the model number of the wheel hub, and constructing 2 or more than 2 training sets;
s2: data processing:
reducing the dimension of each training set through at least one dimension reduction algorithm;
learning the training set after dimensionality reduction through at least one learning algorithm, and establishing 2 or more than 2 model identification data models;
s3: data identification:
when the multi-dimensional characteristic data at least comprising actual weight data of the last procedure is input into a model identification data model corresponding to the last procedure and passes verification completely, outputting the determined model of the hub;
when part of the hub passes the verification, returning to a previous process adjacent to the last process, inputting the multi-dimensional characteristic data at least comprising actual weight data of the previous process into a model identification data model corresponding to the previous process, and outputting the determined hub model after the part passes the verification; and when the other part of the hub model data is not verified, returning to a previous process adjacent to the previous process, and circularly inputting multi-dimensional characteristic data at least comprising actual weight data of the corresponding process into the model identification data model corresponding to the corresponding process until the determined hub model is found.
2. The intelligent weighing-based real-time identification method for hub wheel types of production and processing units according to claim 1, wherein the step S3 comprises the following steps:
s31: when the model identification data model corresponding to the last process Yn is input by the multi-dimensional characteristic data at least comprising the actual weight data of the last process Yn, the model and the theoretical weight data corresponding to the model are preliminarily identified in the error data range of the theoretical weight data of the hub offline of the last process Yn;
s32: judging whether intersection exists between the theoretical weight data preliminarily identified by the last procedure Yn and the actual weight data, if not, determining that the model corresponding to the identified theoretical weight data is the determined hub model, and if so, performing step S33;
s33: returning to the previous procedure Yn-a, inputting multidimensional characteristic data at least comprising actual weight data of the previous procedure Yn-a into a model identification data model corresponding to the previous procedure Yn-a, and preliminarily identifying the model and theoretical weight data corresponding to the model, wherein n represents the number of procedures, a represents the number of cycles, n and a are positive integers, and a is smaller than n;
s34: judging whether intersection exists between the theoretical weight data preliminarily identified in the previous process Yn-a and the actual weight data, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S35;
s35: and judging whether intersection exists between the theoretical weight data preliminarily identified in the previous process Yn-a and the theoretical weight data preliminarily identified in the subsequent process, if not, determining the model corresponding to the identified theoretical weight data as the determined hub model, and if so, performing step S33.
3. The method for identifying the hub wheel type of the production and processing unit in real time based on the intelligent weighing as claimed in claim 1, wherein the multidimensional characteristic data in the step S1 comprises weight data and error data of the offline hub of each process, raw materials for manufacturing the target hub, total weight data of the raw materials, the proportion of each material, consumption of the raw materials of the related process and usage of the related die; the multidimensional characteristic data of the step S3 comprises weight data of offline hubs in each process in the actual production process, raw materials for manufacturing target hubs, total weight data of the raw materials, the proportion of each material, raw material consumption conditions of related processes and use conditions of related dies; the consumption condition of the raw material includes the remaining weight of the material in the corresponding process, and the use condition of the relevant mold includes the thickness of the mold.
4. The method for identifying the hub wheel type of the production and processing unit in real time based on the intelligent weighing as claimed in claim 2, wherein in step S35, when a is 1, it is determined whether the theoretical weight data preliminarily identified by the previous process Yn-1 and the theoretical weight data preliminarily identified by the process Yn intersect; when a is 2, judging whether the theoretical weight data preliminarily identified by the previous process Yn-2 and the theoretical weight data preliminarily identified by the process Yn and the process Yn-1 have intersection or not; when a is n-1, it is determined whether there is an intersection between the theoretical weight data preliminarily identified in the previous process Y1 and the theoretical weight data preliminarily identified in the process Y2, Y3..
5. The intelligent weighing-based real-time identification method for hub wheel types of production and processing units according to claim 1, further comprising the following steps:
s4: judging whether the number of the target wheel hub relevant wheel types identified in the step S3 is greater than 1, if so, performing the step S5, and if not, the identified wheel type is the determined wheel hub type;
s5: acquiring an image of the off-line hub in the last process, and performing feature extraction on the image to obtain feature data to be detected;
s6: the wheel type is identified through a weighing identification data model based on the characteristic data to be detected, the model of the target hub is obtained through data intersection generated by the wheel type and the relevant wheel type identified in the step S3, and the multi-dimensional characteristic data further comprises image information of the line hub in the last process.
6. The method for identifying the wheel type of the hub of the production and processing unit in real time based on the intelligent weighing pair as claimed in claim 5, wherein in step S6, the wheel types identified by the characteristic data to be detected are collected to form a set B1 ═ B11, B12, … and B1n, the relevant wheel types identified in step S3 are collected to form a set B2 ═ B21, B22, … and B2n, and the intersection B1 &B 2 of the two sets is calculated, wherein the elements in the set are the model of the target hub.
7. The intelligent weighing-based real-time identification method for hub wheel types of production and processing units according to claim 5, wherein the step S5 comprises the following steps:
s51: collecting an overlook image of the offline hub in the last process, and summarizing and extracting the outer size, spoke width duty ratio and spoke number of the hub from the overlook image;
s52: collecting strip-shaped bottom images of the hub, splicing the images, and splicing a complete bottom image of the hub;
s53: extracting the size parameters of the central hole of the hub from the bottom view of the hub;
s54: and (3) dividing an image of the hub assembling surface from the bottom image of the hub, and measuring the offset parameter from the hub assembling surface to the peripheral plane at the bottom of the hub by using a non-contact distance measuring instrument.
8. The method for identifying the hub wheel type of the production and processing unit in real time based on the intelligent weighing according to claim 5, wherein the image information of the offline hub in the last process in the step S6 includes the outer dimension of the hub, the duty ratio of the spoke width, the number of spokes, the dimensional parameters of the central hole and the offset distance parameters from the hub assembling surface to the outer peripheral plane of the bottom of the hub, and a fourth mapping is respectively established between the hub type and each parameter in the image information of the offline hub in the corresponding last process in the database.
9. The intelligent weighing-based real-time identification method for hub wheel types of production and processing units according to claim 3, wherein the step S1 comprises the following steps:
s11: collecting model and corresponding multi-dimensional feature data, removing noise and irrelevant data in the multi-dimensional feature data, and performing data aggregation processing on the multi-dimensional feature data;
s12: converting the processed multi-dimensional feature data into a data set based on a data reduction technology;
s13: and establishing mapping between the hub model and the corresponding multi-dimensional characteristic data which is converted into a data set.
10. The intelligent weighing-based real-time identification method for hub wheel types of production and processing units according to claim 9, wherein the step S13 comprises the following steps:
s131: respectively establishing a first mapping between the model number of the hub and a theoretical weight set of the hub which is offline in each corresponding process, wherein the theoretical weight set is a set C belonging to (m-k, m + k) of theoretical weight m and error data +/-k;
s132: establishing a second mapping between the theoretical weight of the off-line hub in each process and the corresponding process name;
s133: and establishing a third mapping between the names of the processes and the raw materials of the corresponding target hub, the total weight of the raw materials, the proportion of the materials, the consumption condition of the raw materials of the related processes or the use condition of the related die.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910595678.1A CN110298408B (en) | 2019-07-03 | 2019-07-03 | Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910595678.1A CN110298408B (en) | 2019-07-03 | 2019-07-03 | Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110298408A CN110298408A (en) | 2019-10-01 |
CN110298408B true CN110298408B (en) | 2022-05-20 |
Family
ID=68030148
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910595678.1A Expired - Fee Related CN110298408B (en) | 2019-07-03 | 2019-07-03 | Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110298408B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111243678B (en) * | 2020-01-07 | 2023-05-23 | 北京唐颐惠康生物医学技术有限公司 | Cell inventory security guarantee method and system based on locking technology |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050185833A1 (en) * | 2004-02-20 | 2005-08-25 | Pla Jesus B. | System for recognizing wheels by artificial vision |
CN103090790A (en) * | 2012-12-21 | 2013-05-08 | 宁波赛恩斯智能科技有限公司 | Automatic identification device and automatic identification method for hub |
-
2019
- 2019-07-03 CN CN201910595678.1A patent/CN110298408B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050185833A1 (en) * | 2004-02-20 | 2005-08-25 | Pla Jesus B. | System for recognizing wheels by artificial vision |
CN103090790A (en) * | 2012-12-21 | 2013-05-08 | 宁波赛恩斯智能科技有限公司 | Automatic identification device and automatic identification method for hub |
Non-Patent Citations (1)
Title |
---|
基于形状匹配及纹理筛选的汽车轮毂型号识别;程淑红等;《仪器仪表学报》;20170930;第2299-2306页 * |
Also Published As
Publication number | Publication date |
---|---|
CN110298408A (en) | 2019-10-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105989538B (en) | Automatic trading system and automatic trading method for financial products | |
CN103424409B (en) | Vision detecting system based on DSP | |
US12001516B2 (en) | Method and assistance system for parameterizing an anomaly detection method | |
WO2018092747A1 (en) | Learned model generation method, learned model generation device, signal data discrimination method, signal data discrimination device, and signal data discrimination program | |
CN113393211B (en) | Method and system for intelligently improving automatic production efficiency | |
CN110298408B (en) | Real-time identification method for hub wheel type of production and processing unit based on intelligent weighing | |
US20210394247A1 (en) | Method and system for forming a stamped component using a stamping simulation model | |
CN113610851B (en) | Packaging decoration quality inspection method and system based on machine vision | |
CN114495498A (en) | Traffic data distribution effectiveness judging method and device | |
CN108073464A (en) | A kind of time series data abnormal point detecting method and device based on speed and acceleration | |
CN113112292A (en) | Supervised commodity intelligent recommendation method and system in bulk commodity transaction | |
CN116522096A (en) | Three-dimensional digital twin content intelligent manufacturing method based on motion capture | |
US11420325B2 (en) | Method, apparatus and system for controlling a robot, and storage medium | |
CN112991093B (en) | Electric larceny detection method and system based on edge calculation | |
CN112488176B (en) | Processing characteristic recognition method based on triangular mesh and neural network | |
Yang et al. | Cherry recognition based on color channel transform | |
CN115294009A (en) | Method and equipment for detecting welding defects of battery tabs based on machine learning and storage medium | |
Hillsman et al. | A semi-automatic mold cost estimation framework based upon geometry similarity | |
Dong | Improvement of the model by preprocessing big data of tapping temperature prediction industry | |
CN114519820B (en) | Automatic citrus screening correction control method and system based on machine vision | |
CN118072113B (en) | Multi-sense paper production intelligent quality control method and system | |
US20230316718A1 (en) | Learning model generating method and inspection device | |
CN118277206A (en) | Visual monitoring method, system and storage medium for running process of cluster GPU | |
CN110287522B (en) | Automatic generation and distribution method of screw holes on insert | |
CN117078287A (en) | Target cost management and control method, system, equipment and medium |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220520 |
|
CF01 | Termination of patent right due to non-payment of annual fee |