CN116681698B - Spring automatic assembly quality detection method and system - Google Patents

Spring automatic assembly quality detection method and system Download PDF

Info

Publication number
CN116681698B
CN116681698B CN202310939134.9A CN202310939134A CN116681698B CN 116681698 B CN116681698 B CN 116681698B CN 202310939134 A CN202310939134 A CN 202310939134A CN 116681698 B CN116681698 B CN 116681698B
Authority
CN
China
Prior art keywords
spring
quality
quality detection
detection result
outputting
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.)
Active
Application number
CN202310939134.9A
Other languages
Chinese (zh)
Other versions
CN116681698A (en
Inventor
唐瑞阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sidelama Machinery Taicang Co ltd
Original Assignee
Sidelama Machinery Taicang Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sidelama Machinery Taicang Co ltd filed Critical Sidelama Machinery Taicang Co ltd
Priority to CN202310939134.9A priority Critical patent/CN116681698B/en
Publication of CN116681698A publication Critical patent/CN116681698A/en
Application granted granted Critical
Publication of CN116681698B publication Critical patent/CN116681698B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/0006—Industrial image inspection using a design-rule based approach
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0499—Feedforward networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component
    • 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)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Image Analysis (AREA)

Abstract

The application relates to the technical field of spring quality detection, and provides a spring automatic assembly quality detection method and system, wherein the method comprises the following steps: outputting a fitting simulation model; acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model; outputting a first image set and a second image set; acquiring a first coincidence quality index; acquiring a second quality index; outputting a first quality detection result and a second quality detection result; the method comprises the steps of generating a quality detection report, solving the technical problem of low quality detection efficiency of spring assembly on a mass-production spring automatic assembly production line, and realizing the technical effects of rapidly and accurately evaluating the automatic assembly quality of springs and guaranteeing the consistency and stability of products by identifying the spring spacing of a overlook reference surface image and evaluating the stability of the surface spacing of a spring inner ring and a spring insert on the mass-production spring automatic assembly production line.

Description

Spring automatic assembly quality detection method and system
Technical Field
The application relates to the technical field related to spring quality detection, in particular to a spring automatic assembly quality detection method and system.
Background
Along with the continuous development of artificial intelligence, the quality detection of spring assembly is continuously intelligent and efficient, and conventionally, a robot is used for carrying out operations such as scanning and photographing on the assembled spring to detect whether indexes such as appearance, size and shape of the spring meet requirements; and detecting the assembled spring by using a sensor, and detecting whether indexes such as elasticity, strength, rigidity and the like of the spring meet the requirements.
The robot detection and the sensor detection are applied to the quality detection of spring assembly, can rapidly detect the quality of the spring, and can automatically operate, thereby improving the production efficiency and the production quality. However, on the one hand, due to the high investment costs, it is not possible to put into and apply to the automatic assembly line of springs for mass production, and on the other hand, it is necessary to equip the corresponding equipment and software and to carry out corresponding technical training, which limits the application in some lines.
In summary, the technical problem of low quality detection efficiency of spring assembly on a mass-produced spring automatic assembly production line exists in the prior art.
Disclosure of Invention
The application provides a spring automatic assembly quality detection method and system, and aims to solve the technical problem of low quality detection efficiency of spring assembly on a spring automatic assembly production line in mass production in the prior art.
In view of the above problems, the present application provides a method and a system for detecting the quality of automatic assembly of springs.
In a first aspect of the present disclosure, a method for detecting automatic assembly quality of a spring is provided, where the method includes: modeling the target assembly by the connection simulation system, and outputting an assembly simulation model; acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model; respectively acquiring multi-angle images of a target assembly according to the first state and the second state, and outputting a first image set and a second image set, wherein each image set comprises a side-looking reference surface image and a top-looking reference surface image; acquiring a first coincidence quality index by recognizing the spring circle distance of the side-looking reference plane image, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs; acquiring a second overlapping quality index by carrying out spring interval identification on the overlook reference surface image, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the spring inner ring and the spring embedded part; outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index; and generating a quality detection report based on the first quality detection result and the second quality detection result.
In another aspect of the present disclosure, there is provided a spring automatic assembly quality detection system, wherein the system comprises: the simulation modeling module is used for being connected with the simulation system to model the target assembly and outputting an assembly simulation model; the state acquisition module is used for acquiring a first state when the spring is relaxed and a second state when the spring is compressed in the assembly simulation model; the image acquisition module is used for respectively acquiring the multi-angle images of the target assembly according to the first state and the second state and outputting a first image set and a second image set, wherein each image set comprises a side-view reference surface image and a top-view reference surface image; the spring circle distance identification module is used for obtaining a first coincidence quality index by carrying out spring circle distance identification on the side-view reference surface image, wherein the first coincidence quality index is used for identifying the distance error degree of two adjacent circles of springs; the spring interval identification module is used for acquiring a second overlapping quality index by carrying out spring interval identification on the overlook reference surface image, wherein the second overlapping quality index is used for identifying the stability degree of the surface interval between the spring inner ring and the spring embedded part; the quality detection result output module is used for outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index; and the quality detection report generation module is used for generating a quality detection report based on the first quality detection result and the second quality detection result.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the connection simulation system is adopted to model the target assembly, so that an assembly simulation model is output; acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model; respectively carrying out multi-angle image acquisition on the target assembly according to the first state and the second state, and outputting a first image set and a second image set, wherein each image set comprises a side-looking reference surface image and a top-looking reference surface image; the method comprises the steps of obtaining a first coincidence quality index by recognizing a spring circle distance of a side-looking reference plane image, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs; acquiring a second overlapping quality index by identifying the spring interval of the overlooking reference surface image, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the spring inner ring and the spring embedded part; outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index; based on the first quality detection result and the second quality detection result, a quality detection report is generated, and the technical effects that the stability degree of the surface spacing between the inner ring of the spring and the spring embedded part is evaluated by identifying the spring spacing on the overlooking reference surface image on a mass production spring automatic assembly production line, the automatic assembly quality of the spring is rapidly and accurately evaluated, and the consistency and stability of products are ensured are realized.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic diagram of a possible flow chart of a spring automatic assembly quality detection method according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a possible process of outputting a first overlapping quality index in a spring automatic assembly quality detection method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a possible flow chart for generating a first quality detection result and a second quality detection result in a spring automatic assembly quality detection method according to an embodiment of the present application;
fig. 4 is a schematic diagram of a possible structure of an automatic spring assembly quality detection system according to an embodiment of the present application.
Reference numerals illustrate: the device comprises a simulation modeling module 100, a state acquisition module 200, an image acquisition module 300, a spring coil distance identification module 400, a spring interval identification module 500, a quality detection result output module 600 and a quality detection report generation module 700.
Detailed Description
The embodiment of the application provides a spring automatic assembly quality detection method and system, which solve the technical problem of low quality detection efficiency of spring assembly on a mass production spring automatic assembly production line, and realize the technical effects of rapidly and accurately evaluating the automatic assembly quality of springs and guaranteeing the consistency and stability of products by identifying the spring spacing of a top-down reference surface image on the mass production spring automatic assembly production line and evaluating the stability of the surface spacing of a spring inner ring and a spring embedded part.
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 detecting automatic assembly quality of a spring, where the method includes:
s10: modeling the target assembly by the connection simulation system, and outputting an assembly simulation model;
s20: acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model;
s30: respectively acquiring multi-angle images of a target assembly according to the first state and the second state, and outputting a first image set and a second image set, wherein each image set comprises a side-looking reference surface image and a top-looking reference surface image;
specifically, compared with the technical means such as robot detection and sensor detection, for the mass production of the spring automatic assembly production line, some more economical technical means need to be considered for quality detection. For example, a simple detection device such as a camera or a photoelectric sensor may be used to detect whether or not the indicators such as the appearance, the size, the shape, etc. of the spring meet the requirements. In addition, algorithms based on techniques such as data analysis and machine learning can be utilized to automatically identify and classify the quality information of the springs. The method has the advantages of ensuring accurate detection and relatively low investment cost, and can be more suitable for application in the spring automatic assembly production line for mass production.
The connection simulation system is used for importing a design drawing of a target assembly into corresponding simulation software, obtaining an assembly simulation model of the target assembly by performing three-dimensional modeling motion simulation on parts, and performing modeling on the target assembly by using computer software to convert the model into a digital model so as to perform various simulation analyses such as strength, rigidity and the like;
the first state refers to a state when the spring is in a relaxed state, and the length of the spring is longer at the moment and is not subjected to compression force; the second state is a state when the spring is compressed, and the length of the spring is short at the moment and is subjected to a certain compression force; acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model, and obtaining the forms of the spring in two states, namely a spring relaxation state and a spring compression state, of a spring part of a target fitting through simulation analysis so as to acquire subsequent images and detect and analyze quality;
respectively acquiring multi-angle images of a target assembly according to the first state and the second state, capturing views of the target assembly under different angles, taking the multi-angle images corresponding to the first state of the target assembly as a first image set, taking the multi-angle images corresponding to the second state of the target assembly as a second image set, and outputting the first image set and the second image set, wherein each image set comprises a side-view reference surface image and a top-view reference surface image, and the side-view reference surface image refers to a spring image shot from a side surface and is used for identifying spring circle distances; the overlooking reference plane image refers to a spring image overlooking from the upper side and is used for identifying the distance between springs;
the background of the side-view reference plane image and the background of the overlook reference plane image are white, gray level conversion in the images correspond to the data of the springs, and based on the gray level conversion, the method specifically comprises the following steps of: and (3) calculating data of the index spring by point operation: g (x, y) =t [ f (x, y) ], f (x, y) being used to characterize the gray scale of the input spring image; g (x, y) is used to characterize the corresponding geometric data for which the image gray data changes meet the definition. T [ ] is used to characterize the gray scale transformation function. The point operation is a point-by-point operation gray level conversion function of pixels, which can be simplified to s=tr, and the point operation is calculated to obtain the data of the spring by comparing the gray level data change of the image.
By combining the image recognition technology and the spring quality index, the automatic quality detection is realized by acquiring a simulation model and a multi-angle image of the target assembly. The method has the advantages of low labor cost, high efficiency, high detection precision and the like, and can effectively improve the production efficiency and the product quality.
S40: acquiring a first coincidence quality index by recognizing the spring circle distance of the side-looking reference plane image, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs;
as shown in fig. 2, step S40 includes the steps of:
s41: acquiring the number of spring turns N which is a positive integer greater than or equal to 0 by identifying the side view reference plane image;
s42: identifying the coil distance between two adjacent coils in the spring coil number N, and outputting N-1 coil distance;
s43: and carrying out fluctuation error analysis according to the N-1 group of circle distances, and outputting the first coincidence quality index.
Specifically, the spring circle distance recognition is carried out on the side view reference plane image, and a first coincidence quality index is obtained, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs; in general, the method comprises the steps of analyzing and identifying a side-view reference surface image of a spring, acquiring the number of turns N of the spring and the circle distance data between N-1 groups of adjacent turns, further calculating a first superposition quality index of the spring, and specifically comprising the following steps:
the number of turns N of the spring of the target assembly is the number of turns on the spring, N is a positive integer greater than or equal to 0, and the number of turns of the spring is identified by the side view reference surface image: opening the side-looking reference plane image by using image processing software and amplifying to a proper size; marking the end-to-end turns of the spring on the image using a selected tool or a painting tool to determine the length and number of turns N of the spring;
the coil distance of the spring refers to the distance between two adjacent coils, the number of coils N is counted to form N-1 group of coil distance data, and the side-looking reference plane image is used for recognition: opening the image of the springs by using image processing software, and adjusting the size and brightness of the image so as to clearly see the detail of each spring ring; selecting a measuring tool or a line segment tool, measuring the distance between two adjacent circles on an image, and recording the numerical value measured each time; repeating the steps until the distances between all adjacent two circles are measured, and obtaining circle distance data between N-1 groups of adjacent circles;
the fluctuation error analysis refers to counting and analyzing the loop distance data to determine the quality condition of the spring; the first coincidence quality index is used for marking the interval error degree of two adjacent circles; carrying out fluctuation error analysis according to the N-1 group of circle distances, calculating the average circle distance between N-1 group of adjacent circles, and taking the average circle distance between N-1 group of adjacent circles as a reference circle distance; calculating the difference between each group of circle distances and the reference circle distance; adding all the differences and averaging to obtain an average deviation; calculating the square of the difference between each set of loop distances and the reference loop distance, and adding the squares of the difference between each set of loop distances and the reference loop distance; dividing the square sum by N-1, and then squaring to obtain a standard deviation; calculating the sum of the standard deviation and the average deviation, and taking the sum of the standard deviation and the average deviation as a first coincidence quality index;
through technical means such as image recognition and fluctuation error analysis, the quality of the spring can be detected rapidly and accurately, parameters such as the number of spring turns and the coil spacing can be output, and meanwhile, the superposition quality index of the spring can be output, so that the comprehensive monitoring and control of the quality of the spring are realized, the production quality of the spring is improved finally, and the defective rate is reduced.
S50: acquiring a second overlapping quality index by carrying out spring interval identification on the overlook reference surface image, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the spring inner ring and the spring embedded part;
step S50 includes the steps of:
s51: acquiring a coil wire diameter of the target assembly;
s52: identifying the overlooking reference plane image, and determining a clearance value of one side of each circle of spring close to the surface of the spring insert;
s53: and based on the coil wire diameters, performing error comparison on the gap values of each layer of coils, and outputting the second overlapping quality index.
Specifically, the spring interval recognition is carried out on the overlook reference plane image, and a second overlapping quality index is obtained, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the inner ring of the spring and the spring embedded part; the coil wire diameter of the spring refers to the wire diameter of the coil material; the gap value refers to the gap size of one side of the spring ring, which is close to the surface of the spring insert, the overlooking angle of the spring and the spring insert is projected into two concentric circles, and the gap value is the difference between the radii corresponding to the two concentric circles obtained by the overlooking angle projection of the spring and the spring insert;
preparing a spring and a measuring tool of a target assembly; placing the spring on a plane to enable the spring to be in a natural state and not to be distorted or deformed; the wire diameter of the spring is measured by using a proper measuring tool such as a caliper and the like and recorded;
importing the overlooking reference plane image into image recognition software; counting and measuring the number of spring coils by using image recognition software, and determining a clearance value of one side of each coil of spring close to the surface of the spring insert; recording gap value data of each circle of spring;
and based on the coil wire diameter of the spring, carrying out error comparison on the gap value of each layer of spring coil: presetting a standard clearance value according to the wire diameter of the spring ring; identifying the gap value of each layer of spring ring through image identification software and carrying out repeated comparison with the standard gap value, namely taking the absolute value of the difference between the gap value of each layer of spring ring and the standard gap value as the error of each layer of spring ring, wherein the corresponding error that the gap value is equal to the standard gap value is 0; determining a second overlapping quality index according to the error of the clearance value of each layer of spring ring and the standard clearance value, for example, comparing the error of the clearance value of each layer of spring ring and the standard clearance value with a tolerance interval, namely, the tolerance range of the error: if errors of the clearance value of each layer of spring ring and the standard clearance value belong to the tolerance interval, setting the second overlapping quality index to be 1; if the error does not belong to the tolerance interval, calculating the sum of squares of errors of the gap values of the spring coils of each layer, which do not belong to the tolerance interval, and the standard gap values, and taking the calculated sum of squares as a second overlapping quality index. By using the technical means of image recognition, fluctuation error analysis and the like, the quality of the spring can be rapidly and accurately detected, and the comprehensive monitoring and control of the quality of the spring can be realized.
Step S50 further includes the steps of:
s541: when the coil wire diameter of the spring is smaller than the preset coil wire diameter of the spring, acquiring a coil wire diameter difference value of the spring;
s542: generating a first precision regulation and control instruction according to the coil diameter difference value of the spring;
s543: and carrying out pixel precision conversion and identification on the overlook reference plane image based on the first precision regulation and control instruction.
Specifically, after the coil wire diameter of the target assembly is obtained, the method further comprises that the coil wire diameter difference value is one of important indexes for measuring the quality of the spring, if the coil wire diameter difference value is too large, the quality of the spring is not up to the standard, and the preset coil wire diameter refers to the standard section diameter of a material arranged during designing or producing the spring;
when the coil wire diameter of the spring is smaller than the preset coil wire diameter of the spring, acquiring a coil wire diameter D of the spring 1 And the wire diameter D of the preset spring coil 2 Is the difference of D 1 -D 2 . The first precision regulation and control instruction is used for regulating production equipment, so that the produced spring meets the preset standard, and the production efficiency and the production quality can be improved;
the first precision regulation and control instruction can be used for regulating the focal length of the image acquisition device to enhance the pixel precision of the overlooking reference plane image, and specifically, the quantity to be regulated is calculated according to the coil diameter difference value; taking the coil diameter difference value and the quantity to be regulated as the instruction content of a first precision regulation instruction to generate the first precision regulation instruction;
the first precision regulating instruction is sent to the image acquisition device, and the focal length of the acquisition device is regulated according to the first precision regulating instruction, so that the pixel precision of the image is improved; performing pixel precision conversion and identification on the adjusted overlook reference plane image, and measuring the coil wire diameter of the spring with higher precision; comparing the measured coil wire diameter with higher precision with a preset coil wire diameter, and calculating a coil wire diameter difference value with higher precision, thereby providing a foundation for correcting the coil wire diameter of the spring;
the adjustment of pixel precision conversion and identification of the overlooking reference plane image is realized, the diameter information of the spring ring is accurately obtained through the pixel precision conversion and identification, and the pixel precision of the image is improved through adjusting the focal length of the acquisition device, so that the detection accuracy and reliability are improved.
S60: outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index;
as shown in fig. 3, step S60 further includes the steps of:
s61: performing spring simulation according to the assembly simulation model, and outputting a test data set;
s62: screening a sample data set according to the test data set, wherein the sample data set is a spring superposition quality sample with quality data larger than preset quality data in assembly quality detection reports of the same batch, and identification data for identifying the ratio of the first quality detection result to the second quality detection result;
s63: training according to the sample data set, outputting a spring quality recognition model, recognizing according to the spring quality recognition model, and generating the first quality detection result and the second quality detection result.
Specifically, according to the fitting simulation model, physical parameters of the spring are set, wherein the physical parameters of the spring include, but are not limited to, materials and elastic modulus of the spring; setting working conditions of the spring in simulation software, such as stress condition and working environment; performing simulation calculation of the spring and outputting a test data set;
the sample data set is a spring superposition quality sample with the quality data larger than the preset quality data in the assembly quality detection report of the same batch, and identification data for identifying the ratio of the first quality detection result to the second quality detection result; screening all spring superposition quality samples with quality data larger than preset quality data from assembly quality detection reports of the same batch on the basis of the test data set, and taking the spring superposition quality samples as a sample data set;
specifically, a simulation program is built in the Simulink, simscan is a single module library in the Simulink, the simscan encapsulates a common model into ready-made modules, based on which training is performed according to the sample data set, and a spring quality recognition model is output, including: based on the feedforward neural network as a model, the sample data set is adopted as construction data, new combination features are constructed based on the historical first coincidence quality index corresponding to the first coincidence quality index and the historical second coincidence quality index corresponding to the second coincidence quality index related to a fitting quality detection report in the sample data set, the combination features are overlapped, and the integrator is commonly used for expressing the first coincidence quality index corresponding to the first stateIs a derivative of the relation Z 1 Differential relation Z between second composite quality indexes corresponding to second states 2 Then, the Simulink is connected through a simple block diagram, and Z can be spelled out by using an addition, subtraction, multiplication and division module 1 And Z 2 According to a similar iterative process, and then assembling to obtain a spring quality identification model;
taking a historical first quality detection result corresponding to a first quality detection result related to a fitting quality detection report in the sample data set and a historical second quality detection result corresponding to a second quality detection result as identification results, transmitting the identification results into a feedforward neural network for model convergence learning, constructing and training to obtain the spring quality recognition model, and determining the spring quality recognition model to provide a model foundation for spring quality recognition;
inputting the spring quality recognition model to recognize the spring quality by taking the first coincidence quality index and the second coincidence quality index as input data, and outputting a first quality detection result based on the first state and a second quality detection result based on the second state; providing a model foundation for spring quality identification.
The embodiment of the application further comprises the steps of:
s64: and identifying the effective spring number of the springs of the target assembly, outputting an effective spring number M, wherein,；
s64: updating the first overlapping quality index and the second overlapping quality index according to the effective spring coil number M, and outputting the updated first overlapping quality index and second overlapping quality index;
s643: and respectively outputting the first quality detection result and the second quality detection result according to the updated first coincidence quality index and the updated second coincidence quality index.
In particular, fatigue failure may occur inevitably after a period of use of the springs of the target assembly. The spring is used for a long timeIn the process, due to repeated stress loading and unloading, tiny defects in materials can be gradually expanded, and finally the breaking or the elastic deformation capacity of the spring is lost, and based on the breaking or the elastic deformation capacity of the spring, the spring of the target assembly is measured and analyzed to obtain spring parameters, wherein the spring parameters comprise but are not limited to spring wire diameter, spring inner diameter, spring outer diameter and free length; calculating the spring rate and the effective spring number M of the spring of the target assembly according to the spring parameters, wherein the effective spring number M refers to the spring number which can generate elastic deformation in the spring,；
and according to the effective spring coil number M, carrying out spring coil distance recognition and spring interval recognition to obtain an updated first coincidence quality index and an updated second coincidence quality index, and updating the first coincidence quality index and the second coincidence quality index at the same time. The updated first superposition quality index reflects the current actual spacing error degree of two adjacent circles of springs of the target assembly; the updated first superposition quality index reflects the stability degree of the current actual spacing between the inner ring of the spring of the target assembly and the surface of the spring embedded part; and respectively outputting the first quality detection result based on the first state and the second quality detection result based on the second state according to the updated first coincidence quality index and the updated second coincidence quality index, synchronously updating the first quality detection result and the second quality detection result in the long-term use process of the spring, evaluating the stability degree of the surface distance between the inner ring of the spring and the spring embedded part, and simultaneously, periodically updating the first quality detection result and the second quality detection result to replace the fatigue failure spring so as to provide a reference for ensuring the normal operation and the safety of the target assembly.
S70: and generating a quality detection report based on the first quality detection result and the second quality detection result.
Step S70 includes the steps of:
s71: collecting a spring operation mode of the target assembly, controlling the simulation system to simulate the assembly simulation model when the spring of the target assembly comprises a tensile state, and outputting a third state when the spring is tensile;
s72: outputting a third quality detection result according to the third state;
s73: and generating a quality detection report based on the first quality detection result, the second quality detection result and the third quality detection result.
Specifically, a quality detection report is generated based on the first quality detection result and the second quality detection result, the spring operation mode of the target assembly part further comprises a third state, the third state is a state when the spring operation process is stretched, and when the spring of the target assembly part comprises a stretching state, stretching test data are collected; simulating the assembly simulation model by using a control simulation system, inputting acquired tensile test data as simulation parameters when the spring stretches, and outputting a third state when the spring stretches; a step of acquiring a first quality detection result based on the spring quality recognition model with reference to the first state and a second quality detection result based on the second state, and outputting a third quality detection result according to the third state; and summarizing and sorting the first quality detection result, the second quality detection result and the third quality detection result, wherein the quality detection report is obtained after summarizing and sorting, and the quality detection report can help a spring manufacturer to know and improve the spring manufacturing process and improve the quality control level of the spring.
In summary, the method and the system for detecting the automatic assembly quality of the spring provided by the embodiment of the application have the following technical effects:
1. the connection simulation system is adopted to model the target assembly, so that an assembly simulation model is output; acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model; respectively carrying out multi-angle image acquisition on the target assembly according to the first state and the second state, and outputting a first image set and a second image set, wherein each image set comprises a side-looking reference surface image and a top-looking reference surface image; the method comprises the steps of obtaining a first coincidence quality index by recognizing a spring circle distance of a side-looking reference plane image, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs; acquiring a second overlapping quality index by identifying the spring interval of the overlooking reference surface image, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the spring inner ring and the spring embedded part; outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index; based on the first quality detection result and the second quality detection result, the application provides the automatic spring assembly quality detection method and system, so that the technical effects of rapidly and accurately evaluating the automatic spring assembly quality and guaranteeing the consistency and stability of products are realized on a mass-production automatic spring assembly production line by identifying the spring spacing of overlooking reference surface images and evaluating the stability of the spacing between the inner ring of the spring and the surface of the spring insert.
2. Due to the fact that when the coil wire diameter of the spring is smaller than the preset coil wire diameter of the spring, the coil wire diameter difference value of the spring is obtained; generating a first precision regulation and control instruction according to the coil wire diameter difference value; and carrying out pixel precision conversion and identification on the overlook reference surface image based on the first precision regulation and control instruction. The adjustment of pixel precision conversion and identification of the overlooking reference plane image is realized, the diameter information of the spring ring is accurately obtained through the pixel precision conversion and identification, and the pixel precision of the image is improved through adjusting the focal length of the acquisition device, so that the detection accuracy and reliability are improved.
Example two
Based on the same inventive concept as the spring automatic assembly quality detection method in the foregoing embodiments, as shown in fig. 4, an embodiment of the present application provides a spring automatic assembly quality detection system, where the system includes:
the simulation modeling module 100 is used for being connected with a simulation system to model a target assembly and outputting an assembly simulation model;
a state acquisition module 200 for acquiring a first state when the spring is relaxed and a second state when the spring is compressed in the fitting simulation model;
the image acquisition module 300 is configured to acquire multiple angles of images of a target assembly according to the first state and the second state, and output a first image set and a second image set, where each image set includes a side view reference plane image and a top view reference plane image;
the spring circle distance recognition module 400 is configured to obtain a first coincidence quality index by performing spring circle distance recognition on the side-view reference plane image, where the first coincidence quality index is used to identify a degree of error of a distance between two adjacent circles of springs;
the spring interval recognition module 500 is configured to obtain a second quality index of overlap by performing spring interval recognition on the top-view reference plane image, where the second quality index of overlap is used to identify a stability degree of a surface interval between the inner ring of the spring and the spring insert;
a quality detection result output module 600, configured to output a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index;
a quality detection report generating module 700, configured to generate a quality detection report based on the first quality detection result and the second quality detection result.
Further, the quality detection report generating module 700 is configured to perform the following steps:
collecting a spring operation mode of the target assembly, controlling the simulation system to simulate the assembly simulation model when the spring of the target assembly comprises a tensile state, and outputting a third state when the spring is tensile;
outputting a third quality detection result according to the third state;
and generating a quality detection report based on the first quality detection result, the second quality detection result and the third quality detection result.
Further, the spring coil pitch recognition module 400 is configured to perform the following steps:
acquiring the number of spring turns N which is a positive integer greater than or equal to 0 by identifying the side view reference plane image;
identifying the coil distance between two adjacent coils in the spring coil number N, and outputting N-1 coil distance;
and carrying out fluctuation error analysis according to the N-1 group of circle distances, and outputting the first coincidence quality index.
Further, the spring interval identification module 500 is configured to perform the following steps:
acquiring a coil wire diameter of the target assembly;
identifying the overlooking reference plane image, and determining a clearance value of one side of each circle of spring close to the surface of the spring insert;
and based on the coil wire diameters, performing error comparison on the gap values of each layer of coils, and outputting the second overlapping quality index.
Further, the spring coil pitch recognition module 400 is further configured to perform the following steps:
and identifying the effective spring number of the springs of the target assembly, outputting an effective spring number M, wherein,；
updating the first overlapping quality index and the second overlapping quality index according to the effective spring coil number M, and outputting the updated first overlapping quality index and second overlapping quality index;
and respectively outputting the first quality detection result and the second quality detection result according to the updated first coincidence quality index and the updated second coincidence quality index.
Further, the quality detection result output module 600 is configured to perform the following steps:
performing spring simulation according to the assembly simulation model, and outputting a test data set;
screening a sample data set according to the test data set, wherein the sample data set is a spring superposition quality sample with quality data larger than preset quality data in assembly quality detection reports of the same batch, and identification data for identifying the ratio of the first quality detection result to the second quality detection result;
training according to the sample data set, outputting a spring quality recognition model, recognizing according to the spring quality recognition model, and generating the first quality detection result and the second quality detection result.
Further, the spring interval identification module 500 is further configured to perform the following steps:
when the coil wire diameter of the spring is smaller than the preset coil wire diameter of the spring, acquiring a coil wire diameter difference value of the spring;
generating a first precision regulation and control instruction according to the coil diameter difference value of the spring;
and carrying out pixel precision conversion and identification on the overlook reference plane image based on the first precision regulation and control instruction.
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 method for implementing an embodiment of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.

Claims (5)

1. A method for automatically assembling a spring for quality detection, the method comprising:
modeling the target assembly by the connection simulation system, and outputting an assembly simulation model;
acquiring a first state when a spring is relaxed and a second state when the spring is compressed in the fitting simulation model;
respectively acquiring multi-angle images of a target assembly according to the first state and the second state, and outputting a first image set and a second image set, wherein each image set comprises a side-looking reference surface image and a top-looking reference surface image;
acquiring a first coincidence quality index by recognizing the spring circle distance of the side-looking reference plane image, wherein the first coincidence quality index is used for marking the distance error degree of two adjacent circles of springs;
acquiring a second overlapping quality index by carrying out spring interval identification on the overlook reference surface image, wherein the second overlapping quality index is used for marking the stability degree of the surface interval between the spring inner ring and the spring embedded part;
outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index;
generating a quality detection report based on the first quality detection result and the second quality detection result;
the spring circle distance identification is carried out on the side-view reference surface image, and a first coincidence quality index is obtained, and the method comprises the following steps:
acquiring the number of spring turns N which is a positive integer greater than or equal to 0 by identifying the side view reference plane image;
identifying the coil distance between two adjacent coils in the spring coil number N, and outputting N-1 coil distance;
carrying out fluctuation error analysis according to the N-1 group of circle distances and outputting the first coincidence quality index;
the spring interval identification is carried out on the overlook reference surface image, so that a second overlapping quality index is obtained, and the method comprises the following steps:
acquiring a coil wire diameter of the target assembly, and generating a preset clearance value according to the coil wire diameter of the target assembly;
identifying the overlooking reference plane image, and determining a clearance value of one side of each circle of spring close to the surface of the spring insert;
based on the preset gap value, performing error comparison on the gap value of each layer of spring ring, and outputting the second overlapping quality index;
the method further comprises the steps of:
and identifying the effective spring number of the springs of the target assembly, outputting an effective spring number M, wherein,；
updating the first overlapping quality index and the second overlapping quality index according to the effective spring coil number M, and outputting the updated first overlapping quality index and second overlapping quality index;
and respectively outputting the first quality detection result and the second quality detection result according to the updated first coincidence quality index and the updated second coincidence quality index.
2. The method of claim 1, wherein the method further comprises:
collecting a spring operation mode of the target assembly, controlling the simulation system to simulate the assembly simulation model when the spring of the target assembly comprises a tensile state, and outputting a third state when the spring is tensile;
outputting a third quality detection result according to the third state;
and generating a quality detection report based on the first quality detection result, the second quality detection result and the third quality detection result.
3. The method of claim 1, wherein the method further comprises:
performing spring simulation according to the assembly simulation model, and outputting a test data set;
screening a sample data set according to the test data set, wherein the sample data set is a spring superposition quality sample with quality data larger than preset quality data in assembly quality detection reports of the same batch, and identification data for identifying the ratio of the first quality detection result to the second quality detection result;
training according to the sample data set, outputting a spring quality recognition model, recognizing according to the spring quality recognition model, and generating the first quality detection result and the second quality detection result.
4. The method of claim 1, wherein after obtaining the coil wire diameter of the target assembly, the method further comprises:
when the coil wire diameter of the spring is smaller than the preset coil wire diameter of the spring, acquiring a coil wire diameter difference value of the spring;
generating a first precision regulation and control instruction according to the coil diameter difference value of the spring;
and carrying out pixel precision conversion and identification on the overlook reference plane image based on the first precision regulation and control instruction.
5. A spring automatic assembly quality inspection system for implementing a spring automatic assembly quality inspection method according to any one of claims 1-4, comprising:
the simulation modeling module is used for being connected with the simulation system to model the target assembly and outputting an assembly simulation model;
the state acquisition module is used for acquiring a first state when the spring is relaxed and a second state when the spring is compressed in the assembly simulation model;
the image acquisition module is used for respectively acquiring the multi-angle images of the target assembly according to the first state and the second state and outputting a first image set and a second image set, wherein each image set comprises a side-view reference surface image and a top-view reference surface image;
the spring circle distance identification module is used for obtaining a first coincidence quality index by carrying out spring circle distance identification on the side-view reference surface image, wherein the first coincidence quality index is used for identifying the distance error degree of two adjacent circles of springs;
the spring interval identification module is used for acquiring a second overlapping quality index by carrying out spring interval identification on the overlook reference surface image, wherein the second overlapping quality index is used for identifying the stability degree of the surface interval between the spring inner ring and the spring embedded part;
the quality detection result output module is used for outputting a first quality detection result based on the first state and a second quality detection result based on the second state according to the first coincidence quality index and the second coincidence quality index;
and the quality detection report generation module is used for generating a quality detection report based on the first quality detection result and the second quality detection result.
CN202310939134.9A 2023-07-28 2023-07-28 Spring automatic assembly quality detection method and system Active CN116681698B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310939134.9A CN116681698B (en) 2023-07-28 2023-07-28 Spring automatic assembly quality detection method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310939134.9A CN116681698B (en) 2023-07-28 2023-07-28 Spring automatic assembly quality detection method and system

Publications (2)

Publication Number Publication Date
CN116681698A CN116681698A (en) 2023-09-01
CN116681698B true CN116681698B (en) 2023-10-10

Family

ID=87787620

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310939134.9A Active CN116681698B (en) 2023-07-28 2023-07-28 Spring automatic assembly quality detection method and system

Country Status (1)

Country Link
CN (1) CN116681698B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113310518A (en) * 2021-05-27 2021-08-27 北京交通大学 Air spring surface quality visual inspection system
CN114218692A (en) * 2021-11-23 2022-03-22 江苏科技大学 Similar part identification system, medium and method based on deep learning and model simulation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113310518A (en) * 2021-05-27 2021-08-27 北京交通大学 Air spring surface quality visual inspection system
CN114218692A (en) * 2021-11-23 2022-03-22 江苏科技大学 Similar part identification system, medium and method based on deep learning and model simulation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于机器视觉的弹簧装配检测;李勇;《计量与测试技术》;第18-20页 *

Also Published As

Publication number Publication date
CN116681698A (en) 2023-09-01

Similar Documents

Publication Publication Date Title
CN107492098B (en) It is a kind of based on PCA and CNN high-temperature forging surface defect in position detecting method
CN111402226A (en) Surface defect detection method based on cascade convolution neural network
CN111784688A (en) Flower automatic grading method based on deep learning
US11521120B2 (en) Inspection apparatus and machine learning method
CN115373370A (en) Method and system for monitoring running state of programmable controller
CN113902710A (en) Method and system for detecting surface defects of industrial parts based on anomaly detection algorithm
US11682112B2 (en) Inspection device and machine learning method
CN116465315A (en) Automatic screen quality detection method and system
CN116681698B (en) Spring automatic assembly quality detection method and system
CN117152152B (en) Production management system and method for detection kit
CN110320802B (en) Complex system signal time sequence identification method based on data visualization
Anchini et al. A comparison between stereo-vision techniques for the reconstruction of 3-D coordinates of objects
CN117052970A (en) Intelligent control system and method for pneumatic ball valve assembly
Ekambaram et al. Identification of defects in casting products by using a convolutional neural network
CN116165213A (en) Stainless steel chain product quality inspection method and system
CN116524004B (en) Method and system for detecting size of steel bar based on HoughLines algorithm
CN117648596B (en) Digital twin and intelligent sensor fusion method and system for building construction
CN117171922B (en) Method and system for parallel correction in steel structure manufacturing
CN117074419B (en) Defect detection method and system for engine gear ring section steel
CN116454583B (en) Production control method of base station filter shell
CN115792141B (en) Method and system for improving balance detection efficiency of water activity meter
CN117074181B (en) Pressure testing method, device and equipment for curved flexible screen and storage medium
CN117193047A (en) Industrial production simulation management system and method based on machine vision
CN117036804A (en) Detection method and system for bookshelf production
CN117217020A (en) Industrial model construction method and system based on digital twin

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