CN112100753A - Big-data prediction system and method for key geometric parameters of self-piercing riveting joint - Google Patents
Big-data prediction system and method for key geometric parameters of self-piercing riveting joint Download PDFInfo
- Publication number
- CN112100753A CN112100753A CN202010769060.5A CN202010769060A CN112100753A CN 112100753 A CN112100753 A CN 112100753A CN 202010769060 A CN202010769060 A CN 202010769060A CN 112100753 A CN112100753 A CN 112100753A
- Authority
- CN
- China
- Prior art keywords
- value
- joint
- self
- network model
- height
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 241000587161 Gomphocarpus Species 0.000 claims abstract description 68
- 238000004364 calculation method Methods 0.000 claims abstract description 38
- 230000008569 process Effects 0.000 claims abstract description 37
- 238000012797 qualification Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 74
- 238000011156 evaluation Methods 0.000 claims description 35
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000013500 data storage Methods 0.000 claims description 6
- 229910045601 alloy Inorganic materials 0.000 claims description 5
- 239000000956 alloy Substances 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 5
- 229910000838 Al alloy Inorganic materials 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000003698 laser cutting Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computational Mathematics (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
A big data prediction system for key geometric parameters of self-piercing riveting joints comprises: the data call module, data input module and data calculation module, the data call module: calling a pre-trained network model corresponding to the type of the mold according to the type of the mold of the joint to be tested; a data input module: inputting the technological parameters of the joint to be tested into the called pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested; a data calculation module: calculating to obtain key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested; according to the invention, a pre-trained network model is selected according to the type of the die to calculate the input process parameters to obtain the nail head height and the bottom thickness value of the joint, and then the qualification judgment is carried out on the result of the predicted value according to the set judgment standard.
Description
Technical Field
The invention relates to prediction of key geometric parameters of a self-piercing riveting joint, in particular to a big-data prediction system and method of the key geometric parameters of the self-piercing riveting joint.
Background
The self-piercing riveting technology is a novel connecting technology for thin plate materials, a semi-hollow rivet is pressed into a plate through a punch in the plate connecting process, the lower layer of the plate is embedded into the rivet leg under the action of the punch and a bottom die after the upper plate is pierced, so that a self-locking effect is formed, the connection of two layers of plates or more than two layers of plates is realized, the self-piercing riveting technology has the characteristics of simple process, no pollution, high connecting strength and the like, and is applied to more and more automobile enterprises.
Due to the characteristics of the self-piercing riveting process, the quality of the joint is mainly influenced by various aspects such as parameters of connecting plates, parameters of rivets, parameters of dies, parameters of driving equipment and the like, a process window is narrow, and the problems of unqualified joint quality such as rivet falling, plate piercing, improper riveting and the like can occur once the parameters are selected improperly. Therefore, before the self-piercing riveting process is actually applied, process exploration is carried out on application materials and structures, and joint quality under different processes is checked. At present, the meridian plane of the sectioning joint is often adopted to measure key geometric parameter values to judge the forming quality of the joint. The three detected evaluation indexes are respectively a self-locking value, a nail head height and a bottom thickness value, wherein the nail head height and the bottom thickness value are visual standards for judging whether the joint is qualified, and the bonding strength of the joint is directly determined by the self-locking value and serves as a main quality evaluation index. The conventional measuring method needs to adopt a laser cutting mode to destroy the self-piercing riveting joint, has high cost and low efficiency, and is not suitable for wide popularization and application.
At present, a series of researches are carried out at home and abroad on how to quickly obtain key geometric parameters of a cutting plane of a self-piercing rivet joint, the connection quality of the self-piercing rivet joint is detected by utilizing narrow-band ultrasound, the riveting quality is evaluated by monitoring the impedance change of a sensor, but the research result cannot judge the internal deformation characteristic of the self-piercing rivet joint, only whether the self-piercing rivet joint has defects can be distinguished, the effect of detecting a thinner joint is extremely poor, the detection process extremely depends on operation skills, and the detection error is larger.
Disclosure of Invention
Aiming at the problems that the internal deformation characteristics of a self-piercing riveting joint cannot be judged in the prior art, whether the self-piercing riveting joint has defects can only be distinguished, the effect of detecting a thinner joint is extremely poor, the detection process extremely depends on the operating skill, and the detection error is large, the invention provides a big-data prediction system for the key geometric parameters of the self-piercing riveting joint, which comprises the following steps: the data input module is used for inputting data;
the data calling module: determining a pre-trained network model corresponding to the type of the mould according to the type of the mould of the joint to be tested;
the data input module: inputting the technological parameters of the joint to be tested into the pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
the data calculation module: calculating to obtain key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested;
the key geometric parameters comprise the height of a nail head of the joint and a bottom thickness value;
the pre-trained network model is obtained by training joint historical process parameters by adopting an error back propagation algorithm.
Preferably, the method further comprises a model training module: the network model is used for dividing historical process parameter data into training samples corresponding to the types of the molds according to the types of the molds, and training the corresponding network models by adopting an error back propagation algorithm based on the training samples corresponding to the types of the molds.
Preferably, the model training module includes: a training sample classification submodule, a training sample submodule, a training network model submodule and a data storage submodule;
the training sample classification submodule: the device is used for dividing historical process parameter data into training samples corresponding to each mould type according to the mould type;
the training sample submodule: the method comprises the steps of forming a training sample by upper plate thickness, lower plate thickness, upper plate hardness, lower plate hardness, upper plate strength, lower plate strength, rivet hardness, rivet length, opening value of a nail foot and a nail foot self-locking value;
the training network model submodule comprises: taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in the training sample corresponding to the mold type as the input of the network model, and taking the opening value and the pin self-locking value of the pins as the output of the network model for training to obtain a trained network model corresponding to the mold type;
the data storage submodule: and storing the trained network model corresponding to the mold type and calling the trained network model by the data calling module.
Preferably, the process parameters include: alloy grade, plate thickness, rivet specification, bottom die depth, riveting speed and nail foot opening value and self-locking value under corresponding process conditions.
Preferably, the data calculation module includes: the nail head height calculation submodule and the bottom thickness value calculation operator module;
the nail head height calculating submodule comprises: calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
the bottom thickness value calculating operator module: and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
Preferably, the method further comprises the following step of: and the method is used for judging the qualification of the key geometric parameters of the joint to be tested according to the set joint evaluation standard.
Preferably, the result prediction module includes: a judgment submodule and an evaluation submodule;
the judgment submodule is used for judging the relationship between the key geometric parameters of the joint to be detected and a threshold value;
and the evaluation submodule is used for determining whether the to-be-detected connector is qualified or not based on the judgment result of the judgment submodule.
Preferably, the judgment sub-module includes: the self-locking value judging unit, the nail head height judging unit and the bottom thickness value judging unit;
the self-locking value judging unit: judging the size of the self-locking value and a set self-locking threshold value;
the nail head height judging unit: judging whether the height of the nail head is within the range of the set nail head height threshold value;
the bottom thickness value judgment unit: and judging the bottom thickness value and the set bottom thickness threshold value.
Preferably, the evaluation sub-module includes: the device comprises a self-locking value evaluation unit, a nail head height evaluation unit and a bottom thickness value evaluation unit;
the self-locking value evaluation unit: when the self-locking value is greater than or equal to the self-locking threshold value, the product is qualified, otherwise, the product is unqualified;
the nail head height evaluation unit: when the height of the nail head is within the range of the set height threshold value of the nail head, the nail head is qualified, otherwise, the nail head is unqualified;
the bottom thickness value evaluation unit: and when the bottom thickness value is larger than or equal to the bottom thickness threshold value, the product is qualified, otherwise, the product is unqualified.
Preferably, the result prediction module further includes a display sub-module, which is used for displaying the key geometric parameters of the to-be-tested connector.
Preferably, the formula for calculating the height of the nail head is as follows:
h＝(Ddie-D0)·tanα/2-LCap (hat)，α＝arcsin(D-D0)/2LFoot
Wherein h is the height of the head of the joint and DDieIs the inner diameter of the die; d0The original diameter of the rivet; l isCap (hat)Is the height of the nail head(ii) a D is the nail foot opening value.
Preferably, the base thickness calculation formula is as follows:
s＝t1+t2+Hdie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFoot
Wherein s is a base thickness value; t is t1Is the thickness of the upper plate; t is t2Is the thickness of the lower plate; hDieIs the mold height; l isFootThe fixed focus height is obtained.
Based on the same inventive concept, the invention provides a big-data prediction method of key geometric parameters of a self-piercing riveting joint, which comprises the following steps:
calling a pre-trained network model corresponding to the type of the mold according to the type of the mold of the joint to be tested;
inputting the technological parameters of the joint to be tested into the called pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
and calculating to obtain the key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested.
Preferably, the network model training includes:
classifying historical data of the training sample according to a plurality of mould types;
the training sample comprises an upper plate thickness, a lower plate thickness, an upper plate hardness, a lower plate hardness, an upper plate strength, a lower plate strength, a rivet hardness, a rivet length, a nail foot opening value and a nail foot self-locking value to form a training sample;
taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in the training sample as the input of the network model, and taking the opening value and the pin self-locking value of the pins as the output of the network model for training to obtain a trained network model;
and storing the trained network model and providing the trained network model for the data calling module to call.
Preferably, the calculating to obtain the key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested includes:
calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
Compared with the prior art, the invention has the beneficial effects that:
1. a big data prediction system for key geometric parameters of self-piercing riveting joints comprises: the data call module, data input module and data calculation module, the data call module: calling a pre-trained network model corresponding to the type of the mold according to the type of the mold of the joint to be tested; a data input module: inputting the technological parameters of the joint to be tested into the called pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested; a data calculation module: calculating to obtain key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested; the invention selects the pre-trained network model to calculate the input process parameters through the type of the mould to obtain the pin head height and the bottom thickness value of the joint, thereby effectively solving the problem of predicting the key geometric parameters of the joint to be measured.
2. The network model is trained through the training sample, then the process parameters of the joint to be tested are input, all the geometric parameters of the joint to be tested can be obtained, and the low-cost and high-efficiency test is effectively achieved through all the geometric parameters and the judgment standard.
Drawings
FIG. 1 is a flow chart of a big data prediction system for key geometric parameters of a self-piercing rivet joint according to the present invention;
FIG. 2 is a schematic diagram of a big data module of a system for predicting key geometric parameters of a self-piercing rivet joint according to the present invention;
FIG. 3 is a schematic diagram of a big data BP neural network model according to the present invention;
FIG. 4 is a schematic diagram illustrating the operation relationship of geometric characteristic values of self-piercing riveting cross sections according to the present invention;
FIG. 5 is a schematic diagram of the geometry of four standard bottom molds according to the present invention;
FIG. 6 is a schematic diagram of an embedded standard eligibility assessment interface in accordance with the present invention.
Detailed Description
The invention will be further explained with reference to the accompanying drawings
Example 1
With reference to fig. 1 and 2, the present invention provides a big data prediction system for key geometric parameters of self-piercing rivet joints, including: the data input module is used for inputting data;
a data calling module: determining a pre-trained network model corresponding to the type of the mould according to the type of the mould of the joint to be tested;
a data input module: inputting the technological parameters of the joint to be tested into a pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
a data calculation module: calculating to obtain key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested;
the key geometric parameters comprise the height of a nail head of the joint and the bottom thickness value;
the pre-trained network model is obtained by training joint historical process parameters by adopting an error back propagation algorithm.
The model training module is further included: the method is used for dividing historical process parameter data into training samples corresponding to the types of the molds according to the types of the molds, and training the corresponding network models by adopting an error back propagation algorithm based on the training samples corresponding to the types of the molds.
The model training module comprises: a training sample classification submodule, a training sample submodule, a training network model submodule and a data storage submodule;
training a sample classification submodule: the device is used for dividing historical process parameter data into training samples corresponding to each mould type according to the mould type;
training a sample submodule: the method comprises the steps of forming a training sample by upper plate thickness, lower plate thickness, upper plate hardness, lower plate hardness, upper plate strength, lower plate strength, rivet hardness, rivet length, opening value of a nail foot and a nail foot self-locking value;
training a network model submodule: taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in a training sample corresponding to the mold type as the input of a network model, and taking the opening value and the pin self-locking value of the pins as the output of the network model for training to obtain a trained network model corresponding to the mold type;
a data storage submodule: and storing the trained network model corresponding to the mold type and calling the trained network model by the data calling module.
The technological parameters comprise: alloy grade, plate thickness, rivet specification, bottom die depth, riveting speed and nail foot opening value and self-locking value under corresponding process conditions.
A data computation module comprising: the nail head height calculation submodule and the bottom thickness value calculation operator module;
nail head height calculating submodule: calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
bottom thickness value calculating operator module: and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
The device also comprises an outcome prediction module: and the method is used for judging the qualification of the key geometric parameters of the joint to be tested according to the set joint evaluation standard.
An outcome prediction module comprising: a judgment submodule and an evaluation submodule;
the judgment submodule is used for judging the relationship between the key geometric parameters of the joint to be detected and the threshold value;
and the evaluation submodule is used for determining whether the joint to be detected is qualified or not based on the judgment result of the judgment submodule.
A judgment submodule comprising: the self-locking value judging unit, the nail head height judging unit and the bottom thickness value judging unit;
a self-locking value judgment unit: judging the size of the self-locking value and a set self-locking threshold value;
nail head height judging unit: judging whether the height of the nail head is within the range of the set nail head height threshold value;
bottom thickness value determination unit: and judging the bottom thickness value and the set bottom thickness threshold value.
An evaluation submodule, comprising: the device comprises a self-locking value evaluation unit, a nail head height evaluation unit and a bottom thickness value evaluation unit;
a self-locking value evaluation unit: when the self-locking value is greater than or equal to the self-locking threshold value, the product is qualified, otherwise, the product is unqualified;
nail head height evaluation unit: when the height of the nail head is within the range of the set height threshold value of the nail head, the nail head is qualified, otherwise, the nail head is unqualified;
bottom thickness value evaluation unit: and when the bottom thickness value is larger than or equal to the bottom thickness threshold value, the product is qualified, otherwise, the product is unqualified.
The result prediction module also comprises a display submodule used for displaying the key geometric parameters of the joint to be tested.
The formula for calculating the height of the nail head is shown as follows:
h＝(Ddie-D0)·tanα/2-LCap (hat)，α＝arcsin(D-D0)/2LFoot
Wherein h is the height of the head of the joint and DDieIs the inner diameter of the die; d0The original diameter of the rivet; l isCap (hat)Is the height of the nail cap; d is the nail foot opening value.
The equation for the bottom thickness is given by:
s＝t1+t2+Hdie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFoot
Wherein s is a base thickness value; t is t1Is the thickness of the upper plate; t is t2Is the thickness of the lower plate; hDieIs the mold height; l isFootThe fixed focus height is obtained.
Example 2
With reference to fig. 2, a big data-based prediction system and method for key geometric parameters of a self-piercing riveting joint include a material database, a training sample database, a data calling module, a data input module, a data calculation module, and a result display module.
The material database comprises material alloy grades, thickness, hardness, yield strength and tensile strength;
the training sample database is obtained through actual process tests and simulation model calculation. Different samples were formed for different die types. Determining a training sample database under four bottom die parameters, namely a phi 3.0-3.4 mm rivet flat die, a phi 3.0-3.4 mm rivet male die, a phi 5.0-5.4 mm rivet flat die and a phi 5.0-5.4 mm rivet male die respectively; the training sample database comprises alloy grades, plate thickness, rivet specification, bottom die depth, riveting speed and nail foot opening value and self-locking value under corresponding process conditions.
The data calling module calls parameters of the bottom die to call different samples;
the data input module is used for inputting process parameters which have great influence on the performance of the joint and mainly comprises an upper plate thickness t1Hardness h of upper plate1Upper plate strength σ1(ii) a Lower plate thickness t2Hardness of lower plate h2Lower plate strength σ2(ii) a Rivet length L0Rivet hardness H0(ii) a The riveting speed V.
With reference to fig. 3 and 4, the data prediction module calculates the input calculation parameters according to a Back Propagation (BP) algorithm established by the training samples, and establishes a network model of the input parameters, the nail foot opening value and the self-locking value.
And the data calculation module obtains the nail head height h and the bottom thickness value s of the joint through calculation according to the upper and lower plate thicknesses, the geometric dimension of the rivet and the depth parameter of the bottom die by utilizing the nail foot opening value D and the self-locking value u predicted by the big data model. The calculation is performed according to the following calculation formula:
h＝(Ddie-D0)·tanα/2-LCap (hat)，s＝t1+t2+HDie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFoot
And the result display module is used for displaying the prediction and calculation results and judging the qualification of the key geometric parameters of the joint to be predicted according to the following joint evaluation standard.
Self-locking value u: not less than 0.2-percent-qualified; less than 0.2-unqualified;
height h of the nail head: h is more than-0.1 and less than 0.1, and qualified; not less than 0.1 or not more than-0.1-disqualified;
bottom thickness value s: not less than 0.15-qualified; less than 0.15, failing.
The mechanical property of the joint can be ensured when the self-locking value u is more than or equal to 0.2 mm; the height h of the nail head is controlled within +/-0.1 mm, so that the good appearance size of the joint can be ensured; the bottom thickness value s is more than or equal to 0.15, so that the sealing performance and the fatigue life of the joint can be ensured.
A big data-based prediction system and method for a geometric characteristic value of a self-piercing riveting joint section comprises the following steps:
s1, calling a sample database by using a data calling module according to the type of the joint bottom die to be predicted;
s2, inputting the process parameters of the joint to be predicted by using the data input module, wherein the process parameters mainly comprise t1、h1、σ1、t2、h2、σ2、D0、V；
S3, analyzing and calculating each item of input self-piercing riveting head data by using a data calculation module to obtain a self-piercing riveting head nail foot opening value D and a self-locking value u;
s4, obtaining values of the nail head height h and the bottom thickness value S by using a calculation formula;
and S5, displaying the prediction and calculation results by using the result display module, and finishing the judgment of key geometric parameters and process qualification according to the joint evaluation standard.
Example 3
The thickness of the 5182-O state aluminum alloy plate is 1.5mm, the rivet specification is phi 5.3 multiplied by 5.5mm, the size of the bottom die is combined with the figure 5, and the depth h is 2 mm. The riveting process parameters of the 5182-O-state aluminum alloy plate are as follows: t is t1＝1.5mm，h1＝80HV，σ1＝280MPa；t2＝1.5mm，h2＝80HV，σ2＝280MPa；D0＝5.3mm，V＝60cm/min。
(1) And calling the phi 5.3 male die by using the data calling module.
(2) Will t1＝1.5mm，h1＝80HV，σ1＝280MPa；t2＝1.5mm，h2＝80HV，σ2＝280MPa；D0The parameter V is 60cm/min and is input into the data input module when the parameter V is 5.3 mm;
(3) calculating the opening value D of the joint nail foot to be 7.32mm and the self-locking value u to be 0.43mm according to a training sample called by a phi 5.3 male die;
(4) using h ═ D (D)Die-D0)·tanα/2-LCap (hat)，s＝t1+t2+HDie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFootThe formula calculates that the nail head height h is 0.05mm, and the bottom thickness s is 0.33 mm.
(5) According to the joint evaluation standard, with reference to fig. 6, the following process parameter conditions are judged: t is t1＝1.5mm，h1＝80HV，σ1＝280MPa；t2＝1.5mm，h2＝80HV，σ2＝280MPa；D0And (5.3) mm, V is 60cm/min, and the key geometric parameters of the joint meet the standard, and the judgment is qualified.
Example 4
The upper plate and the lower plate of the 5754-H22 aluminum alloy plate are both 1.5mm in thickness, the rivet specification is phi 5.3 multiplied by 6mm, the size of a bottom die is combined with the figure 5, and the depth H is 2 mm. The riveting process parameters of the 5754-H22 state aluminum alloy plate are as follows: t is t1＝1.5mm，h1＝75HV，σ1＝260MPa；t2＝1.5mm，h2＝75HV，σ2＝260MPa；D0＝5.3mm，V＝60cm/min。
(1) And calling the phi 5.3 male die by using the data calling module.
(2) Will t1＝1.5mm，h1＝75HV，σ1＝260MPa；t2＝1.5mm，h2＝75HV，σ2＝260MPa；D0The parameter V is 60cm/min and is input into the data input module when the parameter V is 5.3 mm;
(3) according to a training sample called by a phi 5.3 male die, calculating a nail foot opening value D of 7.75 mm; the self-locking value u is 0.65 mm.
(4) By using
h＝(DDie-D0)·tanα/2-LCap (hat)，s＝t1+t2+HDie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFoot
The formula calculates that h is 0.03mm, and s is 0.35 mm.
(5) According to the joint evaluation criteria, with reference to fig. 6, the following process conditions were judged: t is t1＝1.5mm，h1＝75HV，σ1＝260MPa；t2＝1.5mm，h2＝75HV，σ2＝260MPa；D0And (5.3) mm, V is 60cm/min, and the key geometric parameters of the joint meet the standard, and the judgment is qualified.
Example 5
Based on the same inventive concept, the invention provides a big-data prediction method of key geometric parameters of a self-piercing riveting joint, which comprises the following steps:
calling a pre-trained network model corresponding to the type of the mold according to the type of the mold of the joint to be tested;
inputting the technological parameters of the joint to be tested into the called pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
and calculating to obtain the key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested.
The network model training comprises the following steps:
classifying historical data of the training sample according to a plurality of mould types;
the training sample comprises an upper plate thickness, a lower plate thickness, an upper plate hardness, a lower plate hardness, an upper plate strength, a lower plate strength, a rivet hardness, a rivet length, a nail foot opening value and a nail foot self-locking value to form a training sample;
taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in the training sample as the input of the network model, and taking the opening value and the self-locking value of the nail foot as the output of the network model for training to obtain the trained network model;
and storing the trained network model and calling the trained network model by a data calling module.
Calculating to obtain the key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested, wherein the key geometric parameters comprise:
calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The present invention is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are included in the scope of the claims of the present invention which are filed as the application.
Claims (15)
1. A big data prediction system for key geometric parameters of self-piercing riveting joints is characterized by comprising the following components: the data input module is used for inputting data;
the data calling module: determining a pre-trained network model corresponding to the type of the mould according to the type of the mould of the joint to be tested;
the data input module: inputting the technological parameters of the joint to be tested into the pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
the data calculation module: calculating to obtain key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested;
the key geometric parameters comprise the height of a nail head of the joint and a bottom thickness value;
the pre-trained network model is obtained by training joint historical process parameters by adopting an error back propagation algorithm.
2. The system of claim 1, further comprising a model training module to: the network model is used for dividing historical process parameter data into training samples corresponding to the types of the molds according to the types of the molds, and training the corresponding network models by adopting an error back propagation algorithm based on the training samples corresponding to the types of the molds.
3. The system of claim 2, wherein the model training module comprises: a training sample classification submodule, a training sample submodule, a training network model submodule and a data storage submodule;
the training sample classification submodule: the device is used for dividing historical process parameter data into training samples corresponding to each mould type according to the mould type;
the training sample submodule: the method comprises the steps of forming a training sample by upper plate thickness, lower plate thickness, upper plate hardness, lower plate hardness, upper plate strength, lower plate strength, rivet hardness, rivet length, opening value of a nail foot and a nail foot self-locking value;
the training network model submodule comprises: taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in the training sample corresponding to the mold type as the input of the network model, and taking the opening value and the pin self-locking value of the pins as the output of the network model for training to obtain a trained network model corresponding to the mold type;
the data storage submodule: and storing the trained network model corresponding to the mold type and calling the trained network model by the data calling module.
4. The system of claim 3, wherein the process parameters comprise: alloy grade, plate thickness, rivet specification, bottom die depth, riveting speed and nail foot opening value and self-locking value under corresponding process conditions.
5. The system of claim 4, wherein the data computation module comprises: the nail head height calculation submodule and the bottom thickness value calculation operator module;
the nail head height calculating submodule comprises: calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
the bottom thickness value calculating operator module: and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
6. The system of claim 5, further comprising an outcome prediction module to: and the method is used for judging the qualification of the key geometric parameters of the joint to be tested according to the set joint evaluation standard.
7. The system of claim 6, wherein the outcome prediction module comprises: a judgment submodule and an evaluation submodule;
the judgment submodule is used for judging the relationship between the key geometric parameters of the joint to be detected and a threshold value;
and the evaluation submodule is used for determining whether the to-be-detected connector is qualified or not based on the judgment result of the judgment submodule.
8. The system of claim 7, wherein the decision submodule comprises: the self-locking value judging unit, the nail head height judging unit and the bottom thickness value judging unit;
the self-locking value judging unit: judging the size of the self-locking value and a set self-locking threshold value;
the nail head height judging unit: judging whether the height of the nail head is within the range of the set nail head height threshold value;
the bottom thickness value judgment unit: and judging the bottom thickness value and the set bottom thickness threshold value.
9. The system of claim 8, wherein the evaluation sub-module comprises: the device comprises a self-locking value evaluation unit, a nail head height evaluation unit and a bottom thickness value evaluation unit;
the self-locking value evaluation unit: when the self-locking value is greater than or equal to the self-locking threshold value, the product is qualified, otherwise, the product is unqualified;
the nail head height evaluation unit: when the height of the nail head is within the range of the set height threshold value of the nail head, the nail head is qualified, otherwise, the nail head is unqualified;
the bottom thickness value evaluation unit: and when the bottom thickness value is larger than or equal to the bottom thickness threshold value, the product is qualified, otherwise, the product is unqualified.
10. The system of claim 9, wherein the outcome prediction module further comprises a display sub-module for displaying key geometric parameters of the joint under test.
11. The system of claim 10, wherein said pin head height calculation is given by:
h＝(Ddie-D0)·tanα/2-LCap (hat)，α＝arcsin(D-D0)/2LFoot
Wherein h is the height of the head of the joint and DDieIs the inner diameter of the die; d0The original diameter of the rivet; l isCap (hat)Is the height of the nail cap; d is the nail foot opening value.
12. The system of claim 11, wherein the base thickness is calculated as follows:
s＝t1+t2+Hdie-LCap (hat)-LFoot·cosα-h；α＝arcsin(D-D0)/2LFoot
Wherein s is a base thickness value; t is t1Is the thickness of the upper plate; t is t2Is the thickness of the lower plate; hDieIs the mold height; l isFootThe fixed focus height is obtained.
13. A big data prediction method for key geometric parameters of a self-piercing riveting joint is characterized by comprising the following steps:
calling a pre-trained network model corresponding to the type of the mold according to the type of the mold of the joint to be tested;
inputting the technological parameters of the joint to be tested into the called pre-trained network model to obtain the opening value and the pin self-locking value of the pin of the joint to be tested;
and calculating to obtain the key geometric parameters of the joint to be tested according to the opening value of the nail foot of the joint to be tested.
14. The method of claim 13, wherein the network model training comprises:
classifying historical data of the training sample according to a plurality of mould types;
the training sample comprises an upper plate thickness, a lower plate thickness, an upper plate hardness, a lower plate hardness, an upper plate strength, a lower plate strength, a rivet hardness, a rivet length, a nail foot opening value and a nail foot self-locking value to form a training sample;
taking the upper plate thickness, the lower plate thickness, the upper plate hardness, the lower plate hardness, the upper plate strength, the lower plate strength, the rivet hardness and the rivet length in the training sample as the input of the network model, and taking the opening value and the pin self-locking value of the pins as the output of the network model for training to obtain a trained network model;
and storing the trained network model and providing the trained network model for the data calling module to call.
15. The method as claimed in claim 14, wherein the calculating of the key geometric parameters of the joint to be tested from the opening values of the legs of the joint to be tested comprises:
calculating the height of the nail head by adopting a nail head height calculation formula based on the inner diameter of the die, the original diameter of the rivet, the height of the nail head and the opening value of the nail foot;
and calculating the bottom thickness value by adopting a bottom thickness calculation formula based on the thickness of the upper plate, the thickness of the lower plate, the height of the die and the length of the nail foot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010769060.5A CN112100753B (en) | 2020-08-03 | 2020-08-03 | System and method for predicting key geometric parameters of self-punching rivet joint of big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010769060.5A CN112100753B (en) | 2020-08-03 | 2020-08-03 | System and method for predicting key geometric parameters of self-punching rivet joint of big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112100753A true CN112100753A (en) | 2020-12-18 |
| CN112100753B CN112100753B (en) | 2024-03-22 |
Family
ID=73749514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010769060.5A Active CN112100753B (en) | 2020-08-03 | 2020-08-03 | System and method for predicting key geometric parameters of self-punching rivet joint of big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112100753B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541707A (en) * | 2020-12-24 | 2021-03-23 | 安徽巨一科技股份有限公司 | FDS bottom layer plate thickness judgment method and device, electronic equipment and storage medium |
| CN113591234A (en) * | 2021-06-16 | 2021-11-02 | 长三角先进材料研究院 | Self-piercing riveting process simulation model parameter analysis and checking method based on machine learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101908083A (en) * | 2010-05-13 | 2010-12-08 | 上海市特种设备监督检验技术研究院 | Artificial neural network-based Q345 welding joint mechanical property prediction method |
| CN103592114A (en) * | 2013-11-14 | 2014-02-19 | 昆明理工大学 | Method for detecting mechanical property of self-piercing riveting joint |
| CN105479771A (en) * | 2015-12-30 | 2016-04-13 | 吉林大学 | Manufacturing method for carbon fiber composite plate and self-punching riveting die and method for carbon fiber composite plate and aluminum alloy plate |
| CN106931918A (en) * | 2017-03-21 | 2017-07-07 | 武汉理工大学 | A kind of riveted joint geometric parameter detection method |
| CN107415345A (en) * | 2017-07-11 | 2017-12-01 | 昆明理工大学 | A kind of foamed metal sandwich board self-piercing riveting manufacturing process |
| CN109101701A (en) * | 2018-07-20 | 2018-12-28 | 昆明理工大学 | A kind of prediction technique of self-pierce riveting head nail pin opening width |
| WO2020114686A1 (en) * | 2018-12-03 | 2020-06-11 | Asml Netherlands B.V. | Method to predict yield of a semiconductor manufacturing process |
-
2020
- 2020-08-03 CN CN202010769060.5A patent/CN112100753B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101908083A (en) * | 2010-05-13 | 2010-12-08 | 上海市特种设备监督检验技术研究院 | Artificial neural network-based Q345 welding joint mechanical property prediction method |
| CN103592114A (en) * | 2013-11-14 | 2014-02-19 | 昆明理工大学 | Method for detecting mechanical property of self-piercing riveting joint |
| CN105479771A (en) * | 2015-12-30 | 2016-04-13 | 吉林大学 | Manufacturing method for carbon fiber composite plate and self-punching riveting die and method for carbon fiber composite plate and aluminum alloy plate |
| CN106931918A (en) * | 2017-03-21 | 2017-07-07 | 武汉理工大学 | A kind of riveted joint geometric parameter detection method |
| CN107415345A (en) * | 2017-07-11 | 2017-12-01 | 昆明理工大学 | A kind of foamed metal sandwich board self-piercing riveting manufacturing process |
| CN109101701A (en) * | 2018-07-20 | 2018-12-28 | 昆明理工大学 | A kind of prediction technique of self-pierce riveting head nail pin opening width |
| WO2020114686A1 (en) * | 2018-12-03 | 2020-06-11 | Asml Netherlands B.V. | Method to predict yield of a semiconductor manufacturing process |
Non-Patent Citations (1)
| Title |
|---|
| 刘洋等: "基于灰色理论和神经网络的自冲铆接头力学性能预测", 塑性工程学报, vol. 24, no. 4, pages 71 - 76 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541707A (en) * | 2020-12-24 | 2021-03-23 | 安徽巨一科技股份有限公司 | FDS bottom layer plate thickness judgment method and device, electronic equipment and storage medium |
| CN112541707B (en) * | 2020-12-24 | 2024-04-09 | 安徽巨一科技股份有限公司 | FDS bottom plate thickness determination method and device, electronic equipment and storage medium |
| CN113591234A (en) * | 2021-06-16 | 2021-11-02 | 长三角先进材料研究院 | Self-piercing riveting process simulation model parameter analysis and checking method based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112100753B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112100753B (en) | System and method for predicting key geometric parameters of self-punching rivet joint of big data | |
| US20030234239A1 (en) | Method and system for assessing quality of spot welds | |
| CN107478722A (en) | A kind of new resonance acoustics the cannot-harm-detection device | |
| CN111044621B (en) | Nondestructive testing system and method based on sound quality and acoustic characteristics | |
| CN103323526A (en) | Welding line defect detection and identification method based on ultrasonic phased array and support vector machine | |
| JP2022508535A (en) | Methods and equipment for processing workpieces | |
| CN105334105B (en) | A kind of acquisition methods and device of high speed blanking crack initiation Critical Damage threshold values | |
| US20220212245A1 (en) | Method for automatically determining quality of a self-piercing riveting process | |
| CN113486457A (en) | Die casting defect prediction and diagnosis system | |
| CN110749510B (en) | Method for detecting bending property of metal material based on finite element simulation | |
| JP6823025B2 (en) | Inspection equipment and machine learning method | |
| JP2001076022A (en) | Method for designing formed article made of aluminum alloy plate | |
| CN106290152A (en) | A kind of in-situ detection method for composite complex profile bonding quality | |
| CN116579657A (en) | Method for evaluating quality of metal lattice sandwich structure product in additive manufacturing | |
| CN112461920A (en) | Method and device for judging high-temperature alloy spot welding defects based on ultrasonic measurement | |
| WO2023280498A1 (en) | Method and system for generating a test coupon specification for predicting fatigue life of a component | |
| CN113779330A (en) | Ultrasonic detection process parameter intelligent calculation system based on similarity measurement | |
| Okuniewska et al. | Methodology for Diagnosing the Causes of Die-Casting Defects, Based on Advanced Big Data Modelling | |
| Balanovsky et al. | Quality control of welding in titanium panels, made using method of diffusion welding and superplastic forming | |
| CN117900981B (en) | Production process optimization method and system applied to TC4 tape processing | |
| Pacana et al. | Universal Model for the Quality Analysis of Aluminium Alloys Castings | |
| CN116429565B (en) | Method for detecting quality of joint of metal material without rivet connection | |
| JP2006341295A (en) | Method and device for predicting surface strain of press molded product | |
| KR102115959B1 (en) | Quantification method using a sinking index or an overlapping index of metal flow lines | |
| Baran et al. | Comparing deterministic and statistical approaches for predicting “short can” defects in aluminium beverage can production |
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 |