CN113536489B - Method for determining connection configuration and process parameters of component package - Google Patents
Method for determining connection configuration and process parameters of component package Download PDFInfo
- Publication number
- CN113536489B CN113536489B CN202110981544.0A CN202110981544A CN113536489B CN 113536489 B CN113536489 B CN 113536489B CN 202110981544 A CN202110981544 A CN 202110981544A CN 113536489 B CN113536489 B CN 113536489B
- Authority
- CN
- China
- Prior art keywords
- connection configuration
- parameters
- welding
- reliability
- determining
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 133
- 230000008569 process Effects 0.000 title claims abstract description 98
- 238000003466 welding Methods 0.000 claims abstract description 84
- 239000002245 particle Substances 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000003062 neural network model Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims abstract description 11
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000005476 soldering Methods 0.000 claims description 14
- 150000002500 ions Chemical class 0.000 claims description 9
- 229910000679 solder Inorganic materials 0.000 claims description 9
- 230000013016 learning Effects 0.000 claims description 7
- 239000000463 material Substances 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 238000001816 cooling Methods 0.000 claims description 6
- 238000002474 experimental method Methods 0.000 claims description 6
- 239000000758 substrate Substances 0.000 claims description 6
- 230000009194 climbing Effects 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 230000009326 social learning Effects 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 2
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 abstract description 5
- 238000004806 packaging method and process Methods 0.000 abstract description 5
- 238000012795 verification Methods 0.000 abstract description 4
- 238000007418 data mining Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012356 Product development Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012858 packaging process Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- 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
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- 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)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a method for determining connection configuration and process parameters of component packaging, which comprises the following steps: acquiring sample data of connection configuration and process parameters; and (B) step (B): training a neural network model by taking the process parameters as input and the connection configuration as output; step C: obtaining the relation between the connection configuration and the reliability of the welding spots through simulation analysis; step D: and establishing a process parameter-connection configuration-reliability inversion equation, and adjusting the process parameter based on a particle swarm algorithm until a calculation result meets the reliability requirement. The invention has the advantages that: the connection configuration is used as an intermediate parameter of the process parameter and the welding reliability, the relation between the process parameter and the connection reliability is determined through two mappings, an inversion equation is established, the parameters are updated through a particle swarm algorithm, the welding spot reliability under different parameters is calculated, and therefore the connection configuration and the process parameter can be obtained according to the welding spot reliability requirement without real welding verification, and the test cost and time are reduced.
Description
Technical Field
The invention relates to the technical field of electronic component packaging, in particular to a method for determining a connection configuration and a process parameter of component packaging.
Background
Along with the continuous advancement of electronic products to miniaturization, light weight and multifunctional directions, the packaging density of microsystems is higher and higher, the number of devices in a unit area is increased and the welding spots of an interconnection structure are smaller and smaller, the mechanical, electrical and thermal loads carried by the interconnection structure are heavier and heavier, and the reliability requirement on an interconnection interface is increased. According to Rohm aviation data, T/R component failure caused by a connection process factor is up to one third, and the reliability of a microwave component is researched from the connection process perspective. However, in the current connection process research, due process parameters are usually determined empirically in the design and manufacture stage of component packaging connection, actual measurement of the performance of welding spots is performed after sample pieces are tested, if the performance of the welding spots does not reach the standard, the process parameters are required to be changed again in the design stage, the expected performance index of the welding spots can be achieved only by continuous trial and error, and finally, the product development period is long and the input cost is high. The evolution mechanism of the interface structure of the connecting welding spots can be rarely analyzed from the angle of data mining, so that the assembly packaging process is optimized, and the stability of the process is improved; the influence of the microwave component welding spot connection configuration on the connection reliability life is seldom analyzed under the combined action of the data mining and the optimization algorithm.
The invention patent application with publication number of CN104239645A discloses a method and a system for designing vibration-resistant reliability of a micro-assembly component, wherein the relation between the width of a welding line and the natural frequency is analyzed based on a simulation model, the critical width of the welding line corresponding to the critical frequency is obtained, and the critical process parameters of the welding line are determined by taking the critical width of the welding line as a criterion, so that the cost problem of the traditional mode of welding firstly, then testing and then adjusting the parameters is solved, but only the width of the welding line is considered, then the process parameters of the welding are determined, and the influence of chain structure and different process combinations on the reliability of the welding line is not fully considered.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a method for determining welding connection configuration and technological parameters for component connection.
The invention solves the technical problems through the following technical scheme: a method for determining connection configuration and process parameters of a component package includes,
step A: acquiring sample data of connection configuration and process parameters;
and (B) step (B): training a neural network model by taking the process parameters as input and the connection configuration as output;
step C: obtaining the relation between the connection configuration and the reliability of the welding spots through simulation analysis;
step D: and establishing a process parameter-connection configuration-reliability inversion equation, and adjusting the process parameter based on a particle swarm algorithm until a calculation result meets the reliability requirement.
The invention takes the connection configuration as the intermediate parameter of the process parameter and the welding reliability, determines the relation between the process parameter and the connection reliability through two mappings, establishes an inversion equation, updates the parameters through a particle swarm algorithm and calculates the welding spot reliability under different parameters, and realizes the purpose of determining the welding process parameter based on the reliability requirement, thereby obtaining the connection configuration and the process parameter according to the welding spot reliability requirement without real welding verification, and reducing the test cost and time.
Preferably, the method for obtaining sample data of connection configuration and process parameters comprises,
step i: determining geometric parameters and physical parameters of the connection configuration;
step ii: determining key technological parameters of a welding process;
step iii: and (3) designing an experiment, obtaining the numerical value of key technological parameters of the equidistant welding process, and carrying out the welding experiment to obtain the corresponding connection configuration of each group of technological parameters.
Preferably, the geometric parameters of the connection configuration comprise solder joint height, solder joint extension width and solder joint thickness; the physical parameters include material type, material density, elastic modulus and temperature characteristics.
Preferably, the key technological parameters of the welding process comprise peak temperature, cooling rate and solder paste thickness in welding.
Preferably, the method further comprises the step of designing a simulation experiment based on the experimental data in the step iii, simulating the welding spot connection configuration under different process parameters, and expanding the sample data.
Preferably, in the step B, the sample data is preprocessed before training, the sample data is cleaned, interpolated and expanded, the preprocessed sample data is randomly divided, a training set and a testing set are constructed, peak temperature, cooling rate and soldering paste thickness in technological parameters are used as inputs of a neural network model, and soldering spot climbing height, soldering spot extension width and soldering spot thickness are used as outputs of the neural network model.
Preferably, the method for obtaining the connection configuration and the reliability relation of the welding spots through simulation analysis comprises the steps of establishing a corresponding finite element model according to the geometric dimensions of a substrate and components, setting physical parameters, applying corresponding boundary conditions and loads according to loads born by the substrate in an actual environment during working, evaluating the fatigue life of the welding spots under temperature impact based on an Engelmaier-Wild welding spot failure model, determining the association relation between the connection configuration and the stress strain distribution of the welding spots, and establishing a connection configuration and reliability relation network.
Preferably, the process parameter-connection configuration-reliability inversion equation includes,
and (3) a process parameter inversion model:
Find X
s.t.min{x 1 inversion x } is less than or equal to 1 ≤max{x 1 }
min{x 2 Inversion x } is less than or equal to 2 ≤max{x 2 }
…
min{x n Inversion x } is less than or equal to n ≤max{x n }
Connection configuration inversion model:
Find Y
s.t.min{y 1 inversion y 1 ≤max{y 1 }
min{y 2 Inversion y 2 ≤max{y 2 }
…
min{y n Inversion y n ≤max{y n }
Wherein X is a process parameter set, Y is a connection configuration set, Z is the service life of a welding spot, and X is i For the ith process parameter, y i For the ith connection configuration parameter, s.t. represents a constraint, RMSE represents a root mean square value.
Preferably, the method for adjusting the connection configuration based on the particle swarm algorithm is that,
step 1: initializing each parameter of an algorithm, setting the maximum iteration times, self-variable quantity, the maximum speed of ions, inertia weight, position information, self-and group learning factors, randomly initializing the speed and the position in a speed interval and a search space, setting the group size of a particle swarm, initializing a random flight speed of each ion, and setting target values and constraint conditions;
step 2: randomly generating a plurality of groups of connection configuration parameter combinations, and determining stress strain of the welding spots based on the relation between the connection configuration and the reliability of the welding spots to obtain a global optimal solution;
step 3: judging whether the global optimal solution meets a target value or not, or judging whether the maximum iteration times are reached, if yes, outputting the global optimal solution and corresponding connection configuration parameters, if not, updating the structural parameters and corresponding particle speeds, returning to the step 2, and comparing the updated optimal solution with the global optimal solution in the subsequent iteration process to update the global optimal solution;
after the connection configuration is determined, the process parameters are determined based on the same method as above.
Preferably, the optimization target of the particle swarm optimization is the maximum stress and maximum strain value of the dangerous welding spot, and the constraint condition is the interval range of each parameter; the formula for updating speed and position is:
V id ＝ωV id +C 1 random(0,1)(P id -X id )+C 2 random(0,1)(P gd -X id )
X id ＝X id +V id
wherein omega is more than or equal to 0, the larger the numerical value is, the stronger the global optimizing capability is, the smaller the numerical value is, and the stronger the local optimizing capability is; c (C) 1 And C 2 The acceleration constant is that the former is an individual learning factor of each ion, and the latter is a social learning factor of each particle; random (0, 1) represents interval [0,1 ]]Random number in, P id D-th dimension, P, representing the individual extremum of the i-th variable gd D-th dimension, V, representing globally optimal solution id D-th dimension, X representing particle group change rate of i-th variable id The d-th dimension of the particle space value representing the i-th variable.
The connecting configuration and the process parameter determining method of the component package provided by the invention have the advantages that: the connection configuration is used as an intermediate parameter of the process parameter and the welding reliability, the relation between the process parameter and the connection reliability is determined through two mappings, an inversion equation is established, the parameters are updated through a particle swarm algorithm, the welding spot reliability under different parameters is calculated, the purpose of determining the welding process parameter based on the reliability requirement is achieved, and therefore the connection configuration and the process parameter can be obtained according to the welding spot reliability requirement without real welding verification, and the test cost and time are reduced.
Drawings
FIG. 1 is a flow chart of a method for determining connection configuration and process parameters of a component package provided by an embodiment of the present invention;
fig. 2 is a schematic diagram of a connection configuration of a component package and a connection configuration of a process parameter determination method according to an embodiment of the present invention.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in FIG. 1, the embodiment provides a method for determining connection configuration and process parameters of a component package, which comprises
Step A: acquiring sample data of connection configuration and process parameters;
and (B) step (B): training a neural network model by taking the process parameters as input and the connection configuration as output;
step C: obtaining the relation between the connection configuration and the reliability of the welding spots through simulation analysis;
step D: and establishing a process parameter-connection configuration-reliability inversion equation, and adjusting the process parameter based on a particle swarm algorithm until a calculation result meets the reliability requirement.
In the embodiment, the connection configuration is used as the intermediate parameter of the process parameter and the welding reliability, the relation between the process parameter and the connection reliability is determined through two mappings, an inversion equation is established, the parameters are updated through a particle swarm algorithm, the welding spot reliability under different parameters is calculated, the purpose of determining the welding process parameter based on the reliability requirement is achieved, and therefore the connection configuration and the process parameter can be obtained according to the welding spot reliability requirement without actual welding verification, and the test cost and time are reduced.
In particular, the method for determining the connection configuration and the process parameters of the component package provided by the embodiment comprises the following steps,
step A: acquiring sample data of connection configuration and process parameters;
the method of obtaining sample data for connection configuration and process parameters includes,
step i: determining geometric parameters and physical parameters of the connection configuration;
referring to fig. 2, the geometric parameters include a solder joint height H, a solder joint extension degree L, and a solder joint thickness T; the physical parameters comprise material types, material density, elastic modulus, temperature characteristics and the like, wherein the temperature characteristics are characteristic parameters of the material changing along with the temperature.
Step ii: determining key technological parameters of a welding process;
the key technological parameters of the welding process comprise peak temperature, cooling rate and soldering paste thickness in welding.
Step iii: and (3) designing an experiment, obtaining the numerical value of key technological parameters of the equidistant welding process, and carrying out the welding experiment to obtain the corresponding connection configuration of each group of technological parameters.
Further, the method further comprises the steps of designing a simulation experiment through experimental data in the step iii, simulating the connection configuration of the welding spots after melting, solidification and molding under different process parameters, and verifying and correcting by referring to the experimental data so as to expand sample data.
And (B) step (B): training a neural network model by taking the process parameters as input and the connection configuration as output;
preprocessing sample data before training, including cleaning, interpolating and expanding the sample data, randomly dividing the preprocessed sample data to construct a training set and a testing set, and in the embodiment, taking 80% of data as the training set and 20% of data as the testing set to perform model training; the peak temperature, the cooling rate and the soldering paste thickness in the process parameters are used as inputs of the neural network model, and the soldering point climbing height, the soldering point extending width and the soldering point thickness are used as outputs of the neural network model.
And when training is carried out, the convolutional layer of the convolutional neural network learns to obtain characteristic parameters of process parameter training data, then the characteristic parameters are input into a pooling layer for summarization, the convolutional data is output, after 3 times of convolutional operation, the characteristic data of the previous layer is reconstructed in a fully connected layer and then is used as the input of the fully connected network, and fully connected calculation is carried out to obtain output data.
When the difference exists, the weight and the bias in each layer of network are finely adjusted by utilizing the back propagation learning of the neural network, the forward operation is carried out again after the weight and the bias are corrected, the error is calculated, the number of network layers can be increased in the training process, the network depth is deepened within the range of ensuring that the network cannot be fitted, the number of neurons of each layer is regulated, the training period is increased, and the training is stopped when the error meets the expected target or the loss value reaches the minimum and tends to be stable, so that the mapping model of the connection configuration is predicted by the technological parameters under the typical connection technology.
Step C: the connection configuration and the reliability relation of the welding spots are obtained through simulation analysis, and the specific method is that,
according to the geometric dimensions of a substrate and components, a corresponding finite element model is established in ANSYS software, selected physical parameters are set, corresponding boundary conditions and loads are applied according to loads born by the substrate in an actual environment during working, the fatigue life of a welding spot under temperature impact is evaluated based on an Engelmaier-Wild welding spot failure model, the association relation between a connection configuration and the stress-strain distribution of the welding spot is determined, and a connection configuration and reliability relation network is established.
Step D: and establishing a process parameter-connection configuration-reliability inversion equation, and adjusting the process parameter based on a particle swarm algorithm until a calculation result meets the reliability requirement.
The process parameter-connection configuration-reliability inversion equation includes,
and (3) a process parameter inversion model:
Find X
s.t.min{x 1 inversion x } is less than or equal to 1 ≤max{x 1 }
min{x 2 Inversion x } is less than or equal to 2 ≤max{x 2 }
…
min{x n Inversion x } is less than or equal to n ≤max{x n }
Connection configuration inversion model:
Find Y
s.t.min{y 1 inversion y 1 ≤max{y 1 }
min{y 2 Inversion y 2 ≤max{y 2 }
…
min{y n Inversion y n ≤max{y n }
Wherein X is a process parameter set, Y is a connection configuration set, Z is the service life of a welding spot, and X is i For the ith process parameter, y i For the ith connection configuration parameter, s.t. represents constraint, RMSE represents root mean square value, Z i Representing the solder joint life in the t-th iteration.
The method for adjusting the technological parameters based on the particle swarm algorithm comprises the following steps,
step 1: initializing each parameter of an algorithm, setting the maximum iteration times, self-variable quantity, the maximum speed of ions, inertia weight, position information, self-and group learning factors, randomly initializing the speed and the position in a speed interval and a search space, setting the group size of a particle swarm, initializing a random flight speed of each ion, and setting target values and constraint conditions;
step 2: randomly generating a plurality of groups of connection configuration parameter combinations, and determining stress strain of welding spots based on the relation between the connection configuration obtained in the step C and the reliability of the welding spots to obtain a global optimal solution;
step 3: judging whether the global optimal solution meets a target value or not, or judging whether the maximum iteration times are reached, if yes, outputting the global optimal solution and the corresponding connection configuration parameters, if not, updating the structural parameters and the corresponding particle speeds, returning to the step 2, and comparing the updated optimal solution with the global optimal solution in the subsequent iteration process to update the global optimal solution.
The optimization target of the particle swarm optimization is the maximum stress and maximum strain value of the dangerous welding spot, and the constraint condition is the interval range of each parameter; the formula for updating speed and position is:
V id ＝ωV id +C 1 random(0,1)(P id -X id )+C 2 random(0,1)(P gd -X id )
X id ＝X id +V id
wherein omega is more than or equal to 0, the larger the numerical value is, the stronger the global optimizing capability is, the smaller the numerical value is, and the stronger the local optimizing capability is; c (C) 1 And C 2 The acceleration constant is that the former is an individual learning factor of each ion, and the latter is a social learning factor of each particle; random (0, 1) represents interval [0,1 ]]Random number in, P id D-th dimension, P, representing the individual extremum of the i-th variable gd D-th dimension, V, representing globally optimal solution id D-th dimension, X representing particle group change rate of i-th variable id The d-th dimension of the particle space value representing the i-th variable.
And inverting the obtained connection configuration by the same method to determine the technological parameters of the connection configuration, so as to obtain the connection configuration and the technological parameters used for welding.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A method for determining connection configuration and process parameters of component package is characterized in that: comprising the steps of (a) a step of,
step A: acquiring sample data of connection configuration and process parameters;
and (B) step (B): training a neural network model by taking the process parameters as input and the connection configuration as output;
in the step B, preprocessing sample data before training, cleaning, interpolating and expanding the sample data, randomly dividing the preprocessed sample data, constructing a training set and a testing set, taking peak temperature, cooling rate and soldering paste thickness in process parameters as input of a neural network model, and taking soldering spot climbing height, soldering spot extending width and soldering spot thickness as output of the neural network model;
step C: obtaining the relation between the connection configuration and the reliability of the welding spots through simulation analysis;
step D: establishing a process parameter-connection configuration-reliability inversion equation, and adjusting the process parameter based on a particle swarm algorithm until a calculation result meets the reliability requirement;
the process parameter-connection configuration-reliability inversion equation includes,
and (3) a process parameter inversion model:
Find X
s.t.min{x 1 inversion x } is less than or equal to 1 ≤max{x 1 }
min{x 2 Inversion x } is less than or equal to 2 ≤max{x 2 }
…
min{x n Inversion x } is less than or equal to n ≤max{x n }
Connection configuration inversion model:
Find Y
s.t.min{y 1 inversion y 1 ≤max{y 1 }
min{y 2 Inversion y 2 ≤max{y 2 }
...
min{y n Inversion y n ≤max{y n }
Wherein X is a process parameter set, Y is a connection configuration set, Z is the service life of a welding spot, and X is i For the ith process parameter, y i For the ith connection configuration parameter, s.t. represents constraint, RMSE represents root mean square value, Z i Representing the solder joint life in the t-th iteration.
2. The method for determining the connection configuration and process parameters of a component package according to claim 1, wherein: the method of obtaining sample data for connection configuration and process parameters includes,
step i: determining geometric parameters and physical parameters of the connection configuration;
step ii: determining key technological parameters of a welding process;
step iii: and (3) designing an experiment, obtaining the numerical value of key technological parameters of the equidistant welding process, and carrying out the welding experiment to obtain the corresponding connection configuration of each group of technological parameters.
3. The method of determining the connection configuration and process parameters of a component package of claim 2, wherein: the geometric parameters of the connection configuration comprise welding spot climbing height, welding spot extension width and welding spot thickness; the physical parameters include material type, material density, elastic modulus and temperature characteristics.
4. The method of determining the connection configuration and process parameters of a component package of claim 2, wherein: the key technological parameters of the welding process comprise peak temperature, cooling rate and soldering paste thickness in welding.
5. The method of determining the connection configuration and process parameters of a component package of claim 2, wherein: the method further comprises the step of designing a simulation experiment based on the experimental data in the step iii, simulating the welding spot connection configuration under different process parameters, and expanding sample data.
6. The method for determining the connection configuration and process parameters of a component package according to claim 1, wherein: the method for acquiring the connection configuration and welding spot reliability relation through simulation analysis comprises the steps of establishing a corresponding finite element model according to the geometric dimensions of a substrate and components, setting physical parameters, applying corresponding boundary conditions and loads according to loads born by the substrate in an actual environment during working, evaluating the fatigue life of the welding spot under temperature impact based on an Engelmaier-Wild welding spot failure model, determining the association relation between the connection configuration and the welding spot stress strain distribution, and establishing a connection configuration and reliability relation network.
7. The method for determining the connection configuration and process parameters of a component package according to claim 1, wherein: the method for adjusting the connection configuration based on the particle swarm algorithm comprises the following steps,
step 1: initializing each parameter of an algorithm, setting the maximum iteration times, self-variable quantity, the maximum speed of ions, inertia weight, position information, self-and group learning factors, randomly initializing the speed and the position in a speed interval and a search space, setting the group size of a particle swarm, initializing a random flight speed of each ion, and setting target values and constraint conditions;
step 2: randomly generating a plurality of groups of connection configuration parameter combinations, and determining stress strain of the welding spots based on the relation between the connection configuration and the reliability of the welding spots to obtain a global optimal solution;
step 3: judging whether the global optimal solution meets a target value or not, or judging whether the maximum iteration times are reached, if yes, outputting the global optimal solution and corresponding connection configuration parameters, if not, updating the structural parameters and corresponding particle speeds, returning to the step 2, and comparing the updated optimal solution with the global optimal solution in the subsequent iteration process to update the global optimal solution;
after the connection configuration is determined, the process parameters are determined based on the same method as above.
8. The method of determining the connection configuration and process parameters of a component package of claim 7, wherein: the optimization target of the particle swarm optimization is the maximum stress and maximum strain value of the dangerous welding spot, and the constraint condition is the interval range of each parameter; the formula for updating speed and position is:
V id ＝ωV id +C 1 random(0,1)(P id -X id )+C 2 random(0,1)(P gd -X id )
X id ＝X id +V id
wherein omega is more than or equal to 0, the larger the numerical value is, the stronger the global optimizing capability is, the smaller the numerical value is, and the stronger the local optimizing capability is; c (C) 1 And C 2 The acceleration constant is that the former is an individual learning factor of each ion, and the latter is a social learning factor of each particle; random (0, 1) represents interval [0,1 ]]Random number in, P id D-th dimension, P, representing the individual extremum of the i-th variable gd D-th dimension, V, representing globally optimal solution id D-th dimension, X representing particle group change rate of i-th variable id The d-th dimension of the particle space value representing the i-th variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110981544.0A CN113536489B (en) | 2021-08-25 | 2021-08-25 | Method for determining connection configuration and process parameters of component package |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110981544.0A CN113536489B (en) | 2021-08-25 | 2021-08-25 | Method for determining connection configuration and process parameters of component package |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113536489A CN113536489A (en) | 2021-10-22 |
| CN113536489B true CN113536489B (en) | 2023-06-20 |
Family
ID=78092877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110981544.0A Active CN113536489B (en) | 2021-08-25 | 2021-08-25 | Method for determining connection configuration and process parameters of component package |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113536489B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114147380B (en) * | 2021-12-08 | 2023-06-20 | 华东理工大学 | Rotor welding method based on digital twin frame |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6023663A (en) * | 1996-04-08 | 2000-02-08 | Samsung Electronics Co., Ltd. | Method and apparatus for inspecting a solder joint using a correlation neural network |
| CN108875219A (en) * | 2018-06-21 | 2018-11-23 | 桂林电子科技大学 | A kind of BGA welding spot structure parameter optimization method reducing power cycle stress |
| CN109865916A (en) * | 2019-03-19 | 2019-06-11 | 山东大学 | A kind of robot welding process parameter optimizing method based on CBR and RBR |
| CN110405343A (en) * | 2019-08-15 | 2019-11-05 | 山东大学 | A kind of laser welding process parameter optimization method of the prediction model integrated based on Bagging and particle swarm optimization algorithm |
| CN111125941A (en) * | 2019-07-08 | 2020-05-08 | 桂林电子科技大学 | CSP solder joint post-welding residual stress measuring method |
| CN111581862A (en) * | 2020-04-20 | 2020-08-25 | 湖南大学 | Equivalent test method for mechanical property of welding joint microcell |
| CN112257280A (en) * | 2020-10-29 | 2021-01-22 | 桂林电子科技大学 | Liquid-solid phase continuous form prediction method for reflow soldering BGA group soldering points |
| CN113199184A (en) * | 2021-07-05 | 2021-08-03 | 北京航空航天大学 | Weld joint shape prediction method based on improved self-adaptive fuzzy neural network |
-
2021
- 2021-08-25 CN CN202110981544.0A patent/CN113536489B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6023663A (en) * | 1996-04-08 | 2000-02-08 | Samsung Electronics Co., Ltd. | Method and apparatus for inspecting a solder joint using a correlation neural network |
| CN108875219A (en) * | 2018-06-21 | 2018-11-23 | 桂林电子科技大学 | A kind of BGA welding spot structure parameter optimization method reducing power cycle stress |
| CN109865916A (en) * | 2019-03-19 | 2019-06-11 | 山东大学 | A kind of robot welding process parameter optimizing method based on CBR and RBR |
| CN111125941A (en) * | 2019-07-08 | 2020-05-08 | 桂林电子科技大学 | CSP solder joint post-welding residual stress measuring method |
| CN110405343A (en) * | 2019-08-15 | 2019-11-05 | 山东大学 | A kind of laser welding process parameter optimization method of the prediction model integrated based on Bagging and particle swarm optimization algorithm |
| CN111581862A (en) * | 2020-04-20 | 2020-08-25 | 湖南大学 | Equivalent test method for mechanical property of welding joint microcell |
| CN112257280A (en) * | 2020-10-29 | 2021-01-22 | 桂林电子科技大学 | Liquid-solid phase continuous form prediction method for reflow soldering BGA group soldering points |
| CN113199184A (en) * | 2021-07-05 | 2021-08-03 | 北京航空航天大学 | Weld joint shape prediction method based on improved self-adaptive fuzzy neural network |
Non-Patent Citations (4)
| Title |
|---|
| A Novel Vertical Wire-Bonding Compensation Structure Adaptively Modeled and Optimized With GRNN and GA Methods for System in Package;Hao-Ran Zhu et al.;《IEEE Transactions on Electromagnetic Compatibility》;第63卷(第6期);第2-10页 * |
| Process optimization of SnCuNi soldering material using artificial parametric design;Chien-Yi Huang et al.;《Journal of Intelligent Manufacturing》;第25卷;第3-10页 * |
| 基于遗传算法与神经网络微电阻点焊工艺参数优化;高星鹏等;《宇航材料工艺》(第3期);第34-37页 * |
| 结合DFM及焊点形态理论对SMT虚拟组装系统的研究;胡元君;《信息记录材料》;第19卷(第4期);第3-4页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113536489A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107303675A (en) | Optimize the cell controller of the action of the production system with many industrial machineries | |
| CN109783970B (en) | Thermal analysis method for reliability simulation analysis of electronic product | |
| CN103559418A (en) | PoF (physics of failure) based method for calculating mission reliability of electronic product | |
| CN113536489B (en) | Method for determining connection configuration and process parameters of component package | |
| CN112414668B (en) | Wind tunnel test data static bomb correction method, device, equipment and medium | |
| US20170255718A1 (en) | Method and system for determing welding process parameters | |
| CN116305522A (en) | Digital twin-based aircraft structure reliability simulation test model calibration method, device and medium | |
| US9102012B2 (en) | Method and system for determining welding sequences | |
| CN112580228A (en) | Fan blade structural design optimization method for mixed structure of turbofan engine | |
| CN116579207A (en) | Variable reliability model determining method for predicting and optimizing residual stress after welding and spot welding | |
| CN104484511A (en) | Simulation analysis based dynamic characteristic design method for robot structures | |
| CN114564787A (en) | Bayesian optimization method, device and storage medium for target-related airfoil design | |
| CN111756587B (en) | Method for predicting time sequence network link by using GraphSAGE | |
| CN115659895B (en) | Method and device for reducing warping degree of packaged chip, storage medium and electronic equipment | |
| CN117574721A (en) | Process parameter probability model optimization method | |
| US20100063946A1 (en) | Method of performing parallel search optimization | |
| Sineglazov et al. | An algorithm for solving the problem of forecasting | |
| CN116628854A (en) | Wing section aerodynamic characteristic prediction method, system, electronic equipment and storage medium | |
| CN115587688A (en) | Scheduling method and system of assembly type building components based on BIM | |
| CN114819107A (en) | Mixed data assimilation method based on deep learning | |
| CN114580139A (en) | Method for automatically evaluating a simulation model | |
| CN109753018B (en) | Error compensation system and dynamic compensation method based on cloud intelligence | |
| CN110543724A (en) | Satellite structure performance prediction method for overall design | |
| Safikhani et al. | Pareto optimization of two-element wing models with morphing flap using computational fluid dynamics, grouped method of data handling artificial neural networks and genetic algorithms | |
| CN112214843A (en) | Finite element rigidity correction method and device for wind tunnel test wing model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |