SMT Packaging Reliability Optimization Method for Interface Contact Pressure

Abstract

The present invention discloses an SMT packaging reliability optimization method for interface contact pressure, which relates to the technical field of electronic packaging, and includes: determining the geometric parameters, physical properties parameters and design domain of the packaging; sampling within the design domain; constructing a parametric characterization model of the packaging; constructing a solder joint viscoplastic constitutive model, constructing a solder joint cohesive force model and constructing a solder joint temperature cycle load curve; constructing a finite element model under temperature cycle load; performing finite element simulation analysis to form contact pressure sample points; constructing a Kriging surrogate model; obtaining updated geometric parameter sample points; judging whether the convergence stagnates; if not, performing finite element simulation analysis until the convergence stagnates; if so, judging whether the termination criterion is satisfied; if so, outputting the geometric parameters corresponding to the minimum contact pressure; if not, updating the geometric parameter sample points and performing finite element simulation analysis until the convergence stagnates. This application can effectively improve the reliability of SMT packaging.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic packaging, and particularly relates to a method for optimizing the reliability of SMT packaging for interface contact pressure. Background Art

[0002] With the improvement of the integration degree of semiconductor equipment, the electronic packaging structure plays a crucial role in ensuring the normal function of semiconductor equipment, and at the same time puts forward higher requirements for the reliability of electronic packaging. As a widely used form of electronic packaging, Surface Mount Technology (SMT) plays an important role in integrated circuits. However, due to the alternating thermal stress caused by temperature cycling, there are often thermal reliability problems. Especially when the contact pressure between the SMT solder joint interfaces is too large, it means that reliability problems such as cracks or even fractures are likely to occur between the SMT solder joint interfaces. Optimizing the contact pressure between the SMT solder joint interfaces is of great significance for improving the reliability of SMT packaging.

[0003] In the prior art, although traditional finite element simulation methods can provide high-precision mechanical distribution characteristics, in the face of optimization problems that need to be achieved through continuous iteration, a large number of finite element simulations mean extremely high computational costs and time, which are difficult to achieve for the rapid optimization design of electronic products. In addition, the existing optimization of SMT packaging reliability often relies on experience, and there is blindness and subjectivity in the design of structural parameters, and the reliable and robust improvement cannot be guaranteed.

[0004] Therefore, there is an urgent need to develop a method for optimizing the reliability of SMT packaging for interface contact pressure. Summary of the Invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a method for optimizing the reliability of SMT packaging for interface contact pressure. The technical problems to be solved by the present invention are realized through the following technical solutions:

[0006] In a first aspect, the present application provides a method for optimizing the reliability of SMT packaging for interface contact pressure, including:

[0007] Determine the geometric parameters, physical property parameters, and design domain of the package according to the package structure;

[0008] Use the improved Latin hypercube algorithm to sample within the design domain to obtain initial geometric parameter sample points;

[0009] Based on the principle of minimum energy, construct a parametric representation model of the package corresponding to the initial geometric parameter sample points;

[0010] Construct a solder joint viscoplastic constitutive model, construct a solder joint cohesive force model, and construct a solder joint temperature cycle load curve;

[0011] Based on the encapsulation parametric characterization model, the solder joint viscoplastic constitutive model, the solder joint cohesive force model, and the solder joint temperature cycle load curve, a finite element model under temperature cycle load is constructed, and the convergence stagnation threshold and the number of iterations are determined;

[0012] Based on the finite element model, the encapsulation structure corresponding to the initial geometric parameter sample points is subjected to finite element simulation analysis to obtain the maximum contact pressure at the encapsulation solder joint interface, and contact pressure sample points are formed;

[0013] Based on the initial geometric parameter sample points and the contact pressure sample points, a Kriging surrogate model is constructed;

[0014] Based on the wEI filling criterion, by maximizing the wEI value of the current Kriging surrogate model, updated geometric parameter sample points are obtained;

[0015] Based on the convergence stagnation criterion, it is judged whether convergence has stagnated; if not, continue with the finite element simulation analysis until convergence stagnates; if so, judge whether the termination criterion is satisfied; if so, output the geometric parameters corresponding to the minimum contact pressure; if not, based on the MAE criterion, update the geometric parameter sample points again and continue with the finite element simulation analysis until convergence stagnates.

[0016] Advantages of the present invention:

[0017] A method for optimizing the reliability of SMT packaging for interface contact pressure provided by the present invention establishes a parametric characterization model with SMT as the object. Based on this, the temperature cycle load analysis of the SMT packaging structure is realized by using ANSYS. Then, the interface contact pressure of the SMT packaging is gradually optimized through the adaptive filling criterion, improving the reliability of the SMT packaging; minimizing the invocation of unnecessary high-cost simulation calculations, achieving the effect of reducing the interface contact pressure, reducing the possibility of solder joint cracking or fracture, and improving the reliability of the SMT packaging; in addition, this embodiment can well solve the problems of global optimization and minimizing the number of simulation calculation invocations by using ANSYS software, the Kriging surrogate model, and the adaptive filling criterion.

[0018] The following will further elaborate on the present invention in conjunction with the drawings and embodiments. Description of the Drawings

[0019] Figure 1 is a flowchart of a method for optimizing the reliability of SMT packaging for interface contact pressure provided by an embodiment of the present invention;

[0020] Figure 2 is a schematic structural diagram of an SMT packaging structure provided by an embodiment of the present invention;

[0021] Figure 3 It is a schematic diagram of SMT packaging parameters provided by an embodiment of the present invention;

[0022] Figure 4 It is another schematic diagram of SMT packaging parameters provided by an embodiment of the present invention;

[0023] Figure 5 It is a curve schematic diagram of the temperature cycle load of the SMT packaging structure provided by an embodiment of the present invention;

[0024] Figure 6 It is a schematic diagram of a finite element model of the SMT packaging structure provided by an embodiment of the present invention;

[0025] Figure 7 It is a curve schematic diagram of the iterative optimization of the SMT interface contact pressure provided by an embodiment of the present invention. Detailed implementation manners

[0026] The present invention will be further described in detail below with reference to specific embodiments, but the implementation manners of the present invention are not limited thereto.

[0027] To solve the above-mentioned defects in the prior art, the present application provides a packaging reliability optimization method for interface contact pressure, which can reduce the product development cycle, improve reliability, and reduce production and manufacturing costs.

[0028] Please refer to Figures 1 - 6 , Figure 1 It is a flowchart of a method for optimizing the reliability of SMT packaging for interface contact pressure provided by an embodiment of the present invention, Figure 2 It is a schematic structural diagram of an SMT packaging structure provided by an embodiment of the present invention, Figure 3 It is a schematic diagram of SMT packaging parameters provided by an embodiment of the present invention, Figure 4 It is another schematic diagram of SMT packaging parameters provided by an embodiment of the present invention, Figure 5 It is a curve schematic diagram of the temperature cycle load of the SMT packaging structure provided by an embodiment of the present invention, Figure 6 It is a schematic diagram of a finite element model of the SMT packaging structure provided by an embodiment of the present invention. A method for optimizing the reliability of SMT packaging for interface contact pressure provided by the present application includes:

[0029] S101. Determine the geometric parameters, physical property parameters, and design domain of the package according to the package structure;

[0030] S102. Sample in the design domain by using an improved Latin hypercube algorithm to obtain initial geometric parameter sample points;

[0031] S103. Based on the principle of minimum energy, construct a packaging parametric characterization model corresponding to the initial geometric parameter sample points;

[0032] S104. Construct a solder joint viscoplastic constitutive model, a solder joint cohesive force model, and a solder joint temperature cycle load curve;

[0033] S105. Based on the packaging parametric characterization model, the solder joint viscoplastic constitutive model, the solder joint cohesive force model, and the solder joint temperature cycle load curve, construct a finite element model under temperature cycle load, and determine the convergence stagnation threshold and the number of iterations;

[0034] S106. Based on the finite element model, perform finite element simulation analysis on the packaging structure corresponding to the initial geometric parameter sample points to obtain the maximum contact pressure at the packaging solder joint interface, and form contact pressure sample points;

[0035] S107. Based on the initial geometric parameter sample points and the contact pressure sample points, construct a Kriging surrogate model;

[0036] S108. Based on the wEI filling criterion, obtain updated geometric parameter sample points by maximizing the wEI value of the current Kriging surrogate model;

[0037] S109. Judge whether convergence has stagnated based on the convergence stagnation criterion; if not, continue with the finite element simulation analysis, that is, execute step S106 until convergence stagnates; if so, judge whether the termination criterion is satisfied; if the termination condition is satisfied, output the geometric parameters corresponding to the minimum contact pressure; if the termination condition is not satisfied, based on the MAE criterion, update the geometric parameter sample points again and continue with the finite element simulation analysis, that is, execute step S106 until convergence stagnates.

[0038] Specifically, a method for optimizing the reliability of SMT packaging for interface contact pressure provided in this embodiment establishes a parametric characterization model with SMT as the object. Based on this, the temperature cycle load analysis of the SMT packaging structure is realized by using ANSYS. Then, the interface contact pressure of the SMT packaging is gradually optimized through the adaptive filling criterion to improve the reliability of the SMT packaging; it minimizes the invocation of unnecessary high-cost simulation calculations, achieving the effect of reducing the interface contact pressure, reducing the possibility of solder joint cracking or fracture, and improving the reliability of the SMT packaging; in addition, this embodiment can well solve the problems of global optimization and minimizing the number of simulation calculation invocations by using ANSYS software, the Kriging surrogate model, and the adaptive filling criterion.

[0039] The purpose of this application is to provide a method for optimizing the packaging reliability for interface contact pressure, so as to improve the blindness of the traditional SMT interface contact pressure optimization design, reduce the computational cost and computational time of the optimization process, and enhance the thermal reliability of SMT packaging. By constructing a Kriging surrogate model and using the wEI (weighted Expected Improvement) filling criterion and the MAE (Mean Absolute Error) secondary filling criterion, while reducing the invocation of high-cost simulation calculations, the optimization accuracy and efficiency are improved.

[0040] It should be noted that the packaging in this embodiment adopts the Surface Mount Technology (SMT). The SMT packaging structure includes a capacitor 10, a metal terminal 20, a solder joint 30, a solder pad 40, and a substrate. Among them, the capacitor can be a ceramic capacitor. The structure composed of the capacitor 10, the metal terminal 20, the solder joint 30, and the solder pad 40 is located on the substrate (not shown in the figure). The metal terminal 20 envelopes both ends of the capacitor 10, and the metal terminal 20 is fixed to the solder pad 40 through the solder joint 30. According to the packaging structure, the geometric parameters of the packaging are determined, including the capacitor length L1, the capacitor width W1, the capacitor thickness H1, the metal terminal length L2, the solder joint height H3, the solder joint protrusion length L3, the internal length of the solder joint L5, the solder joint width W3, the solder pad length L4, the solder pad width W4, and the gap H2 between the solder pad and the metal terminal. The design domain of the packaging is determined to include the solder joint height wetting length solder joint gap height The physical properties parameters of the packaging are determined, including the solder joint density ρ1, the substrate density ρ2, the solder joint elastic modulus E1, the substrate elastic modulus E2, the solder joint Poisson's ratio μ1, the substrate Poisson's ratio μ2, the solder joint coefficient of thermal expansion CTE1, and the substrate coefficient of thermal expansion CTE2.

[0041] Optionally, for the selection of geometric parameters in this embodiment, please refer to Table 1.

[0042] Table 1 Packaging geometric parameters

[0043]

[0044]

[0045] Optionally, for the selection of physical properties parameters in this embodiment, please refer to Table 2.

[0046] Table 2 Physical properties parameters of array solder joints and packaging structure

[0047]

[0048] Optionally, for the change of the solder joint elastic modulus with temperature in this embodiment, please refer to Table 3.

[0049] Table 3 Variation of the Elastic Modulus of Solder Joints with Temperature

[0050] Temperature (°C) -80 -65 -55 0 25 65 105 EX (MPa) 54497 50994 48658 25812 29973 20629 12455

[0051] Optionally, for the selection of the encapsulation structure design domain in this embodiment, please refer to Table 4.

[0052] Table 4 Variation of the Elastic Modulus of Solder Joints with Temperature

[0053] Geometric parameter <![CDATA[H s (mm)]]> <![CDATA[L w (mm)]]> <![CDATA[H g (mm)]]> Design domain 0.62-0.72 0.9-1.4 0.09-0.14

[0054] In an optional embodiment of the present application, the sampling process of the improved Latin hypercube algorithm includes:

[0055] Dividing the domain value range of the design domain into m equal-length intervals, and randomly selecting a geometric parameter sample point within each interval to form m + 1 sample sets;

[0056] Generating n sets of sampling sets, and obtaining the Euclidean distances between the m + 1 geometric parameter sample points in each set of sampling sets; wherein, the distance between adjacent geometric parameter sample points is defined as the Euclidean distance;

[0057] Comparing the minimum Euclidean distances in each sampling set, and selecting the sampling set corresponding to the maximum Euclidean distance therefrom as the set x0 of the initial geometric parameter sample points.

[0058] Specifically, in this embodiment, the process of obtaining the initial geometric parameter sample points includes:

[0059] The solder joint height of the three-dimensional design domain Wetting length Solder joint gap height Are respectively divided into m equal-length intervals, and a geometric parameter sample point is randomly selected within each equal-length interval, and a total of m + 1 sample sets are formed; it should be noted that the wetting length is the depth length after the solder joint is formed.

[0060] Defining the distance between adjacent geometric parameter sample points as the Euclidean distance, using the above-mentioned Latin hypercube sampling method to simultaneously generate 100 sets of sampling sets, solving the Euclidean distances between the m + 1 geometric parameter sample points in each set of sampling sets, comparing the minimum Euclidean distances in these 100 sets of sampling sets, and selecting a set of sampling sets corresponding to the maximum Euclidean distance therefrom as the set x0 of the initial geometric parameter sample points, and the expression of x0 is:

[0061]

[0062] Optionally, for the geometric parameter design variable sample points selected by the improved Latin hypercube in this embodiment, please refer to Table 5.

[0063] Table 5 Sample Points of Geometric Parameter Design Variables

[0064] <![CDATA[Welding spot height h s / mm]]> <![CDATA[Wetting length l s / mm]]> <![CDATA[Clearance height h g / mm]]> 0.620 0.900 0.090 0.670 1.150 0.115 0.645 1.275 0.103 0.695 1.025 0.128 0.633 1.213 0.134 0.683 0.963 0.109 0.678 1.018 0.099 0.653 1.143 0.136 0.703 1.393 0.111

[0065] In an alternative embodiment of the present application, the encapsulation parametric characterization model is constructed based on solder volume constraint, solder gravitational potential energy constraint, and solder surface potential energy constraint.

[0066] Specifically, in this embodiment, considering that when the solder melts into a liquid state, the entire solder joint system spontaneously tends to have the minimum energy under volume constraint, gravitational potential energy constraint, and surface potential energy constraint, and an encapsulation parametric characterization model is established based on this.

[0067] Among them, under the volume constraint, the relationship between the solder volume V and the geometric parameters is as follows:

[0068] V = 0.5[(W3 + L2)·H2 + L3·H3]·W1;

[0069] Under the gravitational potential energy constraint, the gravitational potential energy E g of the solder itself is expressed as:

[0070]

[0071] ρ is the density of the molten solder, g is the acceleration due to gravity, and z is the coordinate height of the solder joint;

[0072] Under the gravitational potential energy constraint, the expression for the potential energy provided by the external force generated by the component above the solder is:

[0073]

[0074] m is the mass of the component, and h is the height from the bottom surface of the component to the solder pad;

[0075] Under the surface potential energy constraint, the surface potential energy E s1 at the solder-pad contact surface z = 0 is expressed as:

[0076]

[0077] T1 is the equivalent surface tension between the solder and the solder-pad surface;

[0078] Under the surface potential energy constraint, the surface potential energy E s2 at the x-direction perpendicular interface between the solder and the metal terminal is expressed as:

[0079]

[0080] T2 is the equivalent surface tension between the solder and the x-direction perpendicular interface of the metal terminal;

[0081] Under the constraint of surface potential energy, the surface potential energy E of the solder and the vertical interface of the metal terminal in the y-direction s3 and E s4 The expressions are respectively:

[0082]

[0083] T3 and T4 are the equivalent surface tensions between the solder and the vertical interface of the metal pin in the y-direction;

[0084] Under the constraint of surface potential energy, the surface potential energy E of the solder and the vertical interface of the metal terminal in the z-direction s5 can be expressed as:

[0085]

[0086] T5 is the equivalent surface tension between the solder and the vertical interface of the metal pin in the z-direction;

[0087] Under the constraint of surface potential energy, the surface potential energy E of the solder liquid surface itself s6 The expression is:

[0088]

[0089] T is the surface tension of the molten liquid solder itself;

[0090] In summary, under the constraint of surface potential energy, the total surface potential energy E of the solder s is:

[0091]

[0092] In an optional embodiment of the present application, the solder joint viscoplastic constitutive model is constructed based on the solder joint viscoplastic constitutive model parameters, and the solder joint viscoplastic constitutive model parameters include: initial deformation resistance s0, pre-exponential factor A, stress multiplier ξ, strain rate sensitivity index m, strain hardening constant h0, deformation resistance coefficient strain rate sensitivity index n, strain hardening sensitivity coefficient a, and activation energy Q / R;

[0093] The solder joint cohesion model is constructed based on the solder joint cohesion parameters, and the solder joint cohesion parameters include: the normal critical energy release rate is tangential critical energy release rate normal maximum contact stress σ max , tangential maximum contact stress τ max ;

[0094] The temperature cycle load curve is constructed based on the temperature cycle load parameters, and the temperature cycle load parameters include: the thermal cycle temperature range T1℃:T2℃ applied to the package structure, and the reference temperature under zero stress strain is T ref, single cycle period t sig , including heating for t1 minutes, cooling for t2 minutes, high and low temperature insulation for each t minutes, the number of thermal cycle load periods cycles, and the load sub-steps of a single cycle period.

[0095] Specifically, in this embodiment, based on the encapsulation parametric characterization model, the solder joint viscoplastic constitutive model, the solder joint cohesive force model, and the solder joint temperature cycle load curve, a finite element model under temperature cycle load is constructed, and the convergence stagnation threshold and the number of iterations are determined; specifically:

[0096] Determine the parameters of the solder joint viscoplastic constitutive model and construct the solder joint viscoplastic constitutive model; among them, the parameters of the solder joint viscoplastic constitutive model include the initial deformation impedance s0, the pre-exponential factor A, the stress multiplier ξ, the strain rate sensitivity index m, the strain hardening constant h0, the deformation impedance coefficient the strain rate sensitivity index n, the strain hardening sensitivity coefficient a, and the activation energy Q / R; the expression of the constructed solder joint viscoplastic constitutive model is:

[0097] σ = cs, c < 1;

[0098] In the formula, σ is the stress, s is the deformation impedance, and c is a function related to the strain rate and temperature, and its expression is:

[0099]

[0100] In the formula, T is the absolute temperature, R is the gas constant, ε p is the inelastic strain, is the inelastic strain rate. Among them, the solder joint viscoplastic constitutive model describes the steady-state plastic flow as:

[0101]

[0102]

[0103] In the formula, σ * is the saturation stress, s * is the saturation value of s at a given temperature and strain rate;

[0104] According to the above definition, the stress of the solder joint viscoplastic constitutive model can be expressed as:

[0105]

[0106] Determine the solder joint cohesion parameters and construct the solder joint cohesion model; among them, the solder joint cohesion parameters include the normal critical energy release rate as the tangential critical energy release rate the normal maximum contact stress σ max , the tangential maximum contact stress τ max; The expression of the constructed solder joint cohesion model is as follows:

[0107]

[0108]

[0109] In the formula, T n is the normal contact stress, T t is the tangential contact stress, σ max is the maximum normal contact stress, τ max is the maximum tangential contact stress, is the contact gap under the maximum normal contact stress, is the contact gap under the maximum tangential contact stress, is the maximum gap when the interface is completely separated under the normal force, is the maximum gap when the interface is completely separated under the tangential force;

[0110] The normal and tangential critical energy release rates in the cohesion model are and Their expressions are respectively:

[0111]

[0112]

[0113] Determine the temperature cycle load parameters and construct the temperature cycle load curve; among them, the temperature cycle load parameters include: the thermal cycle temperature range T1 °C to T2 °C applied to the package structure, the reference temperature at zero stress strain is T ref , the single cycle period t sig (min), including t1 minutes of heating up, t2 minutes of cooling down, and t minutes of high and low temperature heat preservation for each cycle, the number of thermal cycle load periods cycles, and the load sub-steps steps of a single cycle period.

[0114] Determine the convergence stagnation threshold and the number of iterations, where the convergence stagnation threshold Δ0 = 0.001 and the maximum number of iterations i0 = 100.

[0115] Optionally, for the selection of the solder joint viscoplastic constitutive model parameters in this embodiment, please refer to Table 6.

[0116] Table 6 Solder joint viscoplastic constitutive model parameters

[0117]

[0118] Optionally, for the selection of the solder joint cohesion model parameters in this embodiment, please refer to Table 7.

[0119] Table 7 Solder joint viscoplastic constitutive model parameters

[0120]

[0121] In an alternative embodiment of the present application, the expression of the Kriging surrogate model is:

[0122]

[0123] Wherein, is the response value of any unknown point x within the design domain, s 2 (x) is the variance of any unknown point x within the design domain, x0 is the initial geometric parameter sample point, y0 is the contact pressure sample point; μ is the mean of the Gaussian process, r is the covariance matrix between the initial geometric parameter sample point x0 and any unknown point x within the design domain, and C is the covariance matrix between the initial geometric parameter sample points x0.

[0124] It should be noted that the geometric parameter design variable x0 is the input sample during the model training process, and the output is the contact pressure y0 at the encapsulation interface.

[0125] It should be noted that before this, it is also necessary to extract the maximum contact pressure y0 at the SMT solder joint interface according to the ANSYS thermal cycle load simulation analysis, and its expression is:

[0126]

[0127] In an alternative embodiment of the present application, based on the wEI filling criterion, the process of obtaining the updated geometric parameter sample points by maximizing the wEI value of the current Kriging surrogate model includes:

[0128] The expression of the wEI filling criterion is:

[0129]

[0130]

[0131] Wherein, w is the weight of the global search ability, i is the current iteration number, Φ( ) is the cumulative probability distribution function of the standard normal distribution, φ( ) is the probability density function of the standard normal distribution, f min is the optimal point calculated by the current Kriging surrogate model, s(x) is the standard deviation of any unknown point within the design domain, is the response value of any unknown point x within the design domain, and i0 is the maximum iteration number;

[0132] The obtained updated geometric parameter sample point x a , and its calculation formula is:

[0133] Find: x a

[0134]

[0135] In an alternative embodiment of the present application, the process of determining whether convergence has stagnated based on the convergence stagnation criterion includes:

[0136] Determining whether convergence has stagnated according to the convergence stagnation formula, where the expression of the convergence stagnation formula is:

[0137]

[0138] where min( ) is the minimum value function, max( ) is the maximum value function, and i is the current iteration number; if Δ ≤ Δ0, where Δ0 is the convergence stagnation threshold and Δ0 = 0.001, then convergence has stagnated; otherwise, convergence has not stagnated.

[0139] In an alternative embodiment of the present application, the process of determining whether the termination criterion is satisfied includes:

[0140] Determining whether the current iteration number i is greater than the maximum iteration number; if so, outputting the geometric parameters corresponding to the minimum contact pressure; where the maximum iteration number i0 = 100.

[0141] In an alternative embodiment of the present application, the process of updating the geometric parameter sample points again based on the MAE criterion includes:

[0142] Increasing the geometric parameter sample point x b , and determining the MAE criterion model, whose expression is:

[0143] Find: x b

[0144] Maximize: s 2 (x b );

[0145] where s 2 (x) is the variance of any unknown point x within the design domain.

[0146] In an alternative embodiment of the present application, please refer to Figure 7 , Figure 7 which is a schematic curve diagram of the iterative optimization of the SMT interface contact pressure provided by the embodiments of the present invention. Through iterative optimization based on the wEI criterion and the MAE criterion, it can be seen that the SMT package interface contact pressure gradually decreases during the iteration process, and finally the optimal geometric parameters are output as shown in Table 8. Compared with the initial value, the contact pressure decreases by 9.06%, proving the effectiveness of the SMT package reliability optimization method for interface contact pressure proposed in this paper.

[0147] Table 8 Optimal Geometric Parameters for Interface Contact Pressure

[0148] Parameter name <![CDATA[Welding spot height H s / mm]]> <![CDATA[Wetting length L w / mm]]> <![CDATA[Clearance height H g / mm]]> Contact pressure / MPa Initial parameter value 0.72 1.1 0.12 -23.252 Optimal parameter value 0.703 1.288 0.125 -21.145

[0149] An optimization method for packaging reliability oriented to interface contact pressure provided by the present application aims to improve the blindness of the traditional SMT interface contact pressure optimization design, reduce the computational cost and time in the optimization process, and enhance the thermal reliability of SMT packaging. By constructing a Kriging surrogate model and using the wEI (weighted Expected Improvement) filling criterion and the MAE (Mean Absolute Error) secondary filling criterion, while reducing the high-cost simulation calculation calls, the optimization accuracy and efficiency are improved.

[0150] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A method for optimizing the reliability of SMT packaging for interface contact pressure, characterized in that, Including: Determine the geometric parameters, physical property parameters, and design domain of the package according to the package structure; Use the improved Latin hypercube algorithm to sample within the design domain to obtain initial geometric parameter sample points; Based on the principle of minimum energy, construct a parametric characterization model of the package corresponding to the initial geometric parameter sample points; Construct a solder viscoplastic constitutive model, a solder cohesive force model, and a solder temperature cycle load curve; Based on the parametric characterization model of the package, the solder viscoplastic constitutive model, the solder cohesive force model, and the solder temperature cycle load curve, construct a finite element model under temperature cycle load, and determine the convergence stagnation threshold and the number of iterations; Based on the finite element model, perform finite element simulation analysis on the package structure corresponding to the initial geometric parameter sample points to obtain the maximum contact pressure at the package solder interface, and form contact pressure sample points; Based on the initial geometric parameter sample points and the contact pressure sample points, construct a Kriging surrogate model; Based on the wEI filling criterion, by maximizing the value of the current Kriging surrogate model, an updated geometric parameter sample point is obtained; including: The expression of the wEI filling criterion is: ； ； Among them, is the weight of the global search ability, is the current iteration number, is the cumulative probability distribution function of the standard normal distribution, is the probability density function of the standard normal distribution, is the optimal point calculated by the current Kriging surrogate model, is any unknown point within the design domain is the standard deviation of, is any unknown point within the design domain is the response value of, is the maximum number of iterations; The obtained updated geometric parameter sample points , and its calculation formula is: ； Judge whether convergence has stagnated based on the convergence stagnation criterion; if not, continue with the finite element simulation analysis until convergence stagnates; if so, judge whether the termination criterion is satisfied; if so, output the geometric parameters corresponding to the minimum contact pressure; if not, based on the MAE criterion, update the geometric parameter sample points again and continue with the finite element simulation analysis until convergence stagnates; the process of judging whether convergence has stagnated based on the convergence stagnation criterion includes: Judge whether convergence has stagnated according to the convergence stagnation formula, where the expression of the convergence stagnation formula is: ； Among them, is the minimum value function, is the maximum value function, is the current iteration number; if , is the convergence stagnation threshold, , then the convergence stagnates; otherwise, the convergence does not stagnate.

2. The method for optimizing the reliability of SMT packaging for interface contact pressure according to claim 1, characterized in that, The sampling process of the improved Latin hypercube algorithm includes: Divide the range of domain values of the design domain into equal-length intervals, and randomly select a geometric parameter sample point within each interval to form sample sets; Generate a set of sampling sets, and obtain the Euclidean distances between the geometric parameter sample points in each group of the sampling sets; wherein, the distance between adjacent geometric parameter sample points is defined as the Euclidean distance; Compare the minimum Euclidean distances in each of the sampling sets, and select the sampling set corresponding to the maximum Euclidean distance among them as the set of the initial geometric parameter sample points .

3. The method for optimizing the reliability of SMT packaging for interface contact pressure according to claim 1, characterized in that, The parametric characterization model of the package is constructed under the constraints of solder volume, solder gravitational potential energy, and solder surface potential energy.

4. The method for optimizing the reliability of SMT packaging for interface contact pressure according to claim 1, characterized in that, The solder joint viscoplastic constitutive model is constructed based on the solder joint viscoplastic constitutive model parameters, and the solder joint viscoplastic constitutive model parameters include: initial deformation impedance , pre-exponential factor , stress multiplier , strain rate sensitivity index , strain hardening constant , deformation impedance coefficient , strain rate sensitivity index , strain hardening sensitivity coefficient and activation energy ; The solder joint cohesive force model is constructed based on solder joint cohesive force parameters, and the solder joint cohesive force parameters include: the normal critical energy release rate is , the tangential critical energy release rate , the normal maximum contact stress , the tangential maximum contact stress ; The temperature cycle load curve is constructed based on temperature cycle load parameters, and the temperature cycle load parameters include: the thermal cycle temperature range applied to the package structure ~ , the reference temperature under zero stress strain , a single cycle period , including minutes of temperature rise, minutes of temperature drop, and each minutes of high and low temperature heat preservation, the number of thermal cycle load periods cycles, and the load sub-steps of a single cycle period 5. The method for optimizing the reliability of SMT packaging for interface contact pressure according to claim 1, characterized in that, The expression of the Kriging surrogate model is: ; Among them, is the response value of any unknown point in the design domain , is the variance of any unknown point in the design domain , is the contact pressure sample point; is the mean of the Gaussian process, is the initial geometric parameter sample point and the covariance matrix between any unknown point in the design domain, is the covariance matrix between each initial geometric parameter sample point .

6. The method for optimizing the reliability of SMT packaging for interface contact pressure according to claim 1, characterized in that, The process of judging whether the termination criterion is satisfied includes: Determine the current iteration number whether it is greater than the maximum number of iterations; if so, output the geometric parameters corresponding to the minimum contact pressure; where the maximum number of iterations .

7. The SMT package reliability optimization method for interface contact pressure according to claim 1, characterized in that, The process of updating the geometric parameter sample points again based on the MAE criterion includes: Increase the sample points of geometric parameters , and determine the MAE criterion model, whose expression is: ； Among them, is the variance of any unknown point within the design domain.

Images