A slotted contact separation force calculation method using a fast calculation model

Abstract

The application discloses a slotted contact separation force calculation method using a fast calculation model and relates to the technical field of mechanical property calculation of electric connectors. Input parameters and given tolerances of each parameter are determined; parameter combinations are randomly generated in the given tolerance range through a Monte Carlo algorithm; a finite element model of the slotted contact is established, and an output separation force value is calculated; a support vector regression algorithm is used to compile a fast calculation model; training samples are imported for solution, and the trained fast calculation model is obtained; verification samples are imported and compared with finite element simulation values to verify the accuracy, and if the accuracy meets the requirements, file packaging is performed, otherwise, the number of parameter combinations is supplemented. After data expansion of the input parameters, the separation force results obtained through finite element simulation are used as references, a fast calculation model is compiled based on the support vector regression algorithm and is trained and verified, and the separation force of different slotted contacts of the same batch of electric connectors can be quickly obtained, so that the calculation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mechanical property calculation of electrical connectors, and particularly relates to a slot contact separation force calculation method using a fast calculation model. BACKGROUND

[0002] The slot contact is mainly composed of a slot hole, a pin, a sheath and the like, and is generally designed as a circular slot contact. In order to ensure reliable contact between the slot hole and the pin, in the current production and assembly process, a closing process is needed to cause a certain plastic deformation of the front end of the slot hole inward. When the pin is inserted into the slot hole, the contact normal force and multi-point contact are formed between the pin hole, which has the characteristics of reliable contact and relatively simple processing and assembly.

[0003] The electrical contact of the electrical connector is realized by a contact pair (pin and socket), which is respectively installed in the plug and the socket of the electrical connector. The contact pair is the key to complete the reception and transmission of electrical signals and network signals, and the mechanical performance thereof plays an important role in whether the signals can be stably and reliably received. When the pin and the socket are combined, the spring is forced to elastically deform to generate a certain strength of contact pressure. This action is also called the plugging process, and the separation force is usually used to indirectly reflect whether the electrical connector is in reliable contact. Therefore, the separation force of the contact is one of the important factors affecting the performance of the electrical connector, and is directly related to the performance and use of the electrical connector.

[0004] There are many factors affecting the separation force of the slot contact, such as the Young's modulus, yield strength of the contact material, the inner and outer diameters of the socket, the depth and width of the slot and the like. The traditional separation force calculation method mainly adopts the finite element software simulation method, which cannot quickly calculate the separation force of the contact, resulting in slow calculation speed of the separation force of the contact of the electrical connector. Therefore, how to directly calculate the separation force between the pin and the socket through the related parameters of the contact has become a problem to be solved.

[0005] Therefore, the present application considers the influence of material parameters, size parameters and process parameters on the separation force of the contact of the electrical connector, establishes a fast calculation model of the separation force between the contact pairs of the electrical connector under different parameters through finite element simulation and machine learning algorithm based on support vector regression, and realizes fast calculation of the separation force of the contact of the electrical connector. SUMMARY

[0006] In order to solve the problems in the background art, the present application provides a slot contact separation force calculation method using a fast calculation model. After data expansion of the input parameters, the separation force result of the finite element simulation is taken as a reference, a fast calculation model is written based on the support vector regression algorithm and is trained and verified, and the separation force of different slot contacts of the same batch of electrical connectors can be quickly obtained, thereby improving the calculation efficiency.

[0007] To achieve the above object, the application adopts the following technical scheme: a slot contact separation force calculation method using a fast calculation model, comprising the following steps:

[0008] Step one: according to the process flow and design drawings of the electrical connector, determine the material parameters, size parameters and process parameters of the slot contact as input parameters, and determine the given tolerances of each parameter, the material parameters include Young's modulus, yield strength, tensile strength and elongation, the size parameters include the inner diameter of the jack, the outer diameter of the jack, the slot length and the slot width, the process parameters include the clamp stroke and the placement height, and the separation force value is taken as the output parameter;

[0009] Step two: for the input parameters of step one, randomly generate N groups of parameter combinations within the given tolerance range by the Monte Carlo algorithm, and store them in the Excel file;

[0010] Step three: use ABAQUS software to establish a finite element model of the slot contact, perform finite element calculation on the parameter combinations obtained in step two through model simulation, output the separation force values corresponding to each parameter combination, and store them in the Excel file;

[0011] Step four: use the support vector regression algorithm to write a calculation program for the slot contact separation force in MATLAB as a fast calculation model, the specific operation is as follows:

[0012] The relationship between the input parameters x and the predicted value f(x) of the fast calculation model is established, which is expressed as follows:

[0013] f(x)=w T x+b (1)

[0014] In the formula, w is the normal vector of the hyperplane, and b is the displacement term;

[0015] The error function is expressed as:

[0016]

[0017] In the formula, y is the simulation value of the finite element model, and ε is the bandwidth of the insensitive band;

[0018] The N groups of generated parameter combinations are taken as samples, and the samples are divided into training samples and verification samples according to the requirements;

[0019] The optimization problem of the support vector regression algorithm is expressed as:

[0020]

[0021] In the formula, C is the penalty parameter, i represents the ith sample, l represents the number of training samples, and ξ irelaxation variable, distance between the metric true value and the sensitive band, y i simulation value of the i th sample, x i input parameter of the i th sample;

[0022] introducing Lagrange multiplier, according to Karush-Kuhn-Tucker condition and its supplementary condition, we have:

[0023]

[0024] introducing kernel function to improve the calculation efficiency, the kernel function is expressed as:

[0025]

[0026] Therefore, formula (5) is rewritten as:

[0027]

[0028] In the formula, a i Lagrange multiplier, conjugate complex of a i low-dimensional to high-dimensional mapping relationship;

[0029] Step five: introducing the training sample into MATLAB and solving by support vector regression algorithm to obtain the trained fast calculation model;

[0030] Step six: introducing the verification sample into the trained fast calculation model to verify the accuracy, comparing the output predicted value with the simulation value of the corresponding finite element model, if the overall average error is less than the set threshold, then the fast calculation model is file packaged, if the overall average error is not less than the set threshold, then return to step two to supplement the number of parameter combinations until the requirement is met.

[0031] Further, in the step two, the number of first generated parameter combinations N≥400.

[0032] Further, in the step four, introducing Gaussian kernel function as the kernel function, the expression is as follows:

[0033]

[0034] In the formula, σ>0, is the bandwidth of Gaussian kernel function.

[0035] Further, in the step six, the threshold is set to 10%.

[0036] ​Compared with the prior art, the beneficial effects of the present application are: the present application expands the data of the input parameters by the Monte Carlo algorithm, and uses ABAQUS to perform finite element simulation analysis and calculation of the corresponding separation force value, and a rapid calculation model is written in MATLAB, and according to the support vector regression algorithm, the prediction value of the separation force of the slotted contact of the electrical connector under the condition of given material parameters, size parameters and process parameters is obtained, and the parameter combination after the expansion of the input parameters is trained and verified, so that the subsequent calculation of the separation force of the slotted contact of the same specification does not need to be simulated and analyzed again, but can be directly obtained through the rapid calculation model, solving the problem that the traditional finite element software simulation method is difficult to quickly calculate the separation force of the contact, and effectively improving the calculation efficiency of the separation force. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 is the implementation flowchart of the calculation method of the present application;

[0038] Figure 2 is an error function diagram of the support vector regression algorithm in the calculation method of the present application;

[0039] Figure 3 is a part of the parameter setting diagram in the finite element calculation process of the selected parameter combination in the embodiment. DETAILED DESCRIPTION

[0040] The technical solutions in the present application will be described clearly and completely in the embodiments of the present application combined with the drawings, obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments, based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor are within the protection scope of the present application.

[0041] As shown in Figures 1-2 , a slotted contact separation force calculation method using a rapid calculation model, the implementation process thereof is combined with Figure 1 , including the following steps:

[0042] Step 1: according to the process flow and design drawings of the electrical connector, the material parameters, size parameters and process parameters of the slotted contact are determined as input parameters, and the given tolerances of each parameter are determined, the material parameters include Young's modulus, yield strength, tensile strength and elongation, the size parameters include the inner diameter of the jack, the outer diameter of the jack, the slot length and the slot width, the process parameters include the clamp stroke and the placing height, and the separation force value is taken as the output parameter;

[0043] Step two: Because it is difficult to keep all parameters of the slotted contact unchanged in actual production, N sets of parameter combinations are randomly generated in a given tolerance range by the Monte Carlo algorithm for the input parameters of step one, the first generation number N≥400, which is used to simulate the material parameters, size parameters and process parameters of different slotted contacts in actual production, and stored in an Excel file;

[0044] Step three: a finite element model of the slotted contact is established using ABAQUS software, the parameter combinations obtained in step two are calculated by the model simulation, the separation force values corresponding to each parameter combination are output, and stored in an Excel file;

[0045] Step four: a calculation program for the separation force of the slotted contact is written in MATLAB using the support vector regression algorithm (SVR) as a fast calculation model, which provides input and output windows for input parameters and separation force values, and the specific operation is as follows:

[0046] The prediction of the separation force belongs to the solution of the regression problem, and the relationship between the n-dimensional input parameters (a total of 10 input parameters, n=10) x and the predicted value f(x) of the fast calculation model is established, which is expressed as follows:

[0047] f(x)=w T x+b (1)

[0048] In the formula, w is the normal vector of the hyperplane, and b is the displacement term (bias).

[0049] The error function is expressed as:

[0050]

[0051] As shown in Figure 2 , formula (2) is represented as a near-orbital area in the figure, which is called an insensitive band, and no penalty is given to the data in this area, in which y is the simulation value of the finite element model, and ε is the bandwidth of the insensitive band.

[0052] The generated N sets of parameter combinations are used as samples, and the samples are divided into training samples and validation samples according to the requirements.

[0053] The optimization problem of the support vector regression algorithm can be expressed as:

[0054]

[0055] In the formula, C is a penalty parameter, i represents the ith sample, l represents the number of training samples, and ξ i is a relaxation variable, y represents the distance between the metric real value and the sensitive band, reduces the requirement for the hyperplane, y i x represents the simulation value of the i th sample i x represents the simulation value of the i th sample

[0056] The Lagrange multiplier is introduced, and according to the KKT condition and its supplementary condition, the following can be obtained:

[0057]

[0058] In the formula, a i is the Lagrange multiplier, is the conjugate complex of a i , and is the mapping relationship from low dimension to high dimension.

[0059] In order to linearize the nonlinear characteristic variable, the kernel function is introduced to project the nonlinear sample space into a high-dimensional linear feature space, so as to reduce the difficulty of nonlinear calculation and improve the calculation efficiency. The kernel function is expressed as follows:

[0060]

[0061] Therefore, formula (5) can be rewritten as:

[0062]

[0063] Common kernel functions include linear kernel function, polynomial kernel function, Gaussian kernel function, Laplace kernel function and Sigmoid kernel function, etc. The convergence domain and learning accuracy of the Gaussian kernel function are better than those of several other kernel functions. Therefore, the Gaussian kernel function is preferred as the kernel function of the present scheme, and the expression is as follows:

[0064]

[0065] In the formula, σ>0 is the bandwidth of the Gaussian kernel function;

[0066] Step five: import the training sample into MATLAB and solve it by support vector regression algorithm to obtain the trained fast calculation model;

[0067] Step six: import the verification sample into the trained fast calculation model for accuracy verification, compare the output predicted value with the corresponding simulation value of the finite element model, if the overall average error is less than the set threshold (which can be set to 10%), then the fast calculation model is packaged into a file for subsequent calculation of the separation force value of the same type and different specifications of the slotted contact, if the overall average error is not less than the set threshold, then return to step two to supplement the number of parameter combinations until the requirement is met.

[0068] Embodiment

[0069] (1) According to the process flow and design drawings of the electrical connector, and taking a circular slotted contact as an example, Young's modulus, yield strength, tensile strength, elongation, inner diameter of the jack, outer diameter of the jack, slot length, slot width, clamp travel, and placement height are determined as input parameters, and the given tolerances of the parameters are determined. The data is shown in Table 1:

[0070] Table 1 Center value and tolerance of input parameters

[0071]

[0072] (2) For the input parameters, 450 sets of parameter combinations are randomly generated within the given tolerance range by the Monte Carlo algorithm, which are used to simulate the Young's modulus, yield strength, tensile strength, elongation, inner diameter of the jack, outer diameter of the jack, slot length, slot width, clamp travel, and placement height of different circular slotted contacts in actual production, and are stored in an Excel file;

[0073] (3) The finite element model of the circular slotted contact is established using ABAQUS software, the parameter combinations obtained in step two are calculated by the model simulation, 450 sets of separation force values are output, and are stored in an Excel file. One set of parameter combinations is shown in Table 2:

[0074] Table 2 Parameter combinations

[0075]

[0076] During the finite element calculation process of the parameter combinations, part a) shows the Young's modulus setting, part b) shows the clamp travel setting, and part c) shows the size parameter setting, including the inner diameter of the jack, the outer diameter of the jack, the slot length, and the slot width; Figure 3

[0077] (4) A calculation program for the separation force of the slotted contact is written in MATLAB using the support vector regression algorithm (SVR) as a fast calculation model. The program provides input and output windows for input parameters and separation force values, respectively. Among them, the generated 450 sets of parameter combinations are divided into 440 sets as training samples and 10 sets as validation samples;

[0078] (5) The 440 sets of training samples are imported into MATLAB, and part of the training samples are shown in Table 3:

[0079] Table 3 Training samples (part)

[0080]

[0081] ​The trained fast calculation model is obtained by solving through a support vector regression algorithm;

[0082] (6) The 10 groups of verification samples are introduced into the trained fast calculation model for precision verification, the predicted values output by the fast calculation model are compared with the simulation values of the corresponding finite element models, and the verification results are shown in Table 4:

[0083] Table 3 verification results

[0084]

[0085] It can be seen that the calculation values obtained through the trained fast calculation model have smaller overall average error (about 4.36%, less than 10%) compared with the simulation values of the corresponding finite element models, and it is considered that the precision of the fast calculation model meets the requirements, no additional parameter combination is needed, an app is established and a file package is made, and the separation force values of the same type of different specifications of the slotted contact can be quickly obtained by introducing the input parameters into the fast calculation model.

[0086] It is apparent to those skilled in the art that the present application is not limited to the details of the foregoing exemplary embodiments, and that the present application can be implemented in other embodiments without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all respects as illustrative and not restrictive, the scope of the present application being defined by the appended claims rather than the foregoing description, and it is intended that all changes that come within the meaning and range of equivalency of the claims are embraced therein. Any reference signs in the claims should not be construed as limiting the claims to the figures in which the reference signs are used.

[0087] In addition, it should be understood that although the present specification is described in terms of embodiments, not every embodiment contains only one independent technical solution, and the description of the specification is only for the sake of clarity, and those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be properly combined to form other embodiments that those skilled in the art can understand.

Claims

1. A method for calculating separation force of a slotted contact using a fast computational model, the method comprising: The method comprises the following steps: ​ Step 1: according to the process flow and design drawings of the electrical connector, determine the material parameters, size parameters and process parameters of the slotted contact as input parameters, and determine the given tolerances of each parameter, the material parameters include Young's modulus, yield strength, tensile strength and elongation, the size parameters include the inner diameter of the jack, the outer diameter of the jack, the slot length and the slot width, the process parameters include the clamp stroke and the placement height, and the separation force value is taken as the output parameter; Step 2: for the input parameters of step 1, randomly generate N groups of parameter combinations within the given tolerance range by the Monte Carlo algorithm, and store them in an Excel file; Step 3: use ABAQUS software to establish a finite element model of the slotted contact, perform finite element calculation on the parameter combinations obtained in step 2 through model simulation, output the separation force values corresponding to each parameter combination, and store them in an Excel file; Step 4: use the support vector regression algorithm to write a calculation program for the separation force of the slotted contact in MATLAB as a fast calculation model, the specific operation is as follows: The relationship between the input parameters x and the predicted value f(x) of the fast calculation model is established and expressed as follows: f(x) = w T x + b (1) In the formula, w is the normal vector of the hyperplane, and b is the displacement term; The error function is expressed as: In the formula, y is the simulation value of the finite element model, and ε is the bandwidth of the insensitive band; The N groups of generated parameter combinations are taken as samples, and the samples are divided into training samples and verification samples according to requirements; The optimization problem of the support vector regression algorithm is expressed as: where C is a penalty parameter, i represents the i-th sample, l represents the number of training samples, and ξ i is a slack variable, represents the distance between the metric true value and the sensitive band, y i represents the simulation value of the i-th sample, x i represents the input parameter of the i-th sample; The Lagrange multiplier is introduced, and according to the Karush-Kuhn-Tucker condition and its supplementary condition, we have: The kernel function is introduced to improve the calculation efficiency, and the kernel function is expressed as: Therefore, formula (5) is rewritten as: In the formula, a i is a Lagrange multiplier, is a i conjugate complex of a, is a low-dimensional to high-dimensional mapping relationship Step 5: import the training samples into MATLAB and solve them by the support vector regression algorithm to obtain the trained fast calculation model; Step 6: import the verification samples into the trained fast calculation model to verify the accuracy, compare the output predicted value with the simulation value of the corresponding finite element model, if the overall average error is less than the set threshold, then the fast calculation model is packaged, if the overall average error is not less than the set threshold, then return to step 2 to supplement the number of parameter combinations until the requirements are met.

2. The method of claim 1, wherein the method is a method of calculating a separation force of a slotted contact using a fast computational model. In step 2, the number of the first generated parameter combinations is N≥400.

3. The method of claim 1, wherein: In step 4, the Gaussian kernel function is selected as the kernel function, and the expression is as follows: In the formula, σ>0 is the bandwidth of the Gaussian kernel function.

4. The method of claim 1, wherein: In step 6, the threshold is set to 10%.

Images