Reinforcing steel bar component deformation prediction method based on welding parameters

Abstract

The invention provides a steel bar component deformation prediction method based on welding parameters, the welding parameters and related data of a steel bar component are obtained, the welding parameters at least comprise primary parameters and secondary parameters, and the related data at least comprise the diameter of a steel bar; determining the size of a welding spot according to the welding parameters and the diameter of the steel bar; determining the three-direction rotational rigidity of the welding spot according to the size of the welding spot; determining the overall rigidity of the reinforcing steel bar component according to the three-way rotational rigidity of the welding spot; calculating the deformation of the reinforcing steel bar component according to the overall rigidity of the reinforcing steel bar component; and if the deformation of the reinforcing steel bar component does not meet the preset requirement, adjusting the welding parameters and recalculating until the deformation of the reinforcing steel bar component meets the preset requirement. According to the method, correlation analysis among welding parameters, welding spot characteristics and steel bar component performance is achieved, a scientific basis is provided for optimization of a steel bar welding process, and improvement of structural reliability and durability of a steel bar component is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of welding parameter prediction, and particularly to a method for predicting the deformation of a steel bar component based on welding parameters. Background Art

[0002] The forming methods of steel bar components can be divided into binding and welding. The stiffness of the bound steel bar components is small, and the distance between steel bars is likely to change during transportation and hoisting, and it is relatively difficult to achieve automation compared with welding. Therefore, most steel bar components are formed by welding process. The welded steel bar components are prefabricated in sections in the factory and then transported to the construction site for installation. They need to experience multiple hoisting operations, and when installed in place, the main steel bars of the upper and lower section steel bar components need to be connected through methods such as tapered sleeves. The overall stiffness of the steel bar components will directly affect the hoisting safety and the difficulty of alignment connection. In addition, when the structure is a special-shaped structure, such as having a certain inclination angle, the overall stiffness of the steel bar components will also directly affect the structural linearity and the thickness of the protective layer. Therefore, the calculation of the deformation of the welded steel bar components is particularly important.

[0003] Currently, when calculating the deformation of welded components, the main steel bars and stirrups are usually rigidly connected. However, in fact, the welded joints are not rigid structures, and the main steel bars and stirrups will rotate relative to each other when stressed. Therefore, essentially the welded joints are semi-rigid structures. The traditional rigid connection method will seriously underestimate the deformation of the welded steel bar components, resulting in problems such as excessive hoisting deformation, difficult connection of the main steel bars, non-compliance of the structural linearity with the design requirements, and unqualified thickness of the protective layer. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method for predicting the deformation of a steel bar component based on welding parameters to solve the problem of calculating the deformation of welded steel bar components.

[0005] To solve the above technical problems, the technical solution adopted by the present invention is: a method for predicting the deformation of a steel bar component based on welding parameters, which obtains welding parameters and relevant data of the steel bar component. The welding parameters at least include main parameters and secondary parameters, and the relevant data at least includes the steel bar diameter; Determine the weld size according to the welding parameters and the steel bar diameter; Determine the three-way rotational stiffness of the weld according to the weld size; Determine the overall stiffness of the steel bar component according to the three-way rotational stiffness of the weld; Calculate the deformation of the steel bar component according to the overall stiffness of the steel bar component; If the deformation of the steel bar component does not meet the preset requirements, adjust the welding parameters and recalculate until the deformation of the steel bar component meets the preset requirements.

[0006] In the preferred solution, determining the weld size according to the welding parameters and the steel bar diameter includes: Classify the welding parameters to obtain the main parameters and secondary parameters, where the main parameters include the welding time , and the secondary parameters include the welding voltage and the welding current ; Obtain the main bar diameter and the stirrup diameter from the steel bar diameter ; ; According to the welding time and the main bar diameter and the stirrup diameter establish the first functional relationship ; Amend the first functional relationship through the welding voltage and the welding current , introducing correction coefficients , , to obtain the second functional relationship ; Calculate the weld size according to the second functional relationship .

[0007] In the preferred solution, use a scanning electron microscope and atom probe tomography to analyze the microstructure of the weld, and establish a physical model in combination with materials science theory to assist in determining the correction coefficients , ; At the same time, use the support vector regression algorithm to learn and train a large amount of welding process test data to optimize the first functional relationship .

[0008] In the preferred solution, the weld size determines the three-way rotational stiffness of the weld, including: Obtain the definition of the three-way rotational stiffness of the weld, where the three-way rotational stiffness of the weld includes the rotational stiffness of the stirrup in the plane where the main bar and the stirrup are located , the rotational stiffness of the main bar with the stirrup as the rotation axis and the rotational stiffness of the stirrup with the main bar as the rotation axis ; For the weld size, use a convolutional neural network in deep learning, combined with the microstructure characteristic parameters of the material, to establish the third functional relationship between the weld size and the three-way rotational stiffness of the weld , ; Calculate the three-way rotational stiffness of the weld through the third functional relationship .

[0009] In the preferred solution, a finite element analysis software is used to finely model the solder joints. Combining with the mechanical solution theoretical model of the three-way rotational stiffness of the solder joints, considering the elastic-plastic deformation of materials, the distribution of residual stresses, and micro-defect factors, the stress and deformation conditions of the solder joints under different working conditions are simulated to assist in determining the function .

[0010] In the preferred solution, the three-way rotational stiffness of the solder joints determines the overall stiffness of the steel bar component, including: Obtain the geometric parameters and material properties of the steel bar component. The geometric parameters include the length of the main steel bar and the length of the stirrup, and the material properties include the elastic modulus of the steel bar; establish a mechanical solution model for the three-way rotational stiffness of the solder joints; through the mechanical solution model, combined with the length of the main steel bar, the length of the stirrup, and the elastic modulus of the steel bar, calculate the overall stiffness of the steel bar component; Using the multi-physics field coupling technology, considering the influence of the multi-field interactions of heat-mechanics and force-electricity on the mechanical solution model, optimize the calculation of the overall stiffness.

[0011] In the preferred solution, the calculation of the deformation of the steel bar component by the overall stiffness of the steel bar component includes: Obtain the external load conditions of the steel bar component, and the external load conditions include the acting force applied to the steel bar component ; According to the overall stiffness of the steel bar component and the acting force , establish a deformation calculation model, such as the vertical deformation amount ; Calculate the vertical deformation amount of the steel bar component through the deformation calculation model ; Judge whether the deformation of the steel bar component meets the preset requirements according to the vertical deformation amount .

[0012] In the preferred solution, if the deformation of the steel bar component does not meet the preset requirements, then adjust the welding parameters and recalculate, including: Obtain the preset deformation threshold ; If the deformation of the steel bar component exceeds the deformation threshold , then according to the inverse relationship between the deformation of the steel bar component and the welding parameters, adopt the particle swarm optimization algorithm or the adaptive genetic algorithm, with the goal of minimizing deformation and optimizing welding quality, to determine the adjusted welding parameters; Recalculate the solder joint size, the three-way rotational stiffness of the solder joint, and the overall stiffness of the steel bar component through the adjusted welding parameters until the deformation of the steel bar component is less than or equal to the deformation threshold .

[0013] In the preferred embodiment, the dot size determines the three - dimensional rotational stiffness of the solder joint, including: Obtain the measurement data of the test device, where the test device is used to load the steel bar component and measure the deformation; For the solder joint size, establish a fourth functional relationship through the measurement data , ; Verify the calculation result of the three - dimensional rotational stiffness of the solder joint through the fourth functional relationship, and calculate the calculation error , ; If the calculation result is inconsistent with the measurement data, that is when it exceeds the allowable error range, then adjust the third functional relationship and recalculate the three - dimensional rotational stiffness of the solder joint; In the preferred embodiment, a non - contact measurement technique is used to obtain more accurate measurement data.

[0014] The present invention provides a method for predicting the deformation of a steel bar component based on welding parameters. This method establishes a relationship model among the solder joint size, the three - dimensional rotational stiffness of the solder joint, and the overall stiffness of the steel bar component by obtaining welding parameters and the diameter of the steel bar. Calculate the deformation of the steel bar component according to the external load conditions and compare it with the preset requirements. If the requirements are not met, automatically adjust the welding parameters and recalculate until the deformation requirements are met. The present invention also introduces the measurement data of the test device to verify and optimize the calculation result of the three - dimensional rotational stiffness of the solder joint. Through this method, the welding parameters can be effectively optimized, and the overall performance and quality of the steel bar component can be improved. The present invention realizes the correlation analysis among the welding parameters, the solder joint characteristics, and the performance of the steel bar component, provides a scientific basis for optimizing the steel bar welding process, and helps to improve the structural reliability and durability of the steel bar component. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The following further describes the present invention with reference to the drawings and embodiments: Figure 1 is the deformation prediction process of the steel bar component based on welding parameters of the present invention; Figure 2 is the solder joint size diagram of the present invention; Figure 3 is the schematic diagram of the three - dimensional rotational stiffness of the solder joint of the present invention; Figure 4 is the mechanical solution theoretical model of the three - dimensional rotational stiffness of the solder joint of the present invention; Figure 5 is the schematic diagram of the test device for solving the rotational stiffness Rx of the present invention; Figure 6 is the test schematic diagram for solving the rotational stiffness Ry and Rz of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] Example 1 As Figure 1-6 shown, a method for predicting the deformation of a steel bar component based on welding parameters, which obtains the welding parameters and the relevant data of the steel bar component, the welding parameters at least include main parameters and secondary parameters, and the relevant data at least includes the steel bar diameter; Determine the weld size according to the welding parameters and the steel bar diameter; Determine the three-way rotational stiffness of the weld according to the weld size; Determine the overall stiffness of the steel bar component according to the three-way rotational stiffness of the weld; Calculate the deformation of the steel bar component according to the overall stiffness of the steel bar component; If the deformation of the steel bar component does not meet the preset requirements, adjust the welding parameters and recalculate until the deformation of the steel bar component meets the preset requirements.

[0017] In the preferred solution, determining the weld size according to the welding parameters and the steel bar diameter includes: Classify the welding parameters to obtain the main parameters and secondary parameters, where the main parameters include the welding time , and the secondary parameters include the welding voltage and the welding current ; Obtain the main bar diameter and the stirrup diameter from the steel bar diameter; Establish a first functional relationship based on the welding time with the main bar diameter and the stirrup diameter ; Modify the first functional relationship through the welding voltage and the welding current , and introduce correction factors , to obtain a second functional relationship ; Calculate the weld size according to the second functional relationship.

[0018] In the preferred solution, use a scanning electron microscope and atom probe tomography to analyze the microstructure of the weld, and establish a physical model in combination with materials science theory to assist in determining the correction factors , ; at the same time, use the support vector regression algorithm to learn and train a large amount of welding process test data to optimize the first functional relationship .

[0019] In the preferred embodiment, the solder joint size determines the three - dimensional rotational stiffness of the solder joint, including: Obtain the definition of the three - dimensional rotational stiffness of the solder joint, where the three - dimensional rotational stiffness of the solder joint includes the rotational stiffness of the stirrup within the plane of the main reinforcement and the stirrup , the rotational stiffness of the main reinforcement with the stirrup as the axis of rotation and the rotational stiffness of the stirrup with the main reinforcement as the axis of rotation ; For the solder joint size, use the convolutional neural network in deep learning, combined with the material microstructure characteristic parameters, to establish the third functional relationship between the solder joint size and the three - dimensional rotational stiffness of the solder joint ; Calculate the three - dimensional rotational stiffness of the solder joint through the third functional relationship .

[0020] In the preferred embodiment, use finite - element analysis software to finely model the solder joint. Combining with the mechanical solution theoretical model of the three - dimensional rotational stiffness of the solder joint, considering material elastic - plastic deformation, residual stress distribution, and micro - defect factors, simulate the stress - deformation conditions of the solder joint under different working conditions to assist in determining the function .

[0021] In the preferred embodiment, the three - dimensional rotational stiffness of the solder joint determines the overall stiffness of the steel - bar component, including: Obtain the geometric parameters and material properties of the steel - bar component. The geometric parameters include the length of the main reinforcement and the length of the stirrup, and the material properties include the elastic modulus of the steel bar. Establish a mechanical solution model for the three - dimensional rotational stiffness of the solder joint. Through the mechanical solution model, combined with the length of the main reinforcement, the length of the stirrup, and the elastic modulus of the steel bar, calculate the overall stiffness of the steel - bar component; Use the multi - physical - field coupling technology to consider the influence of thermal - force and force - electricity multi - field interactions on the mechanical solution model and optimize the calculation of the overall stiffness.

[0022] In the preferred embodiment, the overall stiffness of the steel - bar component calculates the deformation of the steel - bar component, including: Obtain the external load conditions of the steel - bar component, where the external load conditions include the force applied to the steel - bar component ; According to the overall stiffness of the steel - bar component and the force , establish a deformation calculation model, such as the vertical deformation amount ; Calculate the vertical deformation amount of the steel - bar component through the deformation calculation model ; According to the vertical deformation amount judge whether the deformation of the steel - bar component meets the preset requirements.

[0023] In the preferred solution, if the deformation of the steel bar component does not meet the preset requirements, the welding parameters are adjusted and recalculated, including: Obtain the preset deformation threshold ; If the deformation of the steel bar component exceeds the deformation threshold , then according to the inverse relationship between the deformation of the steel bar component and the welding parameters, the particle swarm optimization algorithm or the adaptive genetic algorithm is used to determine the adjusted welding parameters with the goal of minimizing deformation and optimizing welding quality; Recalculate the weld size, the three-way rotational stiffness of the weld point, and the overall stiffness of the steel bar component through the adjusted welding parameters until the deformation of the steel bar component is less than or equal to the deformation threshold .

[0024] In the preferred solution, the weld point size determines the three-way rotational stiffness of the weld point, including: Obtain the measurement data of the test device, which is used to load the steel bar component and measure the deformation; For the weld point size, establish the fourth functional relationship through the measurement data , ; Verify the calculation result of the three-way rotational stiffness of the weld point through the fourth functional relationship, and calculate the error , ; If the calculation result is inconsistent with the measurement data, that is when it exceeds the allowable error range, adjust the third functional relationship and recalculate the three-way rotational stiffness of the weld point; In the preferred solution, non-contact measurement technology is used to obtain more accurate measurement data.

[0025] Example 2 Further illustrated in combination with Example 1. As shown in the structure Figure 1-6 , first classify the welding parameters, and regard the welding parameters that have a great influence on the weld point size as the main parameters, and the parameters with less influence as the secondary parameters. Then, for various steel bar diameters (main bars d 1 , stirrups d 2 ), establish the functional relationship between the main welding parameter (such as welding time T ) and the weld point size: D = f ( T , d 1 , d2 ), and correct this functional relationship through welding secondary parameters (such as welding voltage U , welding current A ): D = k U k A f ( T , d 1 , d 2 ). Subsequently, establish the functional relationship between the solder joint size and the three-way rotational stiffness of the solder joint: R i = f ( D ), i = x , y , z , and calculate the overall stiffness of the welded steel bar component according to the three-way rotational stiffness of the solder joint. Finally, calculate the deformation of the steel bar component according to the overall stiffness of the welded steel bar component. When the deformation does not meet the design requirements, change the welding parameters until the deformation of the steel bar component meets the design and construction requirements, Figure 1 as shown

[0026] The welding parameters include welding time T , welding voltage U , welding current A , etc.

[0027] The functional relationship between the main welding parameter and the solder joint size and the correction of this functional relationship by the welding secondary parameters can both be obtained through process tests, that is, measure the solder joint size under the established welding parameters and establish this functional relationship according to statistical theory.

[0028] The three-way rotational stiffness of the solder joint is respectively: the rotational stiffness R x of the stirrup in the plane where the main bar and the stirrup are located, the rotational stiffness R y of the main bar rotating around the stirrup as the axis, and the rotational stiffness R z of the stirrup rotating around the main bar as the axis, Figure 3 as shown

[0029] The three-way rotational stiffness of the solder joint can establish a mechanical solution theoretical model, Figure 4 as shown

[0030] For R x the mechanical solution model, the two ends of the main bar are hinged and constrained, and a force F is applied to the end of the stirrup, and the vertical deformation of the loading end of the stirrup is:

[0031] In the formula, is the deformation amount of the loading end caused by the bending of the stirrup itself; is the deformation amount of the stirrup loading end caused by the bending of the main reinforcement; is the deformation amount of the stirrup loading end caused by the in-plane rotation of the stirrup; L 1 and L 2 are the lengths of the main reinforcement and the stirrup respectively; I 1 and I 2 are the sectional bending moment of inertia of the main reinforcement and the stirrup respectively; E is the elastic modulus of the steel bar.

[0032] For R y mechanical solution model, the two ends of the stirrup are fixed and constrained, and a force F is applied to the end of the main reinforcement. The vertical deformation of the loading end of the main reinforcement is:

[0033] In the formula, is the deformation amount of the loading end caused by the bending of the main reinforcement itself; is the deformation amount of the loading end of the main reinforcement caused by the bending of the stirrup; is the deformation amount of the loading end of the main reinforcement caused by the torsion of the stirrup; is the deformation amount of the loading end of the main reinforcement caused by the rotation of the main reinforcement around the stirrup; G is the torsional modulus of the steel bar; is the sectional torsional moment of inertia of the stirrup.

[0034] Similarly, for R z mechanical solution model, the vertical deformation of the loading end of the stirrup is:

[0035] In the formula, is the deformation amount of the loading end caused by the bending of the stirrup itself; is the deformation amount of the stirrup loading end caused by the bending of the main reinforcement; is the deformation amount of the stirrup loading end caused by the torsion of the main reinforcement; is the deformation amount of the stirrup loading end caused by the rotation of the stirrup around the main reinforcement; is the sectional torsional moment of inertia of the main reinforcement.

[0036] The said Rx The theoretical model for mechanical solution can solve the stiffness through Test Device 1, Figure 5 as shown.

[0037] As described R y and R z the theoretical model for mechanical solution can solve the stiffness through Test Device 2, Figure 6 as shown.

[0038] The functional relationship between the solder joint size and the three - dimensional rotational stiffness of the solder joint can be obtained through statistical theory.

[0039] The welding parameters can predict the deformation of the welded steel bar parts. Conversely, the welding parameters can also be optimized through the deformation of the welded steel bar parts.

[0040] Example 3 Further illustrated in combination with Example 1, as Figure 2 shown, classify the welding parameters to obtain the main parameters and secondary parameters, where the main parameters include the welding time , and the secondary parameters include the welding voltage and the welding current ; For the steel bar diameter, obtain the main bar diameter and the stirrup diameter ; Combined with Figure 2 the information related to the solder joint size, according to the welding time and the main bar diameter and the stirrup diameter establish the first functional relationship ; Through the welding voltage and the welding current correct the first functional relationship, introduce correction coefficients , to obtain the second functional relationship ; Calculate the solder joint size according to the second functional relationship .

[0041] Use scanning electron microscopy (SEM) and atom probe tomography (APT) to analyze the microstructure of the solder joint, establish a physical model in combination with material science theory to assist in determining the correction coefficients , ; at the same time, use the support vector regression algorithm to learn and train a large amount of welding process test data to optimize the first functional relationship .

[0042] During the process of determining the solder joint size, establish the welding time 、the diameter of the main reinforcement and the diameter of the stirrup The first functional relationship between them is one of the key steps. However, this functional relationship established only through simple theoretical derivation may not accurately reflect the actual situation. Therefore, it is necessary to optimize it using the Support Vector Regression (SVR) algorithm. The specific method and algorithm principle are as follows: Data collection: Collect a large amount of welding process test data under different combinations of welding time 、the diameter of the main reinforcement and the diameter of the stirrup . Each data sample contains a set of inputs ( 、 、 ) and the corresponding output (the solder joint size measured actually ). These data cover various possible value ranges to ensure that the algorithm can learn comprehensive rules.

[0043] Data preprocessing: Preprocess the collected data, including operations such as data cleaning and normalization. Data cleaning is used to remove noise, outliers, and duplicate data in the data to ensure data quality. Normalization maps the input data in different ranges to the same interval, such as $[0, 1]$ or , which can speed up the model training speed, improve the convergence and stability of the model, and avoid difficulties in model training caused by too large differences in data magnitudes.

[0044] Principle of SVR algorithm: Support Vector Regression is a regression algorithm developed based on Support Vector Machine (SVM). Its basic idea is to find an optimal regression hyperplane in the feature space so that the distances from most sample points to this hyperplane are within a certain error range. For the linearly separable case, by finding appropriate weight vectors and biases, a linear regression model (in the current problem, represents the input vector, including 、 、 , represents the solder joint size ) is constructed. For non - linear problems, the original data is mapped to a high - dimensional feature space through a kernel function to make it linearly separable in the high - dimensional space. Common kernel functions include linear kernel, polynomial kernel, radial basis kernel (RBF), etc. In this application, select an appropriate kernel function to map the welding parameters and steel bar diameter data to a high - dimensional space, thereby constructing a more complex non - linear regression model.

[0045] Model training: Divide the preprocessed data into a training set and a test set, generally in a ratio of 70% - 30% or 80% - 20%. Use the training set to train the SVR model. By adjusting the model parameters, such as the penalty parameter , kernel function parameter, etc., to minimize the prediction error of the model on the training set. During the training process, the SVR algorithm will continuously adjust the parameters of the regression hyperplane according to the input welding parameters and steel bar diameter data to better fit the actual weld size data.

[0046] Model evaluation: Use the test set to evaluate the trained SVR model. Commonly used evaluation metrics include mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), etc. These metrics can measure the prediction accuracy and generalization ability of the model. If the model performs poorly on the test set, such as having a large error, it is necessary to adjust the model parameters or reselect the kernel function and train and evaluate again until the model reaches satisfactory performance.

[0047] Function relationship optimization: After training and evaluation, obtain an SVR model with good performance. Apply the rules learned by this model to the first function relationship to optimize it. The optimized function relationship can more accurately reflect the relationship between welding time , main bar diameter and stirrup diameter and the weld size , thereby improving the accuracy of weld size calculation and providing a more reliable basis for subsequent prediction of steel bar component deformation.

[0048] Example 4 Further illustrate in combination with Example 1. Combine Figure 3 with the schematic diagram of the three - dimensional rotational stiffness of the weld to obtain the definition of the three - dimensional rotational stiffness of the weld. The three - dimensional rotational stiffness of the weld includes the rotational stiffness of the stirrup in the plane where the main bar and the stirrup are located , the rotational stiffness of the main bar with the stirrup as the rotation axis and the rotational stiffness of the stirrup with the main bar as the rotation axis ; For the weld size, use the convolutional neural network (CNN) in deep learning, combined with the material microstructure characteristic parameters, to establish the third function relationship between the weld size and the three - dimensional rotational stiffness of the weld , ; Calculate the three - dimensional rotational stiffness of the weld through the third function relationship .

[0049] Using finite element analysis software (ANSYS, ABAQUS) to finely model the solder joints, combined with Figure 4 the mechanical solution theoretical model of the three-way rotational stiffness of the solder joints, considering factors such as material elastic-plastic deformation, residual stress distribution, and micro-defects, simulating the stress and deformation conditions of the solder joints under different working conditions, and assisting in determining the function .

[0050] Obtaining the definition of the three-way rotational stiffness of the solder joints: Combining Figure 3 the schematic diagram of the three-way rotational stiffness of the solder joints to clarify the definition of the three-way rotational stiffness of the solder joints.

[0051] The rotational stiffness of the stirrup in the plane where the main reinforcement and the stirrup are located : It describes the ability to resist rotation when the stirrup rotates in the plane where the main reinforcement and the stirrup are located together. For example, when the steel bar component is stressed, the stirrup may have a tendency to rotate in this plane, which measures the ease of this rotation.

[0052] The rotational stiffness of the main reinforcement with the stirrup as the rotation axis : This stiffness represents the characteristic of the main reinforcement to resist rotation when rotating around the stirrup as the rotation axis. In an actual structure, when subjected to a specific external force, the main reinforcement may rotate around the stirrup as the axis, which reflects the hindering effect on this rotation.

[0053] The rotational stiffness of the stirrup with the main reinforcement as the rotation axis : It refers to the ability of the stirrup to resist rotation when rotating around the main reinforcement as the rotation axis. In different stress scenarios, the stirrup will have a tendency to rotate around the main reinforcement as the axis, which is used to quantify the degree of resistance to this rotation.

[0054] Using a convolutional neural network (CNN) to establish a functional relationship Data preparation: Collect a large number of data samples containing solder joint sizes and the corresponding three-way rotational stiffness of the solder joints ( ). At the same time, obtain the material microstructure characteristic parameters, such as grain size, dislocation density, etc., and organize these data into a format suitable for CNN input. For example, normalize the solder joint size and material microstructure characteristic parameters and convert them into numerical feature vectors.

[0055] Construction of CNN Model: CNN is a deep learning model designed specifically for processing data with grid structures such as images. It mainly consists of convolutional layers, pooling layers, and fully connected layers. In this application, although the data being processed is not traditional image data, the solder joint size and material microstructure characteristic parameters can be regarded as a special "data image". The convolutional layer performs convolution operations by sliding the convolution kernel over the data, automatically extracting local features in the data. For example, the convolution kernel can learn certain local variation patterns of the solder joint size and the associated features between material microstructure characteristics. The pooling layer is used to reduce the dimensionality of the data while retaining the main features and preventing overfitting. For instance, the max pooling operation can select the maximum value in the output of the convolutional layer as the input for the next layer. Finally, the fully connected layer synthesizes the features extracted by the previous layers and outputs the predicted three-way rotational stiffness of the solder joint. .

[0056] Model Training: The prepared data is divided into a training set, a validation set, and a test set. The CNN model is trained using the training set, and the parameters of the model (such as the weights and biases of the convolution kernel) are adjusted through the backpropagation algorithm to minimize the error (such as the mean squared error MSE) between the model prediction and the actual value. During the training process, the validation set is used to monitor the performance of the model to avoid overfitting. When the performance of the model on the validation set no longer improves, the training is stopped.

[0057] Establishing a Functional Relationship: After training, the CNN model learns the complex non-linear relationship between the solder joint size, material microstructure characteristic parameters, and the three-way rotational stiffness of the solder joint, thus establishing a third functional relationship (here represents the complex function mapping relationship learned by CNN, which includes the influence of material microstructure characteristic parameters on the relationship).

[0058] Using Finite Element Analysis Software to Assist in Determining the Function

[0059] Model Establishment: Using finite element analysis software (such as ANSYS, ABAQUS), based on Figure 4 the mechanical solution theoretical model of the three-way rotational stiffness of the solder joint, a fine model of the solder joint is built. During the modeling process, the material properties of the solder joint and the steel bar are accurately set, including elastic modulus, Poisson's ratio, etc. At the same time, considering the elastoplastic deformation characteristics of the material, a suitable constitutive model is selected to describe the elastoplastic behavior of the material under stress. For example, for metal materials, a bilinear kinematic hardening model can be selected. In addition, factors such as the residual stress distribution and micro-defects should also be considered, and the initial stress field is added to the model and the geometry and location of the simulated micro-defects are realized.

[0060] Working condition setting and simulation analysis: Different working conditions are set to simulate the deformation of solder joints under various stress conditions. For example, loads such as forces and torques with different directions and magnitudes are applied, and the displacement, strain distribution, and rotation of the solder joints are observed. Through simulation, the numerical results of the three-way rotational stiffness of the solder joints can be obtained considering factors such as material elastic-plastic deformation, residual stress, and microdefects at different solder joint sizes.

[0061] Auxiliary determination of functional relationship: Compare and analyze the results obtained from finite element simulation with those obtained from CNN model training. According to the relationship between the solder joint size and rotational stiffness in the simulation results, adjust and optimize the function established by the CNN model. For example, if it is found that within a certain solder joint size range in the finite element simulation, the change trend of the rotational stiffness is different from that predicted by the CNN model, the structure or parameters of the CNN model can be appropriately adjusted to make it more in line with the actual physical laws, thereby assisting in determining a more accurate function for accurately calculating the three-way rotational stiffness of the solder joint .

[0062] Example 5 Further illustrated in combination with Example 1, the determination of the overall stiffness of the steel bar component according to the three-way rotational stiffness of the solder joint includes: Obtain the geometric parameters and material properties of the steel bar component, where the geometric parameters include the main bar length and the stirrup length , and the material properties include the elastic modulus of the steel bar ; For the three-way rotational stiffness of the solder joint, establish a mechanical solution model. For the mechanical solution model, in combination with Figure 5 rotational stiffness solve the test device, considering the hinged constraints at both ends of the main bar, apply a force at the end of the stirrup, and the vertical deformation at the loading end of the stirrup, where (the deformation amount at the loading end caused by the bending of the stirrup itself), (the deformation amount at the loading end of the stirrup caused by the bending of the main bar), (the deformation amount at the loading end of the stirrup caused by the in-plane rotation of the stirrup), , are the section bending moment of inertia of the main bar and the stirrup respectively; similarly, in combination with Figure 6 rotational stiffness , solve the test device to establish , related models; In the formula, is the deformation amount at the loading end caused by the bending of the stirrup itself; is the deformation amount at the stirrup loading end caused by the bending of the main reinforcement; is the deformation amount at the stirrup loading end caused by the in-plane rotation of the stirrup; L1 and L2 are the lengths of the main reinforcement and the stirrup respectively; I1 and I2 are the section bending moment of inertia of the main reinforcement and the stirrup respectively; E is the elastic modulus of the steel bar.

[0063] Through the mechanical solution model, combined with the length of the main reinforcement , the length of the stirrup and the elastic modulus of the steel bar , calculate the overall stiffness of the steel bar component .

[0064] Utilize the multi-physical field coupling technology, consider the influence of multi-field interactions such as thermal-mechanical and force-electricity on the mechanical solution model, and optimize the calculation of the overall stiffness.

[0065] Calculate the overall stiffness of the steel bar component through the mechanical solution model The calculation of the overall stiffness of the steel bar component is based on the principles of structural mechanics, and is determined by considering the deformation contributions of each component (main reinforcement and stirrup) under loading conditions and their interactions. The three-way rotational stiffness of the solder joint ( , , ) reflects the ability of the solder joint to resist rotation in different directions, while the length of the main reinforcement , the length of the stirrup and the elastic modulus of the steel bar are the key parameters describing the material and geometric characteristics of the steel bar component. These parameters are jointly involved in the mechanical solution model to calculate the overall stiffness.

[0066] The specific method includes: First, establish the corresponding mechanical model according to the rotational stiffness in different directions. Taking as an example, combined with Figure 5 rotational stiffness to solve the test device. When the two ends of the main reinforcement are hinged and restrained and a force is applied to the end of the stirrup , the vertical deformation at the stirrup loading end consists of three parts: . Among them, is based on the beam bending deformation theory in material mechanics to calculate the deformation amount at the loading end caused by the bending of the stirrup itself. Here is the section bending moment of inertia of the stirrup, which reflects the ability of the stirrup to resist bending deformation, is the length of the stirrup, is the elastic modulus of the steel bar, is the applied force, and this formula reflects the influence of the stirrup length, bending moment of inertia and elastic modulus on its own bending deformation; Similarly, based on the beam bending theory, calculate the deformation of the stirrup loading end caused by the bending of the main reinforcement. is the moment of inertia of the main reinforcement section against bending; represents the deformation of the stirrup loading end caused by the in-plane rotation of the stirrup, which plays a role in resisting rotational deformation here.

[0067] Similarly, for and , combined with Figure 6 rotational stiffness , Solve the test device to establish a similar mechanical model, considering their respective corresponding deformation components.

[0068] Then, synthesize the deformation calculation results corresponding to the rotational stiffness in each direction. Since the overall stiffness is a measure of the structure's ability to resist deformation, the , , related deformation calculation results are related to the overall deformation of the steel bar component through certain mechanical relationships. In actual calculations, according to the force characteristics and connection methods of the structure, methods such as the matrix displacement method and force method of structural mechanics may be used to integrate the deformation and stiffness information of each part, so as to calculate the overall stiffness of the steel bar component.

[0069] Optimize the overall stiffness calculation using multi-physics field coupling technology Principle: In the actual welding and use process, the steel bar component is affected by multiple physical fields. There are interactions between multiple fields such as thermal-mechanical and force-electricity. These interactions will affect the mechanical properties and deformation characteristics of the steel bar component, and thus affect the overall stiffness. The multi-physics field coupling technology is to consider the mutual relationships between these different physical fields, modify and improve the traditional mechanical solution model, so that the calculation results are more in line with the actual situation.

[0070] Specific method: During the welding process, a large amount of heat is generated, resulting in an increase in the temperature of the steel bar component. The change in temperature will cause changes in material properties. For example, the elastic modulus $E$ will change with temperature. First, a heat transfer model needs to be established to calculate the change in the temperature distribution of the steel bar component over time during the welding process. Then, according to the thermo-mechanical constitutive relationship of the material, determine the changes in mechanical parameters such as the elastic modulus of the steel bar at different temperatures. Substitute these changed mechanical parameters into the mechanical solution model to recalculate the deformation and overall stiffness of the welded joint. For example, if the increase in temperature causes the elastic modulus to decrease, then under the same force condition, the deformation of the steel bar component will increase and the overall stiffness will decrease.

[0071] Under some special circumstances, the steel bar components may be affected by an electric field or generate an electro - effect during the process of force - induced deformation. The force - electricity coupling effect will affect the mechanical properties and deformation behavior of the material. For the force - electricity coupling, it is necessary to establish a force - electricity coupling constitutive model according to specific physical phenomena and material characteristics. For example, some materials will generate piezoelectric effects when stressed, resulting in a change in the internal charge distribution, which in turn affects the mechanical properties of the material. When calculating the overall stiffness, considering this force - electricity interaction and introducing parameters related to the electric field (such as electric field strength) into the mechanical solution model to correct the calculation of the overall stiffness.

[0072] Taking into account the comprehensive multi - field coupling effects such as thermal - force and force - electricity, comprehensively correct the original mechanical solution model. Verify and calibrate the multi - physical - field coupling model through experimental data or numerical simulation results. When calculating the overall stiffness \(K\), continuously iterate and adjust the model parameters to make the calculation results more accurately reflect the overall stiffness characteristics of the actual steel bar components in a complex physical - field environment, and achieve the optimization of the overall stiffness calculation.

[0073] The above - mentioned embodiments are only the preferred technical solutions of the present invention and should not be regarded as limitations on the present invention. The protection scope of the present invention should be the technical solutions recorded in the claims, including the equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, the equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for predicting deformation of steel bar components based on welding parameters, characterized in that: Acquire welding parameters and relevant data of steel bar components, wherein the welding parameters include at least primary parameters and secondary parameters, and the relevant data include at least steel bar diameter; Determine the size of the weld spot according to the welding parameters and the diameter of the steel bar; Determine the three-dimensional rotational stiffness of the weld according to the weld size; Determine the overall stiffness of the steel bar component based on the three-dimensional rotation stiffness of the weld point; Calculate the deformation of the steel bar component based on the overall stiffness of the steel bar component; If the deformation of the steel bar component does not meet the preset requirements, the welding parameters are adjusted and recalculated until the deformation of the steel bar component meets the preset requirements.

2. A method for predicting deformation of steel bar components based on welding parameters according to claim 1, characterized in that: The welding parameters and the steel bar diameter determine the weld size, including: Classify the welding parameters to obtain primary parameters and secondary parameters, wherein the primary parameters include welding time , the secondary parameters include welding voltage and welding current ; Get the main bar diameter for the steel bar diameter and stirrup diameter ; According to welding time With the main reinforcement diameter and the stirrup diameter Establishing the first functional relationship ; By the welding voltage and the welding current The first functional relationship is corrected by introducing a correction coefficient , , and obtain the second functional relationship ; Calculate the solder joint size according to the second functional relationship .

3. A method for predicting deformation of steel bar components based on welding parameters according to claim 2, characterized in that: Use scanning electron microscopy and atomic probe tomography to analyze the microstructure of the solder joint, and combine material science theory to establish a physical model to assist in determining the correction factor , ; At the same time, the support vector regression algorithm is used to train a large amount of welding process test data to optimize the first function relationship .

4. The method for predicting deformation of steel bar components based on welding parameters according to claim 1 is characterized in that: The size of the weld determines the three-dimensional rotational stiffness of the weld, including: Obtain the definition of the three-dimensional rotational stiffness of the weld, which includes the rotational stiffness of the stirrup in the plane where the main reinforcement and the stirrup are located , the rotational stiffness of the main reinforcement with the stirrup as the rotation axis And the rotational stiffness of the stirrups with the main reinforcement as the rotation axis ; For the size of the solder joint, the convolutional neural network in deep learning is used, combined with the material microstructure characteristic parameters, to establish the solder joint size The third functional relationship with the three-dimensional rotational stiffness of the weld , ; The three-way rotational stiffness of the welding point is calculated by the third functional relationship .

5. A method for predicting deformation of steel bar components based on welding parameters according to claim 4, characterized in that: Finite element analysis software is used to finely model the welds, and the theoretical model is solved by combining the three-dimensional rotational stiffness mechanics of the welds. The elastic-plastic deformation, residual stress distribution and micro-defect factors of the material are considered to simulate the stress and deformation of the welds under different working conditions, and assist in determining the function .

6. The method for predicting deformation of steel bar components based on welding parameters according to claim 1, characterized in that: The three-way rotational stiffness of the weld determines the overall stiffness of the steel bar component, including: Acquire geometric parameters and material properties of the steel bar component, wherein the geometric parameters include the main bar length and the stirrup length, and the material properties include the steel bar elastic modulus; establish a mechanical solution model for the three-dimensional rotational stiffness of the weld; calculate the overall stiffness of the steel bar component by combining the main bar length, the stirrup length and the steel bar elastic modulus through the mechanical solution model; By using multi-physics field coupling technology, the influence of thermal-mechanical and mechanical-electrical multi-field interactions on the mechanical solution model is considered to optimize the overall stiffness calculation.

7. The method for predicting deformation of steel bar components based on welding parameters according to claim 1, characterized in that: The overall stiffness of the steel bar component is used to calculate the deformation of the steel bar component, including: Obtain the external load conditions of the steel bar component, the external load conditions including the forces applied to the steel bar component ; According to the overall stiffness of the steel bar component and the force , establish deformation calculation model, such as vertical deformation ; The vertical deformation of the steel bar component is calculated by the deformation calculation model ; According to the vertical deformation Determine whether the deformation of the steel bar component meets the preset requirements.

8. The method for predicting deformation of steel bar components based on welding parameters according to claim 1, characterized in that: If the deformation of the steel bar component does not meet the preset requirements, the welding parameters are adjusted and recalculated, including: Get the preset deformation threshold ; If the steel bar is deformed Exceeding the deformation threshold , according to the inverse relationship between the deformation of the steel bar component and the welding parameters, a particle swarm optimization algorithm or an adaptive genetic algorithm is used to determine the adjusted welding parameters with the goal of minimizing the deformation and optimizing the welding quality; The size of the weld spot, the three-way rotational stiffness of the weld spot and the overall stiffness of the steel bar component are recalculated by the adjusted welding parameters until the steel bar component is deformed. Less than or equal to the deformation threshold .

9. The method for predicting deformation of steel bar components based on welding parameters according to claim 1, characterized in that: The spot size determines the three-dimensional rotational stiffness of the weld, including: Obtaining measurement data from a test device for loading a steel bar component and measuring deformation; For the solder joint size, a fourth functional relationship is established through the measurement data , ; The calculation result of the three-way rotation stiffness of the solder joint is verified by the fourth functional relationship, and the calculation error is , ; If the calculation result is inconsistent with the measurement data, When the allowable error range is exceeded, the third functional relationship is adjusted and the three-way rotational stiffness of the weld is recalculated.

10. A method for predicting deformation of steel bar components based on welding parameters according to claim 9, characterized in that: Use non-contact measurement technology to obtain more accurate measurement data.

Images