Simulation optimization method for heat source distribution of alloy welding structural member and terminal

Abstract

The invention discloses a simulation optimization method and terminal for heat source distribution of an alloy welding structural member, and the method comprises the steps: obtaining and inputting at least two welding process parameters to a welding heat source model, and obtaining at least two simulation heat source distribution geometric parameters in a one-to-one correspondence manner; acquiring actual geometrical characteristics of transverse and longitudinal sections of welding seams of the alloy welding structural parts in one-to-one correspondence with the welding process parameters; according to all the simulation heat source distribution geometric parameters and the actual geometric features, an objective function used for reducing the difference between the simulation heat source distribution geometric parameters and the corresponding actual geometric features is established, and a heat source distribution optimization model is established according to the objective function; and in the welding simulation process, the heat source distribution optimization model is called to optimize simulation heat source distribution geometric parameters output by the welding heat source model. According to the method, the simulation heat source distribution geometric parameters are optimized through the heat source distribution optimization model, the accuracy of the simulation result of the welding heat source model is improved, the number of trial and error times in the welding process is reduced, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular to a simulation optimization method and a terminal for heat source distribution of an alloy welded structural part. Background Art

[0002] The welding temperature field has an important influence on the deformation and residual stress distribution after welding. For aluminum alloy welded structural parts, the welding temperature field plays a decisive role in the reliability of the structure after welding. In the simulation of aluminum alloy welding, the welding temperature field is mainly affected by the distribution of welding heat source, so it is particularly important to achieve accurate control of the distribution of welding heat source.

[0003] However, the existing geometric parameters of heat source distribution for aluminum alloy welding structures often rely on the operator's experience or need to be determined through a large number of trial and error processes. The precision and accuracy of the simulation conducted in this way are difficult to guarantee, which inevitably increases the number of trial and error in the welding process and seriously affects production efficiency. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a simulation optimization method and terminal for heat source distribution of alloy welded structural parts, reduce the number of trial and error in the welding process, improve the accuracy of welding thermal process simulation, and improve production efficiency.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A simulation optimization method for heat source distribution of alloy welded structural parts comprises the following steps:

[0007] S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence;

[0008] S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one;

[0009] S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function;

[0010] S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

[0011] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0012] A simulation optimization terminal for heat source distribution of alloy welded structural parts, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0013] S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence;

[0014] S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one;

[0015] S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function;

[0016] S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

[0017] The beneficial effects of the present invention are as follows: the present invention provides a simulation optimization method and terminal for the heat source distribution of an alloy welded structural part. Based on a machine learning algorithm, the simulated heat source distribution geometric parameters corresponding to different welding process parameters and the actual geometric characteristics of the weld's transverse and longitudinal sections are used as data basis, an objective function is constructed, and a heat source distribution optimization model is established. The simulated heat source distribution geometric parameters are optimized through the heat source distribution optimization model to narrow the gap between the simulated heat source distribution geometric parameters and the actual geometric characteristics, thereby making the simulation process results of the welding heat source model more accurate, reducing the number of trial and error times in the welding process, and improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the steps of a simulation optimization method for heat source distribution of an alloy welded structural part of the present invention;

[0019] Figure 2 It is a flow chart of a simulation optimization method for heat source distribution of an alloy welded structural part of the present invention;

[0020] Figure 3 The present invention is a system block diagram of a terminal for simulating and optimizing heat source distribution of an alloy welded structural part.

[0021] Description of labels:

[0022] 1. A simulation and optimization terminal for heat source distribution of alloy welded structural parts; 2. Memory; 3. Processor. DETAILED DESCRIPTION

[0023] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0024] Please refer to Figure 1 and Figure 2 , a simulation optimization method for heat source distribution of alloy welded structural parts, comprising the following steps:

[0025] S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence;

[0026] S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one;

[0027] S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function;

[0028] S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

[0029] From the above description, it can be seen that the beneficial effects of the present invention are: based on a machine learning algorithm, the simulated heat source distribution geometric parameters corresponding to different welding process parameters and the actual geometric characteristics of the weld transverse and longitudinal sections are used as data basis, an objective function is built, and a heat source distribution optimization model is established. The simulated heat source distribution geometric parameters are optimized through the heat source distribution optimization model to narrow the gap between them and the actual geometric characteristics, thereby making the simulation process results of the welding heat source model more accurate, reducing the number of trial and error in the welding process, and improving production efficiency.

[0030] Furthermore, before step S3, the following steps are also included:

[0031] According to the simulated heat source distribution geometric parameters, a one-to-one correspondence is simulated for the simulated welding pool to obtain simulated geometric features of the transverse and longitudinal sections of the welding pool;

[0032] The step S3 is replaced by:

[0033] Based on all the simulated geometric features and the actual geometric features, an objective function for reducing the gap between the simulated geometric features and the corresponding actual geometric features is established, and a heat source distribution optimization model is established based on the objective function.

[0034] From the above description, it can be seen that when comparing with the actual geometric features, the geometric parameters of the simulated heat source distribution are first automatically solved to simulate the welding pool, and then the simulated geometric features of the transverse and longitudinal sections of the welding pool are extracted to form a more intuitive comparison with the actual geometric features, thereby improving the efficiency and accuracy of the calculation and analysis.

[0035] Furthermore, before step S4, the following steps are also included:

[0036] Sample data including different simulated geometric features and corresponding actual geometric features are obtained, and the sample data are called to perform model training on the heat source distribution optimization model.

[0037] From the above description, it can be seen that after the heat source distribution optimization model is constructed, a large amount of sample data is used to train the heat source distribution optimization model to ensure that the optimization results of the heat source distribution optimization model are more reasonable, more accurate, and more reliable.

[0038] Furthermore, before step S4, the following steps are also included:

[0039] Obtaining optimized simulated heat source distribution geometric parameters output by the heat source distribution optimization model;

[0040] Compare the optimized simulated heat source distribution geometric parameters with the corresponding actual geometric features to obtain the geometric shape overlap error;

[0041] If the geometric shape overlap error is less than or equal to a preset error value, the heat source distribution optimization model is qualified.

[0042] From the above description, it can be seen that before the heat source distribution optimization model is put into use, error verification of the output results of the heat source distribution optimization model can be used as a criterion for terminating model training to ensure that the optimization process is sufficiently reliable.

[0043] Furthermore, before step S1, the following steps are also included:

[0044] Constructing a double ellipsoid heat source model corresponding to the alloy welding structure, and using the double ellipsoid heat source model as the welding heat source model;

[0045] The expression of the double ellipsoid heat source model is as follows:

[0046]

[0047] Among them, a1, a2, b and c represent shape parameters, η represents absorptivity, P represents power, f1 and f2 represent the ratio of heat between the front and rear parts, v represents the moving speed of the heat source, t represents time, and x, y, z represent the relative position of the integration point and the center of the heat source in the local coordinate system.

[0048] From the above description, it can be seen that when simulating the heat source model of alloy welded structural parts, a double ellipsoid heat source model can be constructed, which has wide applicability and high accuracy.

[0049] Please refer to Figure 3 A simulation optimization terminal 1 for heat source distribution of alloy welded structural parts includes a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, the following steps are implemented:

[0050] S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence;

[0051] S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one;

[0052] S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function;

[0053] S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

[0054] From the above description, it can be seen that the beneficial effects of the present invention are: based on a machine learning algorithm, the simulated heat source distribution geometric parameters corresponding to different welding process parameters and the actual geometric characteristics of the weld transverse and longitudinal sections are used as data basis, an objective function is built, and a heat source distribution optimization model is established. The simulated heat source distribution geometric parameters are optimized through the heat source distribution optimization model to narrow the gap between them and the actual geometric characteristics, thereby making the simulation process results of the welding heat source model more accurate, reducing the number of trial and error in the welding process, and improving production efficiency.

[0055] Furthermore, before step S3, the following steps are also included:

[0056] According to the simulated heat source distribution geometric parameters, a one-to-one correspondence is simulated for the simulated welding pool to obtain simulated geometric features of the transverse and longitudinal sections of the welding pool;

[0057] The step S3 is replaced by:

[0058] Based on all the simulated geometric features and the actual geometric features, an objective function for reducing the gap between the simulated geometric features and the corresponding actual geometric features is established, and a heat source distribution optimization model is established based on the objective function.

[0059] From the above description, it can be seen that when comparing with the actual geometric features, the geometric parameters of the simulated heat source distribution are first automatically solved to simulate the welding pool, and then the simulated geometric features of the transverse and longitudinal sections of the welding pool are extracted to form a more intuitive comparison with the actual geometric features, thereby improving the efficiency and accuracy of the calculation and analysis.

[0060] Furthermore, before step S4, the following steps are also included:

[0061] Sample data including different simulated geometric features and corresponding actual geometric features are obtained, and the sample data are called to perform model training on the heat source distribution optimization model.

[0062] From the above description, it can be seen that after the heat source distribution optimization model is constructed, a large amount of sample data is used to train the heat source distribution optimization model to ensure that the optimization results of the heat source distribution optimization model are more reasonable, more accurate, and more reliable.

[0063] Furthermore, before step S4, the following steps are also included:

[0064] Obtaining optimized simulated heat source distribution geometric parameters output by the heat source distribution optimization model;

[0065] Compare the optimized simulated heat source distribution geometric parameters with the corresponding actual geometric features to obtain the geometric shape overlap error;

[0066] If the geometric shape overlap error is less than or equal to a preset error value, the heat source distribution optimization model is qualified.

[0067] From the above description, it can be seen that before the heat source distribution optimization model is put into use, error verification of the output results of the heat source distribution optimization model can be used as a criterion for terminating model training to ensure that the optimization process is sufficiently reliable.

[0068] Furthermore, before step S1, the following steps are also included:

[0069] Constructing a double ellipsoid heat source model corresponding to the alloy welding structure, and using the double ellipsoid heat source model as the welding heat source model;

[0070] The expression of the double ellipsoid heat source model is as follows:

[0071]

[0072] Among them, a1, a2, b and c represent shape parameters, η represents absorptivity, P represents power, f1 and f2 represent the ratio of heat between the front and rear parts, v represents the moving speed of the heat source, t represents time, and x, y, z represent the relative position of the integration point and the center of the heat source in the local coordinate system.

[0073] From the above description, it can be seen that when simulating the heat source model of alloy welded structural parts, a double ellipsoid heat source model can be constructed, which has wide applicability and high accuracy.

[0074] Please refer to Figure 1 and Figure 2 , Embodiment 1 of the present invention is:

[0075] A simulation optimization method for heat source distribution of alloy welded structural parts comprises the following steps:

[0076] S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence;

[0077] In this embodiment, a double ellipsoid heat source model corresponding to the alloy welding structure is constructed, and the double ellipsoid heat source model is used as the welding heat source model;

[0078] The expression of the double ellipsoid heat source model is as follows:

[0079]

[0080] Wherein, a1, a2, b and c represent shape parameters, η represents absorptivity, P represents power, f1 and f2 represent the ratio of heat between the front and rear parts, v represents the moving speed of the heat source, t represents time, and x, y, z represent the relative position of the integral point and the center of the heat source in the local coordinate system. In other equivalent embodiments, the welding heat source model can also use other heat source models such as a hemispherical heat source model and an ellipsoidal heat source model. In addition, the welding method is not limited, and can be laser welding, resistance welding and other welding methods.

[0081] S2. Acquire the actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural part corresponding to the welding process parameters one by one;

[0082] In this embodiment, according to the simulated heat source distribution geometric parameters, the welding pool is simulated one by one to obtain the simulated geometric features of the horizontal and vertical sections of the welding pool; specifically, a multi-software collaborative calculation method is adopted to simulate the welding pool first, and then extract the simulated geometric features of the horizontal and vertical sections of the welding pool. For example, software such as Abaqus, MATLAB and Python are used at the same time, Abaqus is used for welding thermal analysis simulation calculation, MATLAB is used to extract the geometric feature data of the horizontal and vertical sections of the welding pool, and Python is used for post-processing of the results to realize the automatic solution of the heat source distribution geometric parameters.

[0083] S3. According to all the simulated geometric features and the actual geometric features, an objective function for narrowing the gap between the simulated geometric features and the corresponding actual geometric features is established, and a heat source distribution optimization model is established based on the objective function;

[0084] Among them, regarding the objective function, an example is: a i+1 =f i (w ij ·a i +b ij ). Among them, w ij is the coefficient matrix of the i-th layer, b ij is the bias matrix of the i-th layer, f i is the activation function of the i-th layer, and the multi-layer perceptron (MLP) model is selected, and the Adam optimizer is selected. The objective function takes the simulated geometric features as input, and adjusts the coefficients to make the results closer to the actual geometric features.

[0085] In this embodiment, sample data including different simulated geometric features and corresponding actual geometric features are obtained, and the sample data is used to train the heat source distribution optimization model. In addition, the effect of the training is experimentally verified:

[0086] Obtain the optimized simulated heat source distribution geometric parameters output by the heat source distribution optimization model; compare the optimized simulated heat source distribution geometric parameters with the corresponding actual geometric features for geometric shape overlap to obtain the geometric shape overlap degree error; if the geometric shape overlap degree error is less than or equal to a preset error value, preferably 20%, the heat source distribution optimization model is qualified.

[0087] S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

[0088] Please refer to Figure 3 , Embodiment 2 of the present invention is:

[0089] A simulation optimization terminal 1 for heat source distribution of an alloy welded structural part comprises a memory 2, a processor 3 and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, a simulation optimization method for heat source distribution of an alloy welded structural part of embodiment 1 is implemented.

[0090] In summary, the present invention provides a simulation optimization method and terminal for the heat source distribution of an alloy welded structural part. Based on a machine learning algorithm, the simulated heat source distribution geometric parameters corresponding to different welding process parameters and the actual geometric characteristics of the weld's transverse and longitudinal sections are used as data basis, an objective function is constructed, and a heat source distribution optimization model is established. The simulated heat source distribution geometric parameters are optimized through the heat source distribution optimization model to narrow the gap between the simulated heat source distribution geometric parameters and the actual geometric characteristics. A large amount of sample data is used to train the heat source distribution optimization model, and error comparison is combined as a criterion for training termination, so that the simulation process results of the welding heat source model are more accurate, the number of trial and error in the welding process is reduced, and production efficiency is improved.

[0091] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A simulation optimization method for heat source distribution of alloy welded structural parts, characterized in that: The steps include: S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence; S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one; S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function; S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

2. The simulation optimization method for heat source distribution of alloy welded structural parts according to claim 1, characterized in that: Before step S3, the following steps are also included: According to the simulated heat source distribution geometric parameters, a one-to-one correspondence is simulated for the simulated welding pool to obtain simulated geometric features of the transverse and longitudinal sections of the welding pool; The step S3 is replaced by: Based on all the simulated geometric features and the actual geometric features, an objective function for reducing the gap between the simulated geometric features and the corresponding actual geometric features is established, and a heat source distribution optimization model is established based on the objective function.

3. The simulation optimization method for heat source distribution of alloy welded structural parts according to claim 2, characterized in that: Before step S4, the following steps are also included: Sample data including different simulated geometric features and corresponding actual geometric features are obtained, and the sample data are called to perform model training on the heat source distribution optimization model.

4. The simulation optimization method for heat source distribution of an alloy welded structural part according to claim 1, characterized in that: Before step S4, the following steps are also included: Obtaining optimized simulated heat source distribution geometric parameters output by the heat source distribution optimization model; Compare the optimized simulated heat source distribution geometric parameters with the corresponding actual geometric features to obtain the geometric shape overlap error; If the geometric shape overlap error is less than or equal to a preset error value, the heat source distribution optimization model is qualified.

5. The simulation optimization method for heat source distribution of alloy welded structural parts according to claim 1, characterized in that: Before step S1, the following steps are also included: Constructing a double ellipsoid heat source model corresponding to the alloy welding structure, and using the double ellipsoid heat source model as the welding heat source model; The expression of the double ellipsoid heat source model is as follows: Among them, a1, a2, b and c represent shape parameters, η represents absorptivity, P represents power, f1 and f2 represent the ratio of heat between the front and rear parts, v represents the moving speed of the heat source, t represents time, and x, y, z represent the relative position of the integration point and the center of the heat source in the local coordinate system.

6. A simulation optimization terminal for heat source distribution of alloy welded structural parts, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: S1. Obtain and input at least two welding process parameters into a welding heat source model, and obtain at least two simulated heat source distribution geometric parameters in one-to-one correspondence; S2, obtaining actual geometric characteristics of the cross-section and longitudinal section of the weld of the alloy welded structural component corresponding to the welding process parameters one by one; S3, according to all the simulated heat source distribution geometric parameters and the actual geometric features, establishing an objective function for narrowing the gap between the simulated heat source distribution geometric parameters and the corresponding actual geometric features, and establishing a heat source distribution optimization model based on the objective function; S4. During the welding simulation process, the heat source distribution optimization model is called to optimize the simulated heat source distribution geometric parameters output by the welding heat source model.

7. The simulation optimization terminal for heat source distribution of alloy welded structural parts according to claim 6, characterized in that: Before step S3, the following steps are also included: According to the simulated heat source distribution geometric parameters, a one-to-one correspondence is simulated for the simulated welding pool to obtain simulated geometric features of the transverse and longitudinal sections of the welding pool; The step S3 is replaced by: Based on all the simulated geometric features and the actual geometric features, an objective function for reducing the gap between the simulated geometric features and the corresponding actual geometric features is established, and a heat source distribution optimization model is established based on the objective function.

8. The simulation optimization terminal for heat source distribution of alloy welded structural parts according to claim 7, characterized in that: Before step S4, the following steps are also included: Sample data including different simulated geometric features and corresponding actual geometric features are obtained, and the sample data are called to perform model training on the heat source distribution optimization model.

9. The simulation optimization terminal for heat source distribution of alloy welded structural parts according to claim 6, characterized in that: Before step S4, the following steps are also included: Obtaining optimized simulated heat source distribution geometric parameters output by the heat source distribution optimization model; Compare the optimized simulated heat source distribution geometric parameters with the corresponding actual geometric features to obtain the geometric shape overlap error; If the geometric shape overlap error is less than or equal to a preset error value, the heat source distribution optimization model is qualified.

10. The simulation optimization terminal for heat source distribution of alloy welded structural parts according to claim 6, characterized in that: Before step S1, the following steps are also included: Constructing a double ellipsoid heat source model corresponding to the alloy welding structure, and using the double ellipsoid heat source model as the welding heat source model; The expression of the double ellipsoid heat source model is as follows: Among them, a1, a2, b and c represent shape parameters, η represents absorptivity, P represents power, f1 and f2 represent the ratio of heat between the front and rear parts, v represents the moving speed of the heat source, t represents time, and x, y, z represent the relative position of the integration point and the center of the heat source in the local coordinate system.

Images