A method and system for prestress and thermal vibration aging synergistic regulation of residual stress

Abstract

The application relates to the field of part machining, and discloses a prestress and thermal vibration aging synergic regulation and control residual stress method and system, which comprises the following steps: preparing a part, obtaining the Poisson's ratio and elastic modulus of the part; measuring the initial residual stress of the part; encoding the initial residual stress of the part as input data set; placing the part on a fixed heating furnace table through a clamp; forming a prestress field through stretching of the clamp; and performing thermal aging on the part; the thermal aging process mainly considers material thermal performance factors; meanwhile, a vibration exciter arranged on the workbench starts to work to perform vibration aging on the part; through combination of the prestress technology and the thermal vibration aging technology, the part residual stress regulation and control is integrated and homogenized, the part machining quality and machining precision are improved, relevant parameters of parts which have not been machined can be predicted, the problem that the residual stress of high-strength composite materials is difficult to regulate and control is solved, and the application has great economic value.

Description

Technical Field

[0001] This invention relates to the field of parts processing, and in particular to a method and system for synergistic control of residual stress through prestressing and thermal vibration aging. Background Technology

[0002] Currently, alloy parts often suffer from high residual stress due to deformation during forging and heat treatment, as well as uneven cooling. This leads to deformation during final machining, resulting in scrapped forgings. Extensive research has revealed that the primary factor causing this machining deformation is the initial residual stress within the blank. Therefore, solving the machining deformation problem becomes a matter of controlling this initial residual stress. The manufacturing precision of high-performance alloy parts has become a major issue that needs to be addressed in my country's strategic aerospace projects. Commonly used methods for residual stress control include thermal aging and vibration aging. Both methods are relatively mature. Thermal aging coarsens the grains of components, reducing their strength, while vibration aging can reduce residual stress on the end faces of components. However, both methods have their own drawbacks. For large components, their effective range is limited and their precision is low. Furthermore, residual stress control in high-strength composite materials is difficult and lacks precision. Prestressing, a commonly used engineering technique, can partially offset residual stress generated during component manufacturing, thereby improving the overall performance of the component. Coordinating prestressing and thermal vibration aging to control residual stress presents challenges due to the numerous controllable parameters and the difficulty in ensuring control precision. Therefore, a novel intelligent processing method is needed to achieve coordinated control of residual stress through prestressing and thermal vibration aging. Thus, to address these issues, a method and system for the coordinated control of residual stress through prestressing and thermal vibration aging is crucial. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method and system for synergistic control of residual stress through prestressing and thermal vibration aging, thus solving the problems existing in the background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for synergistically controlling residual stress through prestressing and thermal vibration aging, comprising the following steps: S1. Prepare the part, obtain the Poisson's ratio and elastic modulus of the part; measure the initial residual stress of the part; compile the initial residual stress of the part into an input dataset; S2. The part is placed on a fixed heating furnace platform using a fixture. A prestress field is formed by stretching the fixture to perform thermal aging on the part. The thermal aging process mainly considers the thermal properties of the material, including the heating rate, holding temperature, and cooling rate of the part. At the same time, the vibrator arranged on the worktable is started to work to perform vibration aging on the part. The vibration aging process mainly considers the excitation frequency, excitation time, and excitation amplitude. S3. Record the residual stress value, prestress value, heating rate, holding temperature, cooling rate, excitation frequency, excitation time and excitation amplitude of the parts after aging. S4. Compile the above data into an output dataset; merge the input and output datasets into a single dataset, and divide the dataset into a test set according to a 4:1 ratio. Train the RBF neural network model on the dataset. S5. Based on machine learning technology, optimize the RBF neural network model; this mainly includes: using the Sparrow Search Algorithm (SSA) to adjust the weights, center value, and width of the RBF neural network algorithm; and then assigning the globally optimal parameters to the RBF neural network to improve its predictive performance. S6. Based on the prediction model, the production parameters of the parts are optimized using an improved slime mold optimization algorithm.

[0005] Furthermore, in S2, thermal vibration aging is performed to regulate residual stress during the continuous application of the prestressing field, which differs from conventional residual stress regulation methods.

[0006] Furthermore, in step S4, to improve the stability and prediction performance of the network model, missing value processing, outlier processing, and data value standardization are performed on the dataset.

[0007] Furthermore, in step S5, the weights, center values, and width values ​​of the RBF neural network algorithm are adjusted using the Sparrow Search Algorithm (SSA). Then, the globally optimal parameters are assigned to the RBF neural network to improve its predictive performance. The weights, center values, and width values ​​of the basis functions in an RBF (radial basis function) neural network directly affect its predictive performance. Traditionally, the weights and basis function widths of RBF neural networks are randomly generated, which easily leads to various errors in BP neural network training and makes it difficult to obtain the globally optimal solution. The Sparrow Search Algorithm has advantages such as strong global search capability, simple and fast parameter setting, and excellent convergence performance, which can improve the fit between the predicted and actual values ​​of the RBF neural network.

[0008] Furthermore, the steps in S5 for optimizing the RBF neural network model based on machine learning technology are as follows: S5.1: Establish the RBF neural network by inputting the input and output datasets into the RBF neural network and establishing the RBF prediction model; S5.2: Initialize the weights, center values, and width values ​​of the RBF neural network, using these values ​​as variables in the search space and setting the upper and lower bounds of the search space; S5.3: Initialize the sparrow population and divide them into discoverers and joiners based on their fitness values; S5.4: Calculate the population fitness values ​​and update the optimal position of the discoverers; S5.5: Update the population fitness values, update the positions of the discoverers and joiners, and record the global optimal solution; S5.6: Determine if the stopping condition is met. If the stopping condition is not met, repeat the above steps; otherwise, return the global optimal solution as the optimal weights and basis function width values ​​of the RBF neural network.

[0009] Furthermore, in S6, the output data is optimized using an improved Slime Mould Algorithm (SMA). SMA has the advantages of fast convergence speed, strong global search capability, and simple algorithm implementation, but it also has the disadvantages of low initial population quality and slow convergence speed. The reverse learning algorithm can find the reverse solution based on the current solution. The reverse solution has a higher probability of approximating the optimal solution than the current solution, thereby improving the population quality and thus improving the quality of the algorithm and the optimization results.

[0010] Furthermore, the step in S6 of optimizing production parameters using the SMA algorithm through refraction-based reverse learning is as follows: S6.1, initialize the population using a reverse learning strategy and calculate the fitness value; S6.2, update the slime mold weight parameters; S6.3, determine the position of the offspring and update the position of the offspring; S6.4, calculate the fitness value and update the global optimal solution; S6.5, determine whether the termination condition is met. If not, repeat the above steps. In step S6.3, r is a random number [0,1], and z is the proportion of randomly distributed slime mold individuals to the total population. S(i) represents fitness, and DF represents the best fitness among all iterations.

[0011] According to another aspect of the present invention, a thermal vibration table system is provided, comprising a power supply, a workpiece, a worktable, a heating system, a vibration system, and a clamping system. The heating system includes a heating coupler, a heating furnace, and a temperature controller. The vibration system includes a vibrator and a frequency controller. The clamping device includes a clamp, a guide rail, and a prestress control device arranged longitudinally on the worktable. The prestress control device applies a prestress field to the workpiece.

[0012] This invention provides a method and system for synergistic control of residual stress through prestressing and thermal vibration aging, which has the following beneficial effects: By combining prestressing technology and thermal vibration aging technology, the overall and uniform control of residual stress in parts is achieved, which improves the processing quality and accuracy of parts. It can also predict the relevant parameters of unprocessed parts, solves the problem of difficult control of residual stress in high-strength composite materials by existing technology, and has great economic value. The process involves heating the workpiece to a certain temperature in a furnace, holding it at that temperature for a period of time, and then controlling the cooling process. The basic principle is that the yield strength of a material decreases as temperature increases. When the residual stress inside the workpiece exceeds its yield limit, localized plastic deformation occurs, releasing residual strain and reducing residual stress. Simultaneously, a vibration device applies cyclic load vibration at a specific frequency to the workpiece for a period of time, applying additional alternating stress or deformation. When this additional alternating stress is superimposed on the residual stress, energy is absorbed through internal friction. When this energy reaches or exceeds a certain threshold, the workpiece undergoes microscopic or macroscopic viscoelastic-plastic mechanical changes, thereby reducing and homogenizing the residual stress inside the workpiece and stabilizing its dimensional accuracy. The Sparrow Search algorithm (SSA) is used to adjust the weights, center values, and width values ​​of the RBF neural network algorithm. Then, the globally optimal parameters are assigned to the RBF neural network to improve its prediction performance. The weights, center values, and width values ​​of the basis functions in the RBF (radial basis function) neural network directly affect the prediction performance of the neural network. The weights and basis function widths of the traditional RBF neural network are randomly generated, which can easily lead to various errors in the training of the BP neural network and make it difficult to obtain the globally optimal solution. The Sparrow Search algorithm has the advantages of strong global search capability, simple and fast parameter setting, and excellent convergence performance, which can improve the fit between the predicted value and the actual value of the RBF neural network. The improved Slime Mould Algorithm (SMA) optimizes the production parameters of the output data. SMA has the advantages of fast convergence speed, strong global search capability, and simple algorithm implementation. However, it also has the disadvantages of low initial population quality and slow convergence speed. The reverse learning algorithm can find the reverse solution based on the current solution. The reverse solution has a higher probability of approximating the optimal solution than the current solution, thereby improving the population quality and thus improving the quality of the algorithm and the optimization results. Attached Figure Description

[0013] Figure 1 A flowchart illustrating the operation of a method and system for synergistic control of residual stress through prestressing and thermal vibration aging; Figure 2 This is a flowchart of an algorithm for optimizing an RBF neural network model based on machine learning technology in a method and system for synergistic control of residual stress through prestressing and thermal vibration aging. Figure 3 A flowchart of an improved slime mold algorithm in a method for synergistic control of residual stress through prestressing and thermal vibration aging. Figure 4 This is a schematic diagram of a method for synergistic control of residual stress through prestressing and thermal vibration aging, and the thermal vibration table system within the system.

[0014] In the diagram: 1. Heating coupler; 2. Heating furnace; 3. Temperature controller; 4. Power supply; 5. Prestress control device; 6. Workbench; 7. Guide rail; 8. Fixture; 9. Workpiece; 10. Vibration controller; 11. Vibrator. Detailed Implementation

[0015] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] This invention provides a technical solution: a method for synergistically controlling residual stress through prestressing and thermal vibration aging, comprising the following steps: S1. Prepare the part, obtain the Poisson's ratio and elastic modulus of the part; measure the initial residual stress of the part; compile the initial residual stress of the part into an input dataset.

[0017] S2. The part is placed on a fixed heating furnace platform using a fixture. A prestress field is formed by the stretching of the fixture to thermally age the part. The thermal aging process mainly considers the thermal properties of the material, including the heating rate, holding temperature, and cooling rate of the part. At the same time, the vibrator arranged on the worktable is started to work to vibrate and age the part. The vibration aging process mainly considers the excitation frequency, excitation time, and excitation amplitude. During the continuous application of the prestress field, residual stress is controlled by thermal vibration aging, which is different from conventional residual stress control methods.

[0018] S3. Record the residual stress value, prestress value, heating rate, holding temperature, cooling rate, excitation frequency, excitation time and excitation amplitude of the parts after aging.

[0019] S4. Compile the above data into an output dataset; merge the input and output datasets into a single dataset, and divide the test set according to a 4:1 data ratio. Train the RBF neural network model on the dataset. To improve the stability and prediction performance of the network model, perform missing value processing, outlier processing, and data value standardization on the dataset.

[0020] S5. Optimize the RBF neural network model based on machine learning technology. This mainly includes: using the Sparrow Search Algorithm (SSA) to adjust the weights, center values, and width values ​​of the RBF neural network algorithm; then assigning the globally optimal parameters to the RBF neural network to improve its predictive performance. The weights, center values, and width values ​​of the basis functions in an RBF (radial basis function) neural network directly affect its predictive performance. Traditionally, the weights and basis function widths of RBF neural networks are randomly generated, which easily leads to various errors in BP neural network training and makes it difficult to obtain the globally optimal solution. The Sparrow Search Algorithm has advantages such as strong global search capability, simple and fast parameter setting, and excellent convergence performance, which can improve the fit between the predicted and actual values ​​of the RBF neural network. The steps for optimizing the RBF neural network model based on machine learning technology are as follows: S5.1. Establish the RBF neural network by inputting the input and output datasets into the RBF neural network to establish the RBF prediction model. S5.2 Initialize the weights, center values, and width values ​​of the RBF neural network, using them as variables in the search space, and setting the upper and lower bounds of the search space. S5.3 Initialize the sparrow population and divide individuals into discoverers and joiners based on their fitness values. S5.4 Calculate the population fitness values ​​and update the optimal position of the discoverers. S5.5 Update the population fitness values, update the positions of the discoverers and joiners, and record the global optimal solution. S5.6 Determine if the stopping condition is met. If the stopping condition is not met, repeat the above steps; otherwise, return the global optimal solution as the optimal weights and basis function width values ​​of the RBF neural network.

[0021] S6. Based on the prediction model, the production parameters of the parts are optimized using an improved slime mold optimization algorithm. The improved Slime Mould Algorithm (SMA) optimizes the production parameters of the output data. SMA has advantages such as fast convergence speed, strong global search capability, and simple algorithm implementation, but it also has drawbacks such as low initial population quality and slow convergence speed. The reverse learning algorithm can find the reverse solution based on the current solution. The reverse solution has a higher probability of approximating the optimal solution than the current solution, thus improving the population quality and consequently the algorithm quality and optimization results. The steps for optimizing production parameters using the reverse learning algorithm with SMA are as follows: S6.1. Initialize the population using the reverse learning strategy and calculate the fitness value; S6.2. Update the slime mold weight parameters; S6.3. Determine the position of the offspring and update the offspring position; S6.4. Calculate the fitness value and update the global optimal solution; S6.5. Determine if the termination condition is met. If not, repeat the above steps. In step S6.3, r is a random number [0,1], and z is the proportion of randomly distributed slime mold individuals in the total population. S(i) represents fitness, and DF represents the best fitness among all iterations.

[0022] The method also includes a thermal vibration worktable system, which includes a power supply 4, a workpiece 9, a worktable 6, a heating system, a vibration system, and a clamping system. The heating system includes a heating coupler 1, a heating furnace 2, and a temperature controller 3. The vibration system includes a vibrator 11 and a frequency controller 10. The clamping device includes a clamp 8, a guide rail 7, and a prestress control device 5 arranged longitudinally on the worktable 6. The prestress control device 5 applies a prestress field to the workpiece.

[0023] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0024] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for synergistically controlling residual stress through prestressing and thermal vibration aging, characterized in that, Includes the following steps: S1. Prepare the part, obtain the Poisson's ratio and elastic modulus of the part; measure the initial residual stress of the part; compile the initial residual stress of the part into an input dataset; S2. The part is placed on a fixed heating furnace platform using a fixture. A prestress field is formed by stretching the fixture to perform thermal aging on the part. The thermal aging process mainly considers the thermal properties of the material, including the heating rate, holding temperature, and cooling rate of the part. At the same time, the vibrator arranged on the worktable is started to work to perform vibration aging on the part. The vibration aging process mainly considers the excitation frequency, excitation time, and excitation amplitude. S3. Record the residual stress value, prestress value, heating rate, holding temperature, cooling rate, excitation frequency, excitation time and excitation amplitude of the parts after aging. S4. Compile the above data into an output dataset; merge the input and output datasets into a single dataset, and divide the dataset into a test set according to a 4:1 ratio. Train the RBF neural network model on the dataset. S5. Based on machine learning technology, optimize the RBF neural network model; mainly including: optimizing the weights, center value, and width value of the RBF neural network algorithm through the Sparrow Search Algorithm (SSA); and then assigning the globally optimal parameters to the RBF neural network to improve its prediction performance. S6. Based on the prediction model, the production parameters of the parts are optimized using an improved slime mold optimization algorithm.

2. The method for synergistic control of residual stress by prestressing and thermal vibration aging according to claim 1, characterized in that: In S2, thermal aging is used to regulate residual stress during the continuous application of the prestressing field, which is different from conventional residual stress regulation methods.

3. The method for synergistic control of residual stress by prestressing and thermal vibration aging according to claim 1, characterized in that: In step S4, to improve the stability and prediction performance of the network model, missing value processing, outlier processing, and data value standardization are performed on the dataset.

4. The method for synergistic control of residual stress by prestressing and thermal vibration aging according to claim 1, characterized in that: The steps in S5 for optimizing the RBF neural network model based on machine learning technology are as follows: S5.1: Establish the RBF neural network by inputting the input and output datasets into the RBF neural network and establishing the RBF prediction model; S5.2: Initialize the weights, center values, and width values ​​of the RBF neural network, using these values ​​as variables in the search space and setting the upper and lower bounds of the search space; S5.3: Initialize the sparrow population and divide them into discoverers and joiners based on their fitness values; S5.4: Calculate the population fitness values ​​and update the optimal position of the discoverers; S5.5: Update the population fitness values, update the positions of the discoverers and joiners, and record the global optimal solution; S5.6: Determine if the stopping condition is met. If the stopping condition is not met, repeat the above steps; otherwise, return the global optimal solution as the optimal weights and basis function width values ​​of the RBF neural network.

5. The method for synergistic control of residual stress by prestressing and thermal vibration aging according to claim 1, characterized in that: The steps in S6 for optimizing production parameters using the SMA slime mold algorithm through refraction back learning are as follows: S6.1 Initialize the population using the back learning strategy and calculate the fitness value; S6.2 Update the slime mold weight parameters; S6.3 Determine the position of the offspring and update the position of the offspring; S6.4 Calculate the fitness value and update the global optimal solution; S6.5 Determine whether the termination condition is met. If not, repeat the above steps.

6. A thermal vibration table system, comprising the method for synergistic control of residual stress through prestressing and thermal vibration aging as described in any one of claims 1-5, characterized in that: The device includes a power supply (4), a workpiece (9), a worktable (6), a heating system, a vibration system, and a clamping system. The heating system includes a heating coupler (1), a heating furnace (2), and a temperature controller (3). The vibration system includes a vibrator (11) and a frequency controller (10). The clamping device includes a clamp (8), a guide rail (7), and a prestress control device (5) arranged longitudinally on the worktable (6). The prestress control device (5) applies a prestress field to the workpiece.

Images