Arc height value prediction model training and prediction method, device, medium and product

Abstract

The application discloses a mechanical shot forming arc height value prediction model training method, a mechanical shot forming arc height value prediction method, equipment, a medium and a product. The mechanical shot forming arc height value prediction model training method comprises the following steps: constructing a plurality of parameter sets, using the first and second parameter sets to respectively perform simulation experiments and real experiments, obtaining corresponding shot gun moving speeds and arc height values, and adding the matched parameter sets after correcting the shot gun moving speeds. A curvature radius calculation equation is fitted according to the parameter sets, the curvature radii corresponding to the parameter sets are calculated, and the corresponding predicted arc height values are calculated. The deep learning network model is trained according to the first predicted arc height values of the corrected first parameter sets and the real arc height values of the second parameter sets, and the mechanical shot forming arc height value prediction model is obtained. The technical scheme can solve the technical problem that actual measurement data is difficult to obtain in a large number of real experiments, and has the characteristics of low development cost, convenient and efficient use.

Description

Technical Field

[0001] This invention relates to the field of process engineering technology, and in particular to the training of a prediction model for the forming arc height of mechanical shot peening, and the prediction methods, equipment, media, and products for the forming arc height of mechanical shot peening. Background Technology

[0002] Shot peening, as an advanced and efficient precision forming technology for metal components, has become an indispensable core process in modern aerospace manufacturing due to its outstanding advantages such as high forming accuracy, good surface quality, and excellent fatigue resistance. It has been widely applied in the engineering of key metal components such as wing panels, fuselage skin, and stiffeners. With the continuous improvement of aircraft performance and increasingly complex structures, the requirements for the dimensional accuracy, surface integrity, and structural reliability of metal parts are constantly increasing. Therefore, the stability, controllability, and intelligence level of shot peening directly affect the manufacturing quality, production efficiency, and service safety of aircraft products.

[0003] Currently, the determination and optimization of arc height values ​​in mechanical shot peening still generally adopts a technical approach that combines traditional data-driven and experience-based decision-making. In actual production, process engineers typically make preliminary predictions of parameter windows based on historical experience, process manuals, or numerical simulations to determine the approximate ranges of key parameters such as shot peening pressure, shot flow rate, shot gun movement speed, shot type, nozzle angle and height, material condition, and coverage. These parameters are then corrected and adjusted through multiple real-world test peening experiments to finally obtain the combination of process parameters that meets the forming requirements.

[0004] Existing traditional methods relying on experience and repeated trials have significant technical shortcomings and engineering limitations. Mechanical shot peening involves numerous process parameters and complex coupling relationships; changes in a single parameter or even a small number of parameters can significantly affect the final forming effect. Relying on experience-based estimation and trial-and-error experiments makes it difficult to accurately grasp the inherent mapping laws. At the same time, extensive real-world testing and simulation iterations are not only time-consuming and severely slow down the development cycle, but also consume a large amount of materials, equipment, and manpower. Especially in the production of multi-variety, small-batch aerospace components, the problems of low efficiency, high cost, and difficulty in ensuring consistency are even more prominent, making it difficult to meet the development needs of modern aircraft manufacturing for high efficiency, precision, and low cost. Summary of the Invention

[0005] This invention provides a training method, equipment, medium, and product for predicting the forming arc height of mechanical shot peening, in order to solve the problems of relying on experience, large number of experiments, low efficiency, and high cost in calculating the forming arc height of mechanical shot peening.

[0006] According to one aspect of the present invention, a method for training a mechanical shot peening forming arc height prediction model is provided, comprising: Multiple parameter sets are constructed, and each parameter set is divided into a first group and a second group. Each parameter set includes a set shot peening air pressure value, a set shot peening coverage rate, and a set shot peening test piece thickness. Simulation experiments were conducted using the first parameter sets in the first group to obtain the first spray gun moving speed and simulated arc height values ​​for each first parameter set. Real experiments were conducted using the second parameter sets in the second group to obtain the second spray gun moving speed and real arc height values ​​for each second parameter set. Based on the first spray gun movement speed of each first parameter set and the second spray gun movement speed of each second parameter set, the movement speed of each first spray gun and the movement speed of each second spray gun are corrected, and the correction results are added to the matching first parameter set or second parameter set. Based on the corrected first parameter set and second parameter set, fit each parameter to be fitted in the preset curvature radius fitting formula to obtain the curvature radius calculation equation, and calculate the first curvature radius of each first parameter set and the second curvature radius of each second parameter set according to the curvature radius calculation equation. Calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and correct the first predicted arc height value of each first parameter set based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value. Based on the first predicted arc height value of each first parameter set after correction, and the actual arc height value of each second parameter set, the deep learning network model is trained to obtain the mechanical shot peening forming arc height prediction model.

[0007] According to another aspect of the present invention, a method for predicting the forming arc height of mechanical shot peening is also provided, comprising: Obtain the set of process parameters to be predicted, wherein the set of process parameters includes: target shot peening gas pressure value, target shot peening coverage, and target shot peening test piece thickness; The process parameter set is input into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method described in any one of the embodiments of the present invention; Obtain the mechanical shot peening forming arc height prediction results output by the mechanical shot peening forming arc height prediction model, which are matched with the process parameter set.

[0008] According to another aspect of the present invention, a training device for predicting the height of the forming arc of mechanical shot peening is also provided, comprising: The experimental parameter construction module is used to construct multiple parameter sets and divide each parameter set into a first group and a second group. Each parameter set includes the set shot peening air pressure value, the set shot peening coverage rate, and the set shot peening test piece thickness. The experiment execution module is used to perform simulation experiments using each first parameter set in the first group to obtain the first spray gun moving speed and simulated arc height value of each first parameter set, and to perform real experiments using each second parameter set in the second group to obtain the second spray gun moving speed and real arc height value of each second parameter set. The experimental parameter correction module is used to correct the movement speed of each first spray gun and each second spray gun based on the movement speed of the first spray gun in each first parameter set and the movement speed of the second spray gun in each second parameter set, and add the correction results to the matching first parameter set or second parameter set. The curvature radius calculation module is used to fit each parameter to be fitted in the preset curvature radius fitting formula based on the corrected first parameter set and each second parameter set, to obtain the curvature radius calculation equation, and to calculate the first curvature radius of each first parameter set and the second curvature radius of each second parameter set based on the curvature radius calculation equation. The predicted arc height correction module is used to calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and to correct the first predicted arc height value of each first parameter set based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value. The model training module is used to train the deep learning network model based on the first predicted arc height values ​​of each first parameter set after correction and the actual arc height values ​​of each second parameter set, so as to obtain the mechanical shot peening forming arc height prediction model.

[0009] According to another aspect of the present invention, a mechanical shot peening forming arc height prediction device is also provided, comprising: The process parameter set acquisition module is used to acquire the process parameter set to be predicted, wherein the process parameter set includes: target shot peening gas pressure value, target shot peening coverage, and target shot peening test piece thickness. The arc height prediction module is used to input the set of process parameters into the mechanical shot peening arc height prediction model trained by the mechanical shot peening arc height prediction model training method described in any one of the embodiments of the present invention. The arc height acquisition module is used to acquire the arc height prediction results of mechanical shot peening forming that are matched with the set of process parameters, output by the mechanical shot peening forming arc height prediction model.

[0010] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the mechanical shot peening forming arc height prediction model training method according to any embodiment of the present invention, or to perform the mechanical shot peening forming arc height prediction method according to any embodiment of the present invention.

[0011] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the mechanical shot peening forming arc height prediction model training method according to any embodiment of the present invention, or to implement the mechanical shot peening forming arc height prediction method according to any embodiment of the present invention.

[0012] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of the mechanical shot peening forming arc height prediction model training method or the mechanical shot peening forming arc height prediction method as described in any embodiment of the present invention.

[0013] This invention, through the aforementioned technical solution, constructs multiple parameter sets according to the requirements of mechanical shot peening forming experiments. Simulation and real experiments are conducted by designing different shot peening pressure values, shot peening coverage rates, and shot peening specimen thicknesses to obtain the corresponding spray gun movement speed and experimentally measured arc height values. Using the parameter sets used in the mechanical shot peening forming simulation experiment and the obtained experimental data, the corresponding initial coverage growth coefficient is calculated using the coverage calculation formula. A coverage growth coefficient prediction equation is obtained by fitting a pre-constructed coverage growth coefficient fitting formula based on the obtained data. After calculating the coverage growth coefficients corresponding to each first parameter set and each second parameter set using this equation, a parameter set is further obtained for fitting the curvature radius fitting formula, thereby fitting the curvature radius calculation equation. The curvature calculated according to this equation is ultimately used to calculate the corresponding predicted arc height value. After determining the maximum difference between all predicted arc height values ​​and the matched actual arc height values ​​in the real experiment, a truncated normal distribution of the difference can be further designed to determine the process fluctuation difference of each predicted arc height value. Finally, the training dataset for the deep learning network model is composed of each predicted arc height value corrected by adding the process fluctuation difference and the corresponding parameter set in the simulation experiment. The validation dataset is constructed from the parameter set in the real experiment and the actual measured arc height values. Under the constraints of a pre-constructed loss function, the mechanical shot peening forming arc height prediction model is finally trained. This technical solution combines mechanical shot peening forming simulation experiments and real experiments to correct and compensate for errors in various parameters during the experiment. This results in higher data rigor and better data utilization efficiency during the training of the mechanical shot peening forming arc height prediction model. It effectively solves the technical problem of difficulty in obtaining actual measurement data in a large number of real experiments, and also features low development cost and convenient and efficient use.

[0014] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a mechanical shot peening forming arc height prediction model training method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of a mechanical shot peening forming arc height prediction method provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the prediction results of the forming arc height value of mechanical shot peening applicable to Embodiment 2 of the present invention; Figure 4 This is a schematic diagram of a mechanical shot peening forming arc height prediction model training device provided in Embodiment 3 of the present invention; Figure 5 This is a schematic diagram of the structure of a mechanical shot peening forming arc height prediction device provided in Embodiment 3 of the present invention; Figure 6 This is a schematic diagram of the structure of an electronic device for implementing the mechanical shot peening forming arc height prediction model training method or the mechanical shot peening forming arc height prediction method of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] Example 1 Figure 1 This is a flowchart of a method for training a mechanical shot peening forming arc height prediction model according to Embodiment 1 of the present invention. This embodiment is applicable to situations where a deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model. This method can be executed by a mechanical shot peening forming arc height prediction model training device, which can be implemented in hardware and / or software and is generally configured in a computer device that performs the mechanical shot peening forming arc height prediction model training process. Figure 1 As shown, the method includes: S110. Construct multiple parameter sets and divide each parameter set into a first group and a second group.

[0020] Each parameter set includes the set shot peening pressure value, the set shot peening coverage, and the set shot peening test specimen thickness.

[0021] Understandably, before conducting mechanical shot peening forming experiments, a predetermined number of parameter sets can be constructed based on the experimental accuracy requirements. These parameter sets can include shot peening pressure values, shot peening coverage, and shot peening specimen thickness that meet the experimental needs. Subsequent experiments are then conducted using the experimental parameters from the constructed parameter sets. For example, at least 24 parameter sets can be constructed and divided into two groups, ensuring that each group contains at least 12 parameter sets. The shot peening pressure values ​​in each parameter set can be calculated using a minimum division unit of 0.5 Bar. The value is taken within the Bar range, and the shot peening coverage can be the minimum experimental value allowed by the experiment. The thickness of the obtained shot peening specimen is guaranteed to be consistent with the values ​​of all parameters in the first and second groups.

[0022] S120. Simulation experiments are conducted using the first parameter sets in the first group to obtain the first spray gun moving speed and simulated arc height values ​​for each first parameter set. Real experiments are conducted using the second parameter sets in the second group to obtain the second spray gun moving speed and real arc height values ​​for each second parameter set.

[0023] The first parameter set can refer to the parameter sets used in the mechanical shot peening simulation experiment, which include shot peening pressure, shot peening coverage, and shot peening specimen thickness. The second parameter set can refer to the parameter sets used in the mechanical shot peening simulation experiment, which also include shot peening pressure, shot peening coverage, and shot peening specimen thickness. The first spray gun movement speed can refer to the spray gun movement speed that meets the shot peening coverage condition in the first parameter set, obtained from simulation experiments based on the first parameter set. The second spray gun movement speed can refer to the spray gun movement speed that meets the shot peening coverage condition in the second parameter set, obtained from real experiments based on the second parameter set. The arc height value can refer to the vertical distance between the highest point of the curved surface of the shot peening specimen and the reference plane after bending deformation under shot peening in the mechanical shot peening experiment; it is a core quantitative indicator characterizing the shot peening effect and the degree of deformation.

[0024] Understandably, when conducting shot peening forming experiments, simulation experiments can be performed using all the first parameter sets in the constructed first group. After repeated trial experiments, the first spray gun movement speed that satisfies the specified minimum experimental value of shot peening coverage under different shot peening gas pressures for each first parameter set is obtained, and the corresponding simulated arc height values ​​are measured and recorded. Alternatively, real experiments can be performed using all the second parameter sets in the constructed second group. After repeated trial experiments, the second spray gun movement speed that satisfies the specified minimum experimental value of shot peening coverage under different shot peening gas pressures for each second parameter set is obtained, and the corresponding real arc height values ​​are measured and recorded. The first spray gun movement speed and matched simulated arc height values ​​corresponding to each first parameter set in the first group, and the second spray gun movement speed and matched real arc height values ​​corresponding to each second parameter set in the second group, can serve as dependent data for subsequent calculations.

[0025] S130. Based on the first spray gun movement speed of each first parameter set and the second spray gun movement speed of each second parameter set, correct the first spray gun movement speed and the second spray gun movement speed, and add the correction result to the matching first parameter set or second parameter set.

[0026] It is understandable that the first spray gun movement speeds obtained from experiments based on the first parameter sets, and the second spray gun movement speeds obtained from the second parameters, are obtained through repeated simulation experiments and real experiments, respectively. They have certain deviations and cannot be calculated from known parameter sets. Therefore, a series of calculations can be performed based on the relationship formula between the constructed spray gun movement speed and each parameter in the parameter set to solve for the corrected first spray gun movement speed corresponding to each first parameter set and the corrected second spray gun movement speed corresponding to each second parameter set. The obtained spray gun movement speeds are then integrated into the corresponding parameter sets for subsequent prediction of arc height values.

[0027] Optionally, the movement speeds of the first and second spray guns are corrected based on the movement speeds of the first spray guns in each first parameter set and the movement speeds of the second spray guns in each second parameter set, including: Calculation formula based on the first shot peening coverage parameter The first shot peening coverage rate in each first parameter set Based on the first spray gun movement speed V1 of each first parameter set, the first original coverage growth coefficient of each first parameter set is calculated. ; Calculation formula based on the second shot peening coverage parameter The second shot peening coverage in each set of second parameters Based on the second spray gun movement speed V2 of each second parameter set, the second original coverage growth coefficient of each second parameter set is calculated. ; Based on each first shot peening gas pressure value, the first original coverage growth coefficient matched with each first shot peening gas pressure value, each second shot peening gas pressure value, and the second original coverage growth coefficient matched with each second shot peening gas pressure value, a coverage growth coefficient prediction equation is fitted. Based on the coverage growth coefficient prediction equation, the first target coverage growth coefficient matching each first shot peening pressure value and the second target coverage growth coefficient matching each second shot peening pressure value are recalculated. Substitute each first shot peening air pressure value and the first target coverage growth coefficient matched with each first shot peening air pressure value back into the first shot peening coverage parameter calculation formula to solve for the corrected movement speed of each first spray gun. Substitute each second shot peening air pressure value, and the second target coverage growth coefficient matched with each second shot peening air pressure value, back into the second shot peening coverage parameter calculation formula to solve for the corrected movement speed of each second spray gun.

[0028] The first original coverage growth coefficient can refer to the relative growth rate of shot peening coverage under a unit first shot peening gas pressure intensity as a function of shot peening parameters, and is a key parameter for quantitatively characterizing the efficiency of shot peening coverage improvement. The first shot peening coverage can refer to the shot peening coverage under simulated experimental conditions matching the first shot peening gas pressure and the first spray gun's moving speed. The second original coverage growth coefficient can refer to the relative growth rate of shot peening coverage under a unit second shot peening gas pressure intensity as a function of shot peening parameters, and is a key parameter for quantitatively characterizing the efficiency of shot peening coverage improvement. The second shot peening coverage can refer to the shot peening coverage under real experimental conditions matching the second shot peening gas pressure and the second spray gun's moving speed. The first target coverage growth coefficient can refer to the key parameter used to characterize the efficiency of shot peening coverage improvement, calculated using the coverage growth coefficient prediction equation under a unit first shot peening gas pressure intensity. The second target coverage growth coefficient can refer to the key parameter used to characterize the efficiency of shot peening coverage improvement, calculated using the coverage growth coefficient prediction equation under a unit second shot peening gas pressure intensity.

[0029] Specifically, before correcting the moving speeds of the first and second spray guns, a pre-constructed formula for calculating the first shot peening coverage parameter can be used. Where V1 is the first shot peening pressure value in the first parameter set. The first shot peening coverage, The first original coverage growth coefficient can be calculated using the simulation experimental parameters of each first parameter set, according to the calculation formula for the first shot peening coverage parameter. Similarly, construct the formula for calculating the second shot peening coverage parameter. The second original coverage growth coefficient can be calculated using the actual experimental parameters from each second parameter set, according to the formula for calculating the second shot peening coverage parameter. Based on the first shot peening pressure values ​​and the calculated matching first original coverage growth coefficient, as well as the second shot peening pressure values ​​and the calculated matching second original coverage growth coefficient, a pre-constructed coverage growth coefficient fitting formula can be fitted to obtain a coverage growth coefficient prediction equation. Substituting the first shot peening pressure values ​​into the initial coverage growth coefficient prediction equation, the matching first target coverage growth coefficient can be recalculated. Similarly, the matching second target coverage growth coefficient for each second shot peening pressure value can be calculated. Substituting the first shot peening pressure values ​​and the matching first target coverage growth coefficient back into the first shot peening coverage parameter calculation formula, the corrected movement speed of each first spray gun can be calculated. Similarly, the corrected movement speed of each second spray gun can be calculated.

[0030] Optionally, based on each first shot peening gas pressure value, a first original coverage growth coefficient matching each first shot peening gas pressure value, each second shot peening gas pressure value, and a second original coverage growth coefficient matching each second shot peening gas pressure value, a coverage growth coefficient prediction equation is fitted, including: Construct an initial coverage growth coefficient fitting formula, wherein the coverage growth coefficient fitting formula includes multiple parameters to be fitted; Based on each first shot peening pressure value and the first original coverage growth coefficient matched with each first shot peening pressure value, each parameter to be fitted in the coverage growth coefficient fitting formula is fitted to obtain the initial coverage growth coefficient prediction equation. Substitute each first shot peening gas pressure value into the initial coverage growth coefficient prediction equation to calculate the first predicted coverage growth coefficient that matches each first shot peening gas pressure value. Substitute each second shot peening gas pressure value into the initial coverage growth coefficient prediction equation to calculate the second predicted coverage growth coefficient that matches each second shot peening gas pressure value. The coefficients of determination of the fitting equation are calculated by substituting each first shot peening gas pressure value, the first original coverage growth coefficient and the first predicted coverage growth coefficient matched with each first shot peening gas pressure value, and each second shot peening gas pressure value, the second original coverage growth coefficient and the second predicted coverage growth coefficient matched with each second shot peening gas pressure value into the fitting equation determination coefficient equation. If the coefficient of determination of the fitted equation meets the preset coefficient determination condition, the initial coverage growth coefficient prediction equation shall be used as the coverage growth coefficient prediction equation. Otherwise, update the parameters to be fitted in the coverage growth coefficient fitting formula, return and re-execute the operation of fitting each parameter to be fitted in the updated coverage growth coefficient fitting formula based on each first shot peening pressure value and the first original coverage growth coefficient matched with each first shot peening pressure value, and obtain the initial coverage growth coefficient prediction equation.

[0031] The initial coverage growth coefficient prediction equation can refer to the coverage growth coefficient prediction equation obtained by fitting a pre-constructed coverage growth coefficient fitting formula based on the first shot peening pressure values ​​used in the simulation experiment and the matched first original coverage growth coefficient, after confirming the coefficients in the fitting formula. The first predicted coverage growth coefficient can refer to the predicted coverage growth coefficient calculated by substituting the first shot peening pressure values ​​into the initial coverage growth coefficient prediction equation. The second predicted coverage growth coefficient can refer to the predicted coverage growth coefficient calculated by substituting the second shot peening pressure values ​​into the initial coverage growth coefficient prediction equation. The fitting equation determination coefficient can refer to the coefficient used to describe the cumulative error between each first original coverage growth coefficient, each second original coverage growth coefficient, and the corresponding first predicted coverage growth coefficient. The fitting equation determination coefficient equation can refer to the equation formula used to calculate the fitting equation determination coefficient based on each first original coverage growth coefficient, each second original coverage growth coefficient, and the corresponding first predicted coverage growth coefficient.

[0032] Specifically, in order to obtain the coverage growth coefficient prediction equation, an initial coverage growth coefficient fitting formula including multiple parameters to be fitted can be constructed first. ,in, C and P are parameters to be fitted, and P is the shot peening pressure. Let n be the initial coverage growth coefficient, and n be a constant greater than 1. The initial shot peening pressure values ​​and the matched initial coverage growth coefficient can be substituted into the initialized coverage growth coefficient fitting formula for fitting, thus determining the fitting parameters. And C, the initial coverage growth coefficient prediction equation is obtained. Substituting the first shot peening gas pressure value and the second shot peening gas pressure value into the initial coverage growth coefficient prediction equation, the first predicted coverage growth coefficient matching the first shot peening gas pressure value and the second predicted coverage growth coefficient matching the second shot peening gas pressure value can be calculated. The fitting equation and the coefficient of determination equation are then constructed. ,in, The coefficients of determination for the fitted equation. The shot peening pressure value The corresponding first or second original coverage growth coefficient, The shot peening pressure value The corresponding first or second predicted coverage growth factor, where N is the total number of simulation experiments and real experiments performed. This is the average of the first original coverage growth coefficients obtained in each simulation experiment and the second original coverage growth coefficients obtained in the real experiment.

[0033] The obtained first and second original coverage growth coefficients and their calculated averages are substituted into the coefficient of determination equation of the fitting equation to calculate the coefficient of determination. If the obtained coefficient of determination satisfies the preset coefficient judgment condition, the initial coverage growth coefficient prediction equation can be directly used as the coverage growth coefficient prediction equation. If it does not satisfy the condition, the coverage growth coefficient fitting formula can be reconstructed by increasing the number of terms of the parameters to be fitted in the formula, and the fitting calculation can be performed again according to the above logic until the calculated coefficient of determination satisfies the preset coefficient judgment condition, thus obtaining the coverage growth coefficient prediction equation. For example, if the preset coefficient judgment condition value is 0.95, and the coverage growth coefficient... If the value is greater than 0.95, the initial coverage growth coefficient prediction equation can be directly used as the coverage growth coefficient prediction equation. If this condition is not met, the coverage growth coefficient fitting formula can be reconstructed. Then, the coverage growth coefficient is recalculated and refitted to determine the result.

[0034] S140. Based on the corrected first parameter set and second parameter set, fit each parameter to be fitted in the preset curvature radius fitting formula to obtain the curvature radius calculation equation, and calculate the first curvature radius of each first parameter set and the second curvature radius of each second parameter set according to the curvature radius calculation equation.

[0035] The curvature radius fitting formula can refer to a formula constructed using multiple unknown fitting parameters, as well as shot peening pressure, shot peening coverage, and shot peening specimen thickness, to calculate the curvature radius. The curvature radius calculation equation can refer to an equation obtained by fitting and confirming each fitting parameter based on the shot peening pressure, shot peening coverage, and shot peening specimen thickness, which can be directly used to calculate the curvature radius. The first curvature radius can refer to the curvature radius calculated using the curvature radius calculation equation based on the shot peening pressure, shot peening coverage, and shot peening specimen thickness from the first parameter set. The second curvature radius can refer to the curvature radius calculated using the curvature radius calculation equation based on the shot peening pressure, shot peening coverage, and shot peening specimen thickness from the second parameter set.

[0036] Understandably, after updating and correcting the movement speeds of the first and second spray guns, corrected sets of first and second parameters can be obtained. From these, the parameters to be fitted in the preset radius of curvature fitting formula can be derived, thus obtaining the radius of curvature calculation equation. Substituting the shot peening pressure, shot peening coverage, and shot peening specimen thickness from the corrected first and second parameter sets into the radius of curvature calculation equation, the first radius of curvature corresponding to each first parameter set and the second radius of curvature corresponding to each second parameter set can be calculated.

[0037] Optionally, based on the corrected first parameter sets and second parameter sets, each parameter to be fitted in the preset radius of curvature fitting formula is fitted to obtain the radius of curvature calculation equation, including: Substitute the corrected simulated arc height values ​​from each first parameter set and the actual arc height values ​​from each second parameter set into the arc height calculation formula to solve for the corresponding radius of curvature. Based on the corrected sets of first and second parameters and the corresponding radii of curvature, the parameters to be fitted in the preset radius of curvature fitting formula are fitted to obtain the radius of curvature calculation equation.

[0038] Specifically, after obtaining the corrected first and second parameter sets, a formula for calculating the arc height, representing the quantitative relationship between the radius of curvature and the arc height, can be constructed. Where R is the radius of curvature. Let X be the arc height, and X be the arc length of the shot peening deformation zone. The simulated arc height values ​​from each of the first parameter sets, the actual arc height values ​​from each of the second parameter sets, and the arc length values ​​of the shot peening deformation zone measured during simulation and real experiments are substituted into the arc height calculation formula to inversely calculate the corresponding radius of curvature R. The preset radius of curvature fitting formula is: ,in, , , , , , , , Here are the parameters to be fitted: D is the shot peening air pressure value, V is the shot gun moving speed, P is the shot peening air pressure value, and R is the radius of curvature. Based on the corrected shot gun moving speed, shot peening air pressure value, and shot peening air pressure value in each of the first parameter set and each of the second parameter set after correction, as well as the corresponding radii of curvature, the fitting parameters can be determined by fitting the preset radius of curvature fitting formula, thereby obtaining the radius of curvature calculation equation.

[0039] S150. Calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and correct the first predicted arc height value of each first parameter set according to the difference between the second predicted arc height value of each second parameter set and the matched true arc height value.

[0040] The first predicted arc height value can refer to the arc height value calculated by substituting the first radius of curvature obtained from the radius of curvature calculation equation into the arc height value calculation formula. The second predicted arc height value can refer to the arc height value calculated by substituting the second radius of curvature obtained from the radius of curvature calculation equation into the arc height value calculation formula.

[0041] It is understandable that by substituting the relevant parameters of each first parameter set and each second parameter set into the radius of curvature calculation equation, the corresponding first radius of curvature and each second radius of curvature can be obtained. Then, the first predicted arc height value and the second predicted arc height value with a certain error can be calculated by the arc height value calculation formula. Based on the difference between the second predicted arc height value corresponding to each second parameter set in the real experiment and the actual arc height value obtained by actual measurement in the matching experiment, the first predicted arc height value of each first parameter set obtained in the simulation experiment can be corrected according to the preset correction strategy.

[0042] Optionally, a first predicted arc height value corresponding to each first radius of curvature and a second predicted arc height value corresponding to each second radius of curvature are calculated, and the first predicted arc height value of each first parameter set is corrected based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value, including: The maximum difference is determined based on the second predicted arc height value and the matched true arc height value of each second parameter set, and the proportion of the number of the second parameter sets corresponding to the maximum difference within the preset range of the nearest difference value is determined to the total number of each second parameter set. Based on the maximum difference and the proportion, a truncated normal distribution of the difference is designed. Based on the truncated normal distribution of the difference, a process fluctuation difference is generated to correct and match the first predicted arc height value of each first parameter set. After correction, the first predicted arc height value of each first parameter set with increased process fluctuation difference is obtained.

[0043] The truncated normal distribution refers to a continuous probability distribution obtained by limiting the difference between two sets of normally distributed random variables to a preset range of adjacent differences and then renormalizing it. It is a statistical distribution used to constrain the range of differences and eliminate extreme values. The process fluctuation difference can refer to the error compensation process parameter difference obtained from the first predicted arc height value based on the truncated normal distribution.

[0044] Specifically, in real mechanical shot peening experiments, there is an error between the second predicted arc height value obtained from the arc height calculation formula and the actual measured arc height value. The maximum difference can be determined by comparing the large differences between the second predicted arc height values ​​corresponding to each second parameter set and the matched actual arc height values. Based on the preset nearest neighbor difference, the number of second parameter sets corresponding to the difference between the second predicted arc height value and the matched true arc height value within the range corresponding to the maximum difference can be determined, thereby determining the proportion of each second parameter set to the total number of all second parameter sets. Therefore, it is possible to design a difference-truncation normal distribution. ,in, The arc height value represents the process fluctuation difference, and the distribution of this value satisfies the mean. , upper side The quantile point (i.e., the cutoff point) is Arc height value process fluctuation difference Greater than the maximum difference The probability can be expressed as Among them, the arc height value is the process fluctuation difference. The range of values ​​is Based on the constructed difference-truncated normal distribution, the first predicted arc height value for each first parameter set can be obtained. Corrected process fluctuation difference Thus, the first predicted arc height value after correction for process fluctuation difference is obtained. .

[0045] S160. Based on the first predicted arc height values ​​of each first parameter set after correction, and the actual arc height values ​​of each second parameter set, the deep learning network model is trained to obtain the mechanical shot peening forming arc height prediction model.

[0046] Understandably, the first parameter set in a mechanical shot peening simulation experiment can be designed to meet the required quantity of first parameter sets according to experimental needs, and the matching first predicted arc height values ​​can be calculated. For example, when constructing a specified quantity of first parameter sets for the simulation experiment, M shot peening pressure values, Q shot peening coverage rates, and H shot peening specimen thicknesses can be set to ensure that M... Q H meets the requirement of not less than 500 groups, where the minimum division unit for shot peening coverage data can be 0.1, with a value range of (0.1, 1], and the minimum division unit for shot peening specimen thickness data is 0.5 mm, with a value range of [1, 1]. Therefore, the deep learning network model can be trained based on the constructed first parameter set and the matched first predicted arc height values, and verified based on the second parameter sets and corresponding real arc height values ​​in the above real experiments, thereby obtaining a mechanical shot peening forming arc height prediction model that meets the verification conditions.

[0047] Optionally, based on the first predicted arc height values ​​of each corrected first parameter set and the true arc height values ​​of each second parameter set, a deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model, including: A training dataset is constructed based on each first parameter set and the corresponding corrected first predicted arc height value. A validation dataset is constructed based on each second parameter set and the corresponding true arc height value. A loss function for training the deep learning network model is constructed. Based on the training dataset, validation dataset, and deep learning network model training loss function, a pre-built deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model.

[0048] In this context, the training loss function of a deep learning network model can refer to the objective function constructed during the training process of the deep learning model based on the error between the model's predicted output and the actual experimental measurement of the arc height. This objective function is used to iteratively optimize the network parameters to minimize the prediction bias.

[0049] Specifically, a training dataset can be constructed based on the designed set of first parameters that meets the training requirements, and the calculated corrected first predicted arc height values, to train the deep learning network model. A validation dataset can be constructed based on the set of second parameters and the corresponding real arc height values ​​from real experiments, and the training loss function for the deep learning network model can be constructed as follows: ,in, , Y represents the predicted arc height value obtained by the deep learning network model based on the second set of input parameters, where Y is the actual arc height value. The constructed deep learning network model has an input layer with 3 neurons, corresponding to the three inputs: shot peening pressure, shot peening coverage, and shot peening specimen thickness. The memory hidden layer can be designed as 6 layers, each with 24 neurons. The hidden layers can be connected to a fully connected layer using the sigmoid activation function, resulting in a total of 24 neurons. The output layer consists of a single neuron, forming a single-output structure, and the output value is the arc height value. The parameters used for training the network are: batch size = 8, learning rate = 0.0002, and epochs = 400. The deep learning network model obtained after training meets the requirements and can be used as a prediction model for the arc height value of mechanical shot peening. The distribution of its prediction results is shown below. Figure 3 As shown, the X-axis represents the experiment number, and the Y-axis represents the arc height value (in mm).

[0050] This invention, through the aforementioned technical solution, constructs multiple parameter sets according to the requirements of mechanical shot peening forming experiments. Simulation and real experiments are conducted by designing different shot peening pressure values, shot peening coverage rates, and shot peening specimen thicknesses to obtain the corresponding spray gun movement speed and experimentally measured arc height values. Using the parameter sets used in the mechanical shot peening forming simulation experiment and the obtained experimental data, the corresponding initial coverage growth coefficient is calculated using the coverage calculation formula. A coverage growth coefficient prediction equation is obtained by fitting a pre-constructed coverage growth coefficient fitting formula based on the obtained data. After calculating the coverage growth coefficients corresponding to each first parameter set and each second parameter set using this equation, a parameter set is further obtained for fitting the curvature radius fitting formula, thereby fitting the curvature radius calculation equation. The curvature calculated according to this equation is ultimately used to calculate the corresponding predicted arc height value. After determining the maximum difference between all predicted arc height values ​​and the matched actual arc height values ​​in the real experiment, a truncated normal distribution of the difference can be further designed to determine the process fluctuation difference of each predicted arc height value. Finally, the training dataset for the deep learning network model is composed of each predicted arc height value corrected by adding the process fluctuation difference and the corresponding parameter set in the simulation experiment. The validation dataset is constructed from the parameter set in the real experiment and the actual measured arc height values. Under the constraints of a pre-constructed loss function, the mechanical shot peening forming arc height prediction model is finally trained. This technical solution combines mechanical shot peening forming simulation experiments and real experiments to correct and compensate for errors in various parameters during the experiment. This results in higher data rigor and better data utilization efficiency during the training of the mechanical shot peening forming arc height prediction model. It effectively solves the technical problem of difficulty in obtaining actual measurement data in a large number of real experiments, and also features low development cost and convenient and efficient use.

[0051] Example 2 Figure 2 This is a flowchart of a method for predicting the forming arc height of mechanical shot peening according to Embodiment 2 of the present invention. This embodiment is applicable to situations where the forming arc height of mechanical shot peening is predicted using a mechanical shot peening forming arc height prediction model. This method can be executed by a mechanical shot peening forming arc height prediction device, which can be implemented in hardware and / or software and is generally configured in a computer device that performs the process of predicting the forming arc height of mechanical shot peening. Figure 2 As shown, the method includes: S210. Obtain the set of process parameters to be predicted.

[0052] The process parameters include: target shot peening gas pressure, target shot peening coverage, and target shot peening test piece thickness.

[0053] Here, the process parameter set refers to the collection of relevant data used as input information for the mechanical shot peening arc height prediction model when predicting the arc height. The target shot peening pressure value refers to the shot peening pressure value set when predicting the arc height under specified mechanical shot peening experimental conditions. The target shot peening coverage rate refers to the shot peening coverage rate set when predicting the arc height under specified mechanical shot peening experimental conditions. The target shot peening specimen thickness refers to the thickness of the shot peening specimen selected when predicting the arc height under specified mechanical shot peening experimental conditions.

[0054] Understandably, before using the mechanical shot peening forming arc height prediction model to predict the mechanical shot peening forming arc height, the corresponding set of process parameters can be determined according to the experimental needs. The set of process parameters can include the target shot peening gas pressure value, target shot peening coverage, and target shot peening specimen thickness determined to fit the current experimental scenario.

[0055] S220. Input the set of process parameters into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method.

[0056] Understandably, in the process of training a pre-built deep learning network model to obtain a mechanical shot peening forming arc height prediction model, the training dataset includes data from the process parameter set obtained after a series of processing steps. The shot peening pressure, shot peening coverage, and shot peening test piece thickness data are designed as input information for the mechanical shot peening forming arc height prediction model. Therefore, when predicting the arc height, the process parameter set adapted to the current experimental scenario can be used as input information to the mechanical shot peening forming arc height prediction model for prediction processing.

[0057] S230. Obtain the mechanical shot peening forming arc height prediction results output by the mechanical shot peening forming arc height prediction model, which are matched with the process parameter set.

[0058] It is understandable that the process parameter set adapted to the current experimental scenario is input into the mechanical shot peening forming arc height prediction model and the output is the predicted arc height value. The obtained arc height value is matched with the target shot peening gas pressure value, target shot peening coverage, and target shot peening specimen thickness in the process parameter set. Then, different process parameter sets can be set according to the needs of different experimental scenarios, and the corresponding mechanical shot peening forming arc height prediction result can be obtained through the mechanical shot peening forming arc height prediction model.

[0059] This invention, through the above-described technical solution, enables the acquisition of a set of process parameters—including target shot peening pressure, target shot peening coverage, and target shot peening specimen thickness—suitable for the experimental scenario after training the mechanical shot peening forming arc height prediction model using the mechanical shot peening forming arc height prediction model training method. The acquired process parameter set is then input into the mechanical shot peening forming arc height prediction model for processing, outputting a mechanical shot peening forming arc height prediction result that matches the process parameter set. This technical solution, by inputting the process parameter set into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method, can improve the efficiency and accuracy of arc height prediction, is simple to operate and easy to implement, and effectively lowers the barrier to entry.

[0060] Example 3 Figure 4 This is a schematic diagram of a training device for a mechanical shot peening forming arc height prediction model provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes: an experimental parameter construction module 410, an experimental execution module 420, an experimental parameter correction module 430, a radius of curvature calculation module 440, a predicted arc height correction module 450, and a model training module 460.

[0061] The experimental parameter construction module 410 is used to construct multiple parameter sets and divide each parameter set into a first group and a second group. Each parameter set includes a set shot peening pressure value, a set shot peening coverage rate, and a set shot peening test piece thickness.

[0062] The experiment execution module 420 is used to perform simulation experiments using each first parameter set in the first group to obtain the first spray gun moving speed and simulated arc height value of each first parameter set, and to perform real experiments using each second parameter set in the second group to obtain the second spray gun moving speed and real arc height value of each second parameter set.

[0063] The experimental parameter correction module 430 is used to correct the movement speed of each first spray gun and each second spray gun according to the movement speed of the first spray gun in each first parameter set and the movement speed of the second spray gun in each second parameter set, and add the correction result to the matching first parameter set or second parameter set.

[0064] The radius of curvature calculation module 440 is used to fit each parameter to be fitted in the preset radius of curvature fitting formula according to the corrected first parameter set and each second parameter set, to obtain the radius of curvature calculation equation, and to calculate the first radius of curvature of each first parameter set and the second radius of curvature of each second parameter set according to the radius of curvature calculation equation.

[0065] The predicted arc height correction module 450 is used to calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and to correct the first predicted arc height value of each first parameter set based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value.

[0066] The model training module 460 is used to train the deep learning network model based on the first predicted arc height values ​​of each first parameter set after correction and the actual arc height values ​​of each second parameter set, so as to obtain the mechanical shot peening forming arc height prediction model.

[0067] This invention, through the aforementioned technical solution, constructs multiple parameter sets according to the requirements of mechanical shot peening forming experiments. Simulation and real experiments are conducted by designing different shot peening pressure values, shot peening coverage rates, and shot peening specimen thicknesses to obtain the corresponding spray gun movement speed and experimentally measured arc height values. Using the parameter sets used in the mechanical shot peening forming simulation experiment and the obtained experimental data, the corresponding initial coverage growth coefficient is calculated using the coverage calculation formula. A coverage growth coefficient prediction equation is obtained by fitting a pre-constructed coverage growth coefficient fitting formula based on the obtained data. After calculating the coverage growth coefficients corresponding to each first parameter set and each second parameter set using this equation, a parameter set is further obtained for fitting the curvature radius fitting formula, thereby fitting the curvature radius calculation equation. The curvature calculated according to this equation is ultimately used to calculate the corresponding predicted arc height value. After determining the maximum difference between all predicted arc height values ​​and the matched actual arc height values ​​in the real experiment, a truncated normal distribution of the difference can be further designed to determine the process fluctuation difference of each predicted arc height value. Finally, the training dataset for the deep learning network model is composed of each predicted arc height value corrected by adding the process fluctuation difference and the corresponding parameter set in the simulation experiment. The validation dataset is constructed from the parameter set in the real experiment and the actual measured arc height values. Under the constraints of a pre-constructed loss function, the mechanical shot peening forming arc height prediction model is finally trained. This technical solution combines mechanical shot peening forming simulation experiments and real experiments to correct and compensate for errors in various parameters during the experiment. This results in higher data rigor and better data utilization efficiency during the training of the mechanical shot peening forming arc height prediction model. It effectively solves the technical problem of difficulty in obtaining actual measurement data in a large number of real experiments, and also features low development cost and convenient and efficient use.

[0068] Optionally, the experimental parameter correction module 430 can be specifically used to: calculate the coverage parameter based on the first shot peening formula. The first shot peening coverage rate in each first parameter set Based on the first spray gun movement speed V1 of each first parameter set, the first original coverage growth coefficient of each first parameter set is calculated. The calculation formula based on the second shot peening coverage parameter. The second shot peening coverage in each set of second parameters Based on the second spray gun movement speed V2 of each second parameter set, the second original coverage growth coefficient of each second parameter set is calculated. Based on the first shot peening air pressure value, the first original coverage growth coefficient matching the first shot peening air pressure value, the second shot peening air pressure value, and the second original coverage growth coefficient matching the second shot peening air pressure value, a coverage growth coefficient prediction equation is fitted. Based on the coverage growth coefficient prediction equation, the first target coverage growth coefficient matching the first shot peening air pressure value and the second target coverage growth coefficient matching the second shot peening air pressure value are recalculated. The first shot peening air pressure value and the first target coverage growth coefficient matching the first shot peening air pressure value are substituted back into the first shot peening coverage parameter calculation formula to solve for the corrected movement speed of each first spray gun. The second shot peening air pressure value and the second target coverage growth coefficient matching the second shot peening air pressure value are substituted back into the second shot peening coverage parameter calculation formula to solve for the corrected movement speed of each second spray gun.

[0069] Optionally, the experimental parameter correction module 430 can also be specifically used for: constructing an initial coverage growth coefficient fitting formula, wherein the coverage growth coefficient fitting formula includes multiple parameters to be fitted. Based on each first shot peening pressure value and a first original coverage growth coefficient matching each first shot peening pressure value, each parameter to be fitted in the coverage growth coefficient fitting formula is fitted to obtain an initial coverage growth coefficient prediction equation. Substituting each first shot peening pressure value into the initial coverage growth coefficient prediction equation, a first predicted coverage growth coefficient matching each first shot peening pressure value is calculated. Substituting each second shot peening pressure value into the initial coverage growth coefficient prediction equation, a second predicted coverage growth coefficient matching each second shot peening pressure value is calculated. Substituting each first shot peening pressure value, the first original coverage growth coefficient matching each first shot peening pressure value, and the first predicted coverage growth coefficient, as well as each second shot peening pressure value, the second original coverage growth coefficient matching each second shot peening pressure value, and the second predicted coverage growth coefficient into the fitting equation determination coefficient equation, the fitting equation determination coefficient is calculated. If the coefficient of determination of the fitted equation meets the preset coefficient determination condition, the initial coverage growth coefficient prediction equation is used as the coverage growth coefficient prediction equation. Otherwise, the parameters to be fitted in the coverage growth coefficient fitting formula are updated, and the operation of fitting the updated coverage growth coefficient fitting formula to obtain the initial coverage growth coefficient prediction equation is re-executed based on each first shot peening pressure value and the first original coverage growth coefficient matched with each first shot peening pressure value.

[0070] Optionally, the radius of curvature calculation module 440 can be specifically used to: substitute the simulated arc height values ​​from each corrected first parameter set and the actual arc height values ​​from each second parameter set into the arc height calculation formula to solve for the corresponding radius of curvature. Based on the corrected first parameter sets, second parameter sets, and corresponding radii of curvature, fit each parameter to be fitted in the preset radius of curvature fitting formula to obtain the radius of curvature calculation equation.

[0071] Optionally, the arc height correction module 450 can be specifically used to: determine the maximum difference based on the second predicted arc height value and the matched actual arc height value of each second parameter set, and determine the proportion of the number of corresponding second parameter sets within the preset range of the maximum difference to the total number of each second parameter set. Based on the maximum difference and the proportion, a truncated normal distribution of the difference is designed; based on the truncated normal distribution of the difference, a process fluctuation difference is generated to correct and match the first predicted arc height value of each first parameter set; after correction, the first predicted arc height value of each first parameter set with increased process fluctuation difference is obtained.

[0072] Optionally, the model training module 460 can be specifically used to: construct a training dataset based on each first parameter set and the corresponding corrected first predicted arc height value; construct a validation dataset based on each second parameter set and the corresponding true arc height value; and construct a deep learning network model training loss function. Based on the training dataset, validation dataset, and deep learning network model training loss function, a pre-constructed deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model.

[0073] The mechanical shot peening forming arc height prediction model training device provided in this embodiment of the invention can execute the mechanical shot peening forming arc height prediction model training method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0074] Example 4 Figure 5 This is a schematic diagram of a mechanical shot peening forming arc height prediction device provided in Embodiment 4 of the present invention. Figure 5 As shown, the device includes: a process parameter set acquisition module 510, an arc height prediction module 520, and an arc height acquisition module 530.

[0075] The process parameter set acquisition module 510 is used to acquire the process parameter set to be predicted, wherein the process parameter set includes: target shot peening gas pressure value, target shot peening coverage, and target shot peening test piece thickness.

[0076] The arc height prediction module 520 is used to input the set of process parameters into the mechanical shot peening arc height prediction model trained by the mechanical shot peening arc height prediction model training method described in any one of the embodiments of the present invention.

[0077] The arc height acquisition module 530 is used to acquire the arc height prediction results of mechanical shot peening forming that are matched with the process parameter set, output by the mechanical shot peening forming arc height prediction model.

[0078] This invention, through the above-described technical solution, enables the acquisition of a set of process parameters—including target shot peening pressure, target shot peening coverage, and target shot peening specimen thickness—suitable for the experimental scenario after training the mechanical shot peening forming arc height prediction model using the mechanical shot peening forming arc height prediction model training method. The acquired process parameter set is then input into the mechanical shot peening forming arc height prediction model for processing, outputting a mechanical shot peening forming arc height prediction result that matches the process parameter set. This technical solution, by inputting the process parameter set into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method, can improve the efficiency and accuracy of arc height prediction, is simple to operate and easy to implement, and effectively lowers the barrier to entry.

[0079] The mechanical shot peening forming arc height prediction device provided in this embodiment of the invention can execute the mechanical shot peening forming arc height prediction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0080] Example 5 Figure 6 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0081] like Figure 6As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0082] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0083] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as training a mechanical shot peening forming arc high value prediction model or a mechanical shot peening forming arc high value prediction method.

[0084] In some embodiments, the mechanical peening forming arc high value prediction model training or mechanical peening forming arc high value prediction method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the mechanical peening forming arc high value prediction model training or mechanical peening forming arc high value prediction method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured by any other suitable means (e.g., by means of firmware) to perform the mechanical peening forming arc high value prediction model training method or the mechanical peening forming arc high value prediction method, wherein: The training method for the mechanical shot peening forming arc height prediction model includes: Multiple parameter sets are constructed, and each parameter set is divided into a first group and a second group. Each parameter set includes a set shot peening air pressure value, a set shot peening coverage rate, and a set shot peening test piece thickness. Simulation experiments were conducted using the first parameter sets in the first group to obtain the first spray gun moving speed and simulated arc height values ​​for each first parameter set. Real experiments were conducted using the second parameter sets in the second group to obtain the second spray gun moving speed and real arc height values ​​for each second parameter set. Based on the first spray gun movement speed of each first parameter set and the second spray gun movement speed of each second parameter set, the movement speed of each first spray gun and the movement speed of each second spray gun are corrected, and the correction results are added to the matching first parameter set or second parameter set. Based on the corrected first parameter set and second parameter set, fit each parameter to be fitted in the preset curvature radius fitting formula to obtain the curvature radius calculation equation, and calculate the first curvature radius of each first parameter set and the second curvature radius of each second parameter set according to the curvature radius calculation equation. Calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and correct the first predicted arc height value of each first parameter set based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value. Based on the first predicted arc height value of each first parameter set after correction, and the actual arc height value of each second parameter set, the deep learning network model is trained to obtain the mechanical shot peening forming arc height prediction model.

[0085] Furthermore, the method for predicting the forming arc height of mechanical shot peening includes: Obtain the set of process parameters to be predicted, wherein the set of process parameters includes: target shot peening gas pressure value, target shot peening coverage, and target shot peening test piece thickness; The process parameter set is input into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method described in any one of the embodiments of the present invention; Obtain the mechanical shot peening forming arc height prediction results output by the mechanical shot peening forming arc height prediction model, which are matched with the process parameter set.

[0086] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0087] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0088] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0089] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0090] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0091] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0092] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0093] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A training method for a mechanical shot peening forming arc height prediction model, characterized in that, include: Multiple parameter sets are constructed, and each parameter set is divided into a first group and a second group. Each parameter set includes a set shot peening air pressure value, a set shot peening coverage rate, and a set shot peening test piece thickness. Simulation experiments were conducted using the first parameter sets in the first group to obtain the first spray gun moving speed and simulated arc height values ​​for each first parameter set. Real experiments were conducted using the second parameter sets in the second group to obtain the second spray gun moving speed and real arc height values ​​for each second parameter set. Based on the first spray gun movement speed of each first parameter set and the second spray gun movement speed of each second parameter set, the movement speed of each first spray gun and the movement speed of each second spray gun are corrected, and the correction results are added to the matching first parameter set or second parameter set. Based on the corrected first parameter set and second parameter set, fit each parameter to be fitted in the preset curvature radius fitting formula to obtain the curvature radius calculation equation, and calculate the first curvature radius of each first parameter set and the second curvature radius of each second parameter set according to the curvature radius calculation equation. Calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature, and correct the first predicted arc height value of each first parameter set based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value. Based on the first predicted arc height value of each first parameter set after correction, and the actual arc height value of each second parameter set, the deep learning network model is trained to obtain the mechanical shot peening forming arc height prediction model.

2. The method according to claim 1, characterized in that, Based on the first spray gun movement speed of each first parameter set and the second spray gun movement speed of each second parameter set, the movement speeds of the first and second spray guns are corrected, including: Calculation formula based on the first shot peening coverage parameter The first shot peening coverage rate in each first parameter set Based on the first spray gun movement speed V1 of each first parameter set, the first original coverage growth coefficient of each first parameter set is calculated. ; Calculation formula based on the second shot peening coverage parameter The second shot peening coverage in each set of second parameters Based on the second spray gun movement speed V2 of each second parameter set, the second original coverage growth coefficient of each second parameter set is calculated. ; Based on each first shot peening gas pressure value, the first original coverage growth coefficient matched with each first shot peening gas pressure value, each second shot peening gas pressure value, and the second original coverage growth coefficient matched with each second shot peening gas pressure value, a coverage growth coefficient prediction equation is fitted. Based on the coverage growth coefficient prediction equation, the first target coverage growth coefficient matching each first shot peening pressure value and the second target coverage growth coefficient matching each second shot peening pressure value are recalculated. Substitute each first shot peening air pressure value and the first target coverage growth coefficient matched with each first shot peening air pressure value back into the first shot peening coverage parameter calculation formula to solve for the corrected movement speed of each first spray gun. Substitute each second shot peening air pressure value, and the second target coverage growth coefficient matched with each second shot peening air pressure value, back into the second shot peening coverage parameter calculation formula to solve for the corrected movement speed of each second spray gun.

3. The method according to claim 2, characterized in that, Based on each first shot peening gas pressure value, a first original coverage growth coefficient matching each first shot peening gas pressure value, each second shot peening gas pressure value, and a second original coverage growth coefficient matching each second shot peening gas pressure value, a coverage growth coefficient prediction equation is fitted, including: Construct an initial coverage growth coefficient fitting formula, wherein the coverage growth coefficient fitting formula includes multiple parameters to be fitted; Based on each first shot peening pressure value and the first original coverage growth coefficient matched with each first shot peening pressure value, each parameter to be fitted in the coverage growth coefficient fitting formula is fitted to obtain the initial coverage growth coefficient prediction equation. Substitute each first shot peening gas pressure value into the initial coverage growth coefficient prediction equation to calculate the first predicted coverage growth coefficient that matches each first shot peening gas pressure value. Substitute each second shot peening gas pressure value into the initial coverage growth coefficient prediction equation to calculate the second predicted coverage growth coefficient that matches each second shot peening gas pressure value. The coefficients of determination of the fitting equation are calculated by substituting each first shot peening gas pressure value, the first original coverage growth coefficient and the first predicted coverage growth coefficient matched with each first shot peening gas pressure value, and each second shot peening gas pressure value, the second original coverage growth coefficient and the second predicted coverage growth coefficient matched with each second shot peening gas pressure value into the fitting equation determination coefficient equation. If the coefficient of determination of the fitted equation meets the preset coefficient determination condition, the initial coverage growth coefficient prediction equation shall be used as the coverage growth coefficient prediction equation. Otherwise, update the parameters to be fitted in the coverage growth coefficient fitting formula, return and re-execute the operation of fitting each parameter to be fitted in the updated coverage growth coefficient fitting formula based on each first shot peening pressure value and the first original coverage growth coefficient matched with each first shot peening pressure value, and obtain the initial coverage growth coefficient prediction equation.

4. The method according to any one of claims 1-3, characterized in that, Based on the corrected sets of first and second parameters, the parameters to be fitted in the preset radius of curvature fitting formula are obtained, resulting in the radius of curvature calculation equation, including: Substitute the corrected simulated arc height values ​​from each first parameter set and the actual arc height values ​​from each second parameter set into the arc height calculation formula to solve for the corresponding radius of curvature. Based on the corrected sets of first and second parameters and the corresponding radii of curvature, the parameters to be fitted in the preset radius of curvature fitting formula are fitted to obtain the radius of curvature calculation equation.

5. The method according to any one of claims 1-3, characterized in that, Calculate the first predicted arc height value corresponding to each first radius of curvature and the second predicted arc height value corresponding to each second radius of curvature. Then, based on the difference between the second predicted arc height value of each second parameter set and the matched true arc height value, correct the first predicted arc height value of each first parameter set, including: The maximum difference is determined based on the second predicted arc height value and the matched true arc height value of each second parameter set, and the proportion of the number of the second parameter sets corresponding to the maximum difference within the preset range of the nearest difference value is determined to the total number of each second parameter set. Based on the maximum difference and the proportion, a truncated normal distribution of the difference is designed. Based on the truncated normal distribution of the difference, a process fluctuation difference is generated to correct and match the first predicted arc height value of each first parameter set. After correction, the first predicted arc height value of each first parameter set with increased process fluctuation difference is obtained.

6. The method according to any one of claims 1-3, characterized in that, Based on the corrected predicted arc height values ​​of each first parameter set and the actual arc height values ​​of each second parameter set, the deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model, including: A training dataset is constructed based on each first parameter set and the corresponding corrected first predicted arc height value. A validation dataset is constructed based on each second parameter set and the corresponding true arc height value. A loss function for training the deep learning network model is constructed. Based on the training dataset, validation dataset, and deep learning network model training loss function, a pre-built deep learning network model is trained to obtain a mechanical shot peening forming arc height prediction model.

7. A method for predicting the forming arc height of mechanical shot peening, characterized in that, include: Obtain the set of process parameters to be predicted, wherein the set of process parameters includes: target shot peening gas pressure value, target shot peening coverage, and target shot peening test piece thickness; The process parameter set is input into the mechanical shot peening forming arc height prediction model trained by the mechanical shot peening forming arc height prediction model training method as described in any one of claims 1-6; Obtain the mechanical shot peening forming arc height prediction results output by the mechanical shot peening forming arc height prediction model, which are matched with the process parameter set.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the mechanical shot peening forming arc height prediction model training method or the mechanical shot peening forming arc height prediction method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the mechanical shot peening forming arc height prediction model training method or the mechanical shot peening forming arc height prediction method as described in any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the mechanical shot peening forming arc height prediction model training method or the mechanical shot peening forming arc height prediction method according to any one of claims 1-7.

Images