Zero-stress test block model obtaining method and device, medium and equipment

Abstract

The invention discloses a zero-stress test block model obtaining method and device, a medium and equipment, and relates to the technical field of nondestructive detection.The method comprises the steps that ultrasonic measurement is conducted on a solid test block under different stress levels, ultrasonic propagation signals are collected, training data are formed according to corresponding residual stress, and a stress prediction model is constructed; the model can correspondingly output the characteristics of the ultrasonic propagation signal according to the input stress value, finally, zero stress is used as input, the characteristics of the ultrasonic propagation signal under the stress level are output, the zero-stress test block model without the solid test block is constructed, the tedious process of solid test block preparation can be avoided, and the production efficiency is improved. In the subsequent detection, comparison can be completed only by calling corresponding electronic test block data, the detection efficiency is improved, the electronic test block restores the zero stress state as far as possible through the neural network model, the detection error introduced by solid test block preparation is avoided, the detection precision is improved, and the effect of ultrasonic residual stress measurement is improved.

Description

Technical Field

[0001] This application relates to the field of nondestructive testing technology, specifically to a method, apparatus, medium, and equipment for obtaining a zero-stress test block model. Background Technology

[0002] In the manufacturing of high-end equipment such as aerospace and defense weaponry, residual stress has a profound impact on the geometric accuracy, service performance, and reliability of components. Specifically, the initial residual stress inside the blank after forming (generated during casting / forging / additive manufacturing) breaks the equilibrium during material removal, creating a stress gradient that causes elastic recovery deformation of the component. During machining, the thermo-mechanical coupling effect of cutting introduces machining residual stress, which superimposes with the original stress field of the blank to form a complex stress field. This stress field gradually releases its equilibrium under the load of static loading, inducing warping, torsion, and other deformations in thin-walled structures, leading to parts exceeding tolerance limits or being scrapped during assembly.

[0003] To control the residual stress within thin-walled components to a low level, accurate measurement of residual stress is a necessary prerequisite for evaluating process optimization and stress relief effects. Ultrasonic measurement methods based on the acoustoelastic principle are widely used for residual stress measurement and evaluation in production sites due to their good field applicability, portability, fast measurement speed, and low cost. However, because this technology uses physical test blocks, each measurement of a new batch of products requires the preparation of new test blocks and certain stress relief processes. This results in a long preparation cycle, high costs for verifying stress relief effects, and the residual stress measured by acoustic time calibration of the prepared test blocks is only a relative value; the test blocks themselves may still have some degree of tensile / compressive residual stress, which cannot effectively guide subsequent risk assessment. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, medium, and equipment for obtaining zero-stress test block models, aiming to solve the problem of poor performance of ultrasonic measurement of residual stress in the prior art.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, embodiments of this application provide a method for obtaining a zero-stress test block model, applied to ultrasonic residual stress measurement, comprising the following steps: Based on the target measurement object, a target stress prediction model is determined. The target stress prediction model is trained based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value. Using zero stress as input, the characteristics of ultrasonic propagation signals under zero stress are output by the target stress prediction model to obtain a zero-stress test block model.

[0006] In one possible implementation of the first aspect, before determining the target stress prediction model based on the target measurement object, the method further includes: Ultrasonic propagation signals were collected from the state measurement area of ​​the target physical specimen under different stress levels; the different stress levels included the initial zero stress level. Based on the ultrasonic propagation signal, the residual stress in the state measurement area under different stress levels is obtained; Using the residual stress in the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, a target stress prediction model is trained to obtain the model.

[0007] In one possible implementation of the first aspect, before training the target stress prediction model using the residual stress of the state measurement region at different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the measurement results of the blind hole measurement method on the target physical specimen, the residual stress under the initial zero stress level is corrected to obtain the initial calibration stress of the state measurement area. Using the residual stress in the state measurement region under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, a target stress prediction model is trained and obtained, including: Using the residual stress and initial calibration stress of the state measurement area under different stress levels as inputs and the characteristics of the ultrasonic propagation signal as outputs, a target stress prediction model is trained to obtain the model.

[0008] In one possible implementation of the first aspect, the residual stress in the state measurement region at different stress levels is obtained based on the ultrasonic propagation signal, including: Based on the ultrasonic propagation signal, the propagation sound time of the state measurement area under different stress levels is obtained; among which, the propagation sound time includes the propagation sound time under the initial zero stress level; The residual stress of the state measurement region under different stress levels is obtained by comparing the propagation time of sound in the state measurement region under different stress levels with the propagation time of sound in the initial zero stress level.

[0009] In one possible implementation of the first aspect, before obtaining the residual stress of the state measurement region at different stress levels based on the propagation time of the acoustic sound in the state measurement region at different stress levels and the propagation time of the acoustic sound in the initial zero stress level, the method further includes: Based on the different grain size states of the target physical specimen, the propagation time of sound under the initial zero stress level is compensated to obtain the propagation time of sound under zero stress. Based on the propagation time of sound in the state measurement region under different stress levels and the propagation time of sound in the initial zero stress level, the residual stress in the state measurement region under different stress levels is obtained, including: The residual stress in the state measurement area under different stress levels is obtained by comparing the propagation time of sound in the state measurement area under different stress levels with the propagation time of sound at zero stress under the initial zero stress level.

[0010] In one possible implementation of the first aspect, before training the target stress prediction model using the residual stress of the state measurement region at different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the sinusoidal wave signal of the ultrasonic propagation signal, the amplitude, frequency, and phase parameters are extracted to obtain the characteristics of the ultrasonic propagation signal.

[0011] In one possible implementation of the first aspect, a zero-stress test block model is obtained by using zero stress as input and utilizing the characteristics of the ultrasonic propagation signal under zero stress output by the target stress prediction model, including: Using zero stress as input, the amplitude and phase of the ultrasonic propagation signal under zero stress are output using the target stress prediction model; Based on the amplitude and phase of the ultrasonic propagation signal under zero stress, the target sinusoidal signal is plotted to obtain a zero-stress test block model.

[0012] Secondly, embodiments of this application provide a zero-stress test block model acquisition device, applied to ultrasonic residual stress measurement, comprising: The determination module is used to determine the target stress prediction model based on the target measurement object. The target stress prediction model is trained based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value. The module is used to obtain the characteristics of the ultrasonic propagation signal under zero stress by using the target stress prediction model as input and outputting the model.

[0013] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when loaded and executed by a processor, implements the method for obtaining a zero-stress test block model as provided in any of the first aspects above.

[0014] Fourthly, embodiments of this application provide an electronic device, including a processor and a memory, wherein, Memory is used to store computer programs; The processor is used to load and execute a computer program to cause the electronic device to perform the method for obtaining a zero-stress test block model as provided in any of the first aspects above.

[0015] Compared with the prior art, the beneficial effects of this application are: This application proposes a method, apparatus, medium, and device for obtaining a zero-stress test block model. The method includes: determining a target stress prediction model based on the target measurement object; wherein the target stress prediction model is trained based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical test block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value; taking zero stress as input, using the target stress prediction model to output the ultrasonic propagation signal characteristics under zero stress, thereby obtaining a zero-stress test block model. This application utilizes ultrasonic measurements of physical test blocks under different stress levels to collect ultrasonic propagation signals. Based on the corresponding residual stress, training data is generated to construct a stress prediction model. This model can output the characteristics of the ultrasonic propagation signal corresponding to the input stress value. Finally, using zero stress as input, it outputs the characteristics of the ultrasonic propagation signal at that stress level, constructing a zero-stress test block model without a physical test block. This avoids the cumbersome process of preparing physical test blocks. Subsequent testing only requires calling the corresponding electronic test block data for comparison, improving testing efficiency. Furthermore, the electronic test block, through a neural network model, restores the zero-stress state as closely as possible, avoiding detection errors introduced by physical test block preparation, improving detection accuracy, and enhancing the effect of ultrasonic measurement of residual stress. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of this application; Figure 2 A flowchart illustrating the method for obtaining a zero-stress test block model provided in this application embodiment; Figure 3 A schematic diagram of the ultrasonic measurement principle in the method for obtaining a zero-stress test block model provided in the embodiments of this application; Figure 4 A schematic diagram of the ultrasonic acquisition system in the method for obtaining a zero-stress test block model provided in the embodiments of this application; Figure 5 A schematic diagram illustrating the principle of the blind hole measurement method in the zero-stress test block model acquisition method provided in the embodiments of this application; Figure 6 This is a schematic diagram of the range measured by the ultrasonic testing method in the method for obtaining a zero-stress test block model provided in the embodiments of this application; Figure 7 A schematic diagram of the module for obtaining a zero-stress test block model provided in the embodiments of this application; The markings in the diagram are: 101-processor, 102-communication bus, 103-network interface, 104-user interface, 105-memory, 1-base, 2-column, 3-target physical test block, 4-weight, 5-ultrasonic testing wedge. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0018] See attached document Figure 1 , attached Figure 1 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the embodiments of this application. The electronic device may include: a processor 101, such as a central processing unit (CPU), a communication bus 102, a user interface 104, a network interface 103, and a memory 105. The communication bus 102 is used to realize the connection and communication between these components. The user interface 104 may include a display screen and an input unit such as a keyboard. Optionally, the user interface 104 may also include a standard wired interface and a wireless interface. The network interface 103 may optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface). The memory 105 may be a storage device independent of the aforementioned processor 101. The memory 105 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as at least one disk storage device. The processor 101 may be a general-purpose processor, including a central processing unit, a network processor, etc., or it may be a digital signal processor, an application-specific integrated circuit, a field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component.

[0019] Those skilled in the art will understand that the appendix Figure 1 The structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0020] As attached Figure 1 As shown, the memory 105, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a zero-stress test block model acquisition device.

[0021] In the appendix Figure 1In the electronic device shown, the network interface 103 is mainly used for data communication with the network server; the user interface 104 is mainly used for data interaction with the user; the processor 101 and the memory 105 in this application can be set in the electronic device. The electronic device calls the zero-stress test block model acquisition device stored in the memory 105 through the processor 101 and executes the zero-stress test block model acquisition method provided in the embodiment of this application.

[0022] In the manufacturing of high-end equipment such as aerospace and defense weaponry, residual stress has a profound impact on the geometric accuracy, service performance, and reliability of components. Specifically, the initial residual stress inside the blank after forming (generated during casting / forging / additive manufacturing) breaks the equilibrium during material removal, creating a stress gradient that causes elastic recovery deformation of the component. During machining, the thermo-mechanical coupling effect of cutting introduces machining residual stress, which superimposes with the original stress field of the blank to form a complex stress field. This stress field gradually releases its equilibrium under the load of static loading, inducing warping, torsion, and other deformations in thin-walled structures, leading to parts exceeding tolerance limits or being scrapped during assembly.

[0023] To control the residual stress inside thin-walled components to a low level, common methods include iterative optimization of process parameters for key machining operations. This involves matching appropriate cutting speeds, feed rates, depths of cut, and machining paths to reduce the amplitude of residual stress after machining. Alternatively, stress relief treatments can be implemented during the blanking or roughing stages to reduce and homogenize the distribution of residual stress. Therefore, accurate measurement of residual stress is a necessary prerequisite for evaluating the effectiveness of process optimization and stress relief.

[0024] Ultrasonic measurement methods based on the principle of acoustoelasticity are widely used in production sites for residual stress measurement and evaluation due to their good field applicability, portability, fast measurement speed, and low cost. For example, patent application CN116907709 discloses a multi-directional electromagnetic ultrasonic stress detection device. This device uses guide wheels to detect stress on multiple machined and unmachined surfaces at different locations on the component under test. A necessary step is to pre-calibrate the zero-stress acoustic time, that is, using the parameters of a relatively zero-stress specimen as a benchmark, calibrating the zero-stress specimen to obtain the zero-stress acoustic time, and then calculating the residual stress.

[0025] Similarly, patent application CN116698268A discloses a method for detecting the pressure (stress) of a liquid medium inside a pipe based on the propagation time of ultrasonic transverse waves. A necessary prerequisite is selecting a "zero" stress region on the pipe wall to calibrate the sound wave propagation time, and then comparing the sound wave propagation time of the measured region to calculate the (pressure) stress. Patent application CN105158342A discloses a water-immersion ultrasonic residual stress measuring device, explicitly requiring that the calibration sample and the test sample be of the same grade and specification, or that a portion of the material to be tested be taken as the calibration sample.

[0026] The current national standards for ultrasonic residual stress testing, GB / T32073-2015 "Non-destructive Testing of Residual Stress by Ultrasonic Critical Refraction Longitudinal Wave" and GB / T43900-2024 "Non-destructive Testing of Steel Products by Ultrasonic Testing of Torsional Residual Stress Distribution in Shaft Components," both specify that the calculation of residual stress involves treating the calibration area of ​​the zero-stress specimen as having a theoretical absolute stress of zero. The time difference between the propagation time of the sound in the calibration area and that in the tested area is calculated and multiplied by the tensile coefficient to obtain the residual stress data. The measured residual stress data is a relative value, indicating whether the measured area is greater than or less than the value in the calibration area of ​​the zero-stress specimen. To ensure the residual stress of the zero-stress specimen, pretreatment using the high-energy acoustic beam modulation or stress-relief annealing methods described in GB / T38811 should be employed.

[0027] The paper "Secondary Shot Peening Induced Stress and Deformation Engineering Prediction Method" focuses on the ultrasonic measurement of residual stress in aluminum alloy pre-stretched plates. It emphasizes that the zero-stress specimen is prepared by cutting a rectangular sample with a length of 100 mm, a width of 90 mm, and a thickness of 20 mm from the same batch of products of the pre-stretched plate being tested. The residual stress in the four edge areas is measured using the blind hole method until the maximum stress value is less than 50 MPa. The sample is then considered as a zero-stress specimen.

[0028] The above methods still have the following problems: 1) The materials used to prepare the zero-stress specimens must be from the same batch and grade as the material being tested. Each time a new grade or condition is measured, a new specimen needs to be prepared and subjected to a certain stress-relief process. The preparation cycle is long and the cost of verifying the stress-relief effect is high. 2) The residual stress measured by the acoustic calibration of the zero-stress specimen is a relative value. It only indicates how much higher or lower the stress in the measured area is than that in the calibration area of ​​the zero-stress specimen. If the zero-stress specimen itself has a certain amplitude of tensile / compressive residual stress, the measured value may be abnormal (exceeding the yield strength or approaching zero) if the stress in the measured area is too large or too small, which cannot effectively guide the subsequent deformation risk assessment. 3) Although the existing technology proposes to calibrate the initial stress of the zero-stress specimen using the destructive blind hole method, the measurement principle of the blind hole method is different from that of the ultrasonic method. The residual stress measurement value of the ultrasonic method is a scalar without direction, while the measurement value of the blind hole method is a two-dimensional vector in a plane calculated by three strain gauges, which also has a certain error. 4) Existing technologies all require zero stress to be a solid test block, which needs to be stored specially and will be affected by environmental loads to release its own stress slowly, causing calibration time drift.

[0029] To address the aforementioned problems in the existing technology, refer to the appendix. Figure 2 Based on the hardware device of the foregoing embodiments, embodiments of this application provide a method for obtaining a zero-stress test block model, applied to ultrasonic residual stress measurement, including the following steps: S10: Determine the target stress prediction model based on the target measurement object; wherein, the target stress prediction model is obtained by training based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value.

[0030] In the specific implementation process, the target measurement object is the component that needs residual stress detection, such as a thin-walled component of an aircraft. Based on the object to be measured, a corresponding stress prediction model is determined, namely the target stress prediction model. The target stress prediction model is based on a target physical specimen prepared from the same grade and batch of material as the target measurement object, and is obtained through the methods described in this application. Different stress prediction models are established for different measurement objects; during non-destructive testing, only the corresponding model needs to be selected.

[0031] The training data for the stress prediction model is obtained by measuring the residual stress at different stress levels and mapping it to the characteristics of the ultrasonic propagation signal. Through training, the stress prediction model can output the corresponding characteristics of the ultrasonic propagation signal based on the input stress value.

[0032] In one embodiment, before determining the target stress prediction model based on the target measurement object, the method further includes: Ultrasonic propagation signals were collected from the state measurement area of ​​the target physical specimen under different stress levels; the different stress levels included the initial zero stress level. Based on the ultrasonic propagation signal, the residual stress in the state measurement area under different stress levels is obtained; Using the residual stress in the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, a target stress prediction model is trained to obtain the model.

[0033] In practical implementation, the principle of ultrasonic residual stress measurement is based on the feedback of residual stress data from changes in the speed of sound propagation, assuming a linear relationship between propagation speed and stress within a certain range. (See attached...) Figure 3 As shown, the principle of ultrasonic measurement of residual stress in critically refracted longitudinal waves is illustrated using an example: Once the dimensions of the ultrasonic wedge are determined, the incident piezoelectric element and the receiving piezoelectric element emit, propagate, and receive the critically refracted longitudinal wave at a fixed distance. When the ultrasonic wedge is placed in the calibration area of ​​the component being measured, a propagation time of the ultrasonic wave will be obtained. t 0. Then, place the ultrasonic wedge in the measurement area of ​​the component being tested. Under the action of residual tensile / compressive stress, the propagation time of the sound will increase / decrease, thereby obtaining the measurement sound time. t .in accordance with The relative residual stress value was calculated. , K This is the tensile stress coefficient.

[0034] However, residual stress is not the only factor affecting the propagation speed of ultrasound in materials. Changes in material hardness, grain size, and temperature can also alter the sound propagation speed, introducing detection errors and affecting accuracy. Therefore, when selecting physical test blocks, it is essential to ensure they are from the same batch and grade as the material being tested, and to maintain consistent forming processes and heat treatment conditions. After sampling, the blocks can be prepared into a long beam shape for easy testing.

[0035] The ultrasonic acquisition system that can be used for ultrasonic acquisition of the target solid specimen is shown in the attached figure. Figure 4As shown, the system includes a base 1, a column 2, a target physical test block 3, weights 4, and an ultrasonic wedge 5. The base 1 is considered a rigid body, serving to fix and support the column 2. The column 2 is also considered a rigid body, fixed to the base 1 and connected to the target physical test block 3. One end of the target physical test block 3 is fixedly connected to the column 2, while the other end extends horizontally and is freely suspended, forming a cantilever beam structure. The target physical test block 3 has no specific size restrictions, only requiring its thickness to be greater than twice the stress detection depth and to form a cantilever beam structure after connecting to the base 1 and the column 2. The weights 4 are suspended below the free end of the target physical test block 3 via connecting ropes or other means, and their weight can only cause a small deflection deformation in the cantilever beam, i.e., the maximum deflection does not exceed one-five-hundredth of the length of the target physical test block 3. Alternatively, a force sensor combined with a hydraulic transmission device can be used to apply downward pressure to the target physical test block 3 above the free end. The ultrasonic wedge 5 is placed at the center of the target physical test block 3 to detect changes in the ultrasonic wave propagation signal. The specific operating principle of ultrasonic measurement is as follows: First, the base 1, column 2, and target specimen 3 are fixedly connected and constrained to form a cantilever beam structure. A weight 4 is applied downwards to the free end of the cantilever beam, resulting in a small deflection. The lower surface of the target specimen 3 is compressed, generating tensile stress to resist the compression, while the upper surface is stretched, generating compressive stress to resist the stretch.

[0036] Then, the maximum deformation (deflection) at the free end of the target physical specimen 3 is obtained using measuring tools such as dial indicators / height measuring instruments. From the formula The stress on the upper surface can be calculated, where E is the elastic modulus, L is the length of the target solid specimen 3, and c is half the thickness of the target solid specimen 3.

[0037] Finally, by continuously changing the weight of the applied weight 4, different stress changes can be generated within the small deflection elastic deformation range, and the deformation is very small. The entire long beam range can be regarded as a plane, which has no effect on ultrasonic testing (ignoring the change in sound wave propagation path under small deformation). By placing the ultrasonic wedge 5 at the center of the upper surface of the target physical test block 3, ultrasonic wave propagation signal data under different stress states can be collected.

[0038] Existing ultrasonic testing instruments primarily identify changes in ultrasonic propagation time by analyzing changes in the characteristic values ​​of the propagation signal. Taking the critical refraction longitudinal wave method as an example, one or two sinusoidal signals from the zero-stress test block calibration are taken as the headwaves. First, the sampled signals are processed using Lagrange difference / cubic difference / spline difference to improve sampling accuracy. Then, the headwave signal from the tested area is similarly acquired and processed using difference analysis. The acoustic time difference is calculated using a cross-correlation algorithm (considering that the critical refraction longitudinal wave at the tested location is similar to the critical refraction longitudinal wave under zero stress, with only a certain delay and identical other information) / peak algorithm (comparing the times at the peak values ​​of the two critical refraction longitudinal waves to calculate the acoustic time difference), thus solving for the acoustic time. Therefore, when collecting and recording ultrasonic propagation signal data under different stress states, the same steps are followed. The acoustic time without added weights is recorded as the zero-stress acoustic time. The sampling signal data is processed using a difference method to improve sampling accuracy. Each time a different weight is added, the corresponding stress is calculated based on the deflection, and the sampling signal data is recorded and processed using the difference method. That is, based on the ultrasonic propagation signal, the residual stress in the state measurement area under different stress levels is obtained, including: Based on the ultrasonic propagation signal, the propagation sound time of the state measurement area under different stress levels is obtained; among which, the propagation sound time includes the propagation sound time under the initial zero stress level; The residual stress of the state measurement region under different stress levels is obtained by comparing the propagation time of sound in the state measurement region under different stress levels with the propagation time of sound in the initial zero stress level.

[0039] The ultrasonic propagation signals under different stress states and their corresponding residual stress values ​​are obtained. Through supervised training, the stress prediction model can output the characteristics of the ultrasonic propagation signal according to the input stress value.

[0040] In one embodiment, before training a target stress prediction model using the residual stress in the state measurement region at different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the measurement results of the blind hole measurement method on the target physical specimen, the residual stress under the initial zero stress level is corrected to obtain the initial calibration stress of the state measurement area.

[0041] In practice, since the physical specimen may contain residual stress, it cannot be truly considered as a zero-stress state. Therefore, the aforementioned steps record this state as the initial zero-stress level. To more accurately assess the deformation risk caused by residual stress, the residual stress in the initial state is corrected.

[0042] The principle of the blind hole measurement method is as follows: Figure 5As shown, the blind hole method for measuring residual stress requires strain gauges to be arranged at certain intervals along three directions (0°, 45°, and 90°) in the measured area. After drilling, the corresponding strain gauges are collected. According to the formula:

[0043] Two principal stress vectors in the plane are obtained, where A represents strain release coefficient 1 and B represents strain release coefficient 2.

[0044] The measurement range of ultrasonic testing methods is shown in the appendix. Figure 6 The figure shows the average stress in the space occupied by the distance between the two probes and the propagation depth, which is a scalar. Therefore, it is necessary to convert the measured blind hole stress value into an equivalent stress related to yield deformation to evaluate the initial calibration stress level of the ultrasonic measurement area. According to GB / T32073-2015 "Nondestructive Testing of Residual Stress - Ultrasonic Critical Refraction Longitudinal Wave Test Method", the ultrasonic testing frequency... f With detection depth d The relationship between them is , This is a correction factor. To measure the true stress level on the upper surface of the target specimen 3, without the weight 4 loaded, holes were drilled at both ends of the probe placement position using the blind hole method. The drilling depth was the same as the detection depth for the corresponding ultrasonic testing frequency. The corresponding... Then, using the formula:

[0045] The von Mises equivalent stress at the two punching locations was calculated, and the average value was taken as the true stress in the undeformed state, thus completing the correction of the residual stress under the initial zero stress level.

[0046] Based on the aforementioned steps, using the residual stress in the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, a target stress prediction model is trained and obtained, including: Using the residual stress and initial calibration stress of the state measurement area under different stress levels as inputs and the characteristics of the ultrasonic propagation signal as outputs, a target stress prediction model is trained to obtain the model.

[0047] In one embodiment, before obtaining the residual stress of the state measurement region at different stress levels based on the propagation sound time of the state measurement region at different stress levels and the propagation sound time at the initial zero stress level, the method further includes: Based on the different grain size states of the target physical specimen, the propagation time of sound under the initial zero stress level is compensated to obtain the propagation time of sound under zero stress.

[0048] In practical implementation, traditional zero-stress test blocks are usually only sampled from the edge bosses of the blank, or processed from sheet metal of the same grade, and then heat-treated to relieve stress before being considered as zero-stress test blocks. However, in the manufacturing of large-size, high-precision aerospace thin-walled parts, free forgings, die forgings, and castings are often used as blanks for machining. During the blank forming process, the uneven distribution of temperature gradients and the influence of process parameters result in non-uniform grain structure in the depth direction. Therefore, after removing the surface and near-surface allowances, the grain size of the core region differs from that of the zero-stress test block sampled from the surface / boob, introducing measurement errors. For this type of testing scenario, solid test blocks can be prepared by sampling from sheet metal of the same grade, and then subjected to different heat aging treatments to achieve different grain size states. The grain size can be determined through metallographic analysis. By introducing a grain size compensation algorithm for sound propagation velocity, the propagation sound time under the initial zero-stress level can be compensated to obtain the zero-stress sound time.

[0049] Based on the aforementioned steps, the residual stress of the state measurement region at different stress levels is obtained by comparing the propagation time of sound in the state measurement region at different stress levels with the propagation time of sound in the initial zero stress level, including: The residual stress in the state measurement area under different stress levels is obtained by comparing the propagation time of sound in the state measurement area under different stress levels with the propagation time of sound at zero stress under the initial zero stress level.

[0050] In one embodiment, before training a target stress prediction model using the residual stress in the state measurement region at different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the sinusoidal wave signal of the ultrasonic propagation signal, the amplitude, frequency, and phase parameters are extracted to obtain the characteristics of the ultrasonic propagation signal.

[0051] In the specific implementation process, data features are extracted from the sinusoidal signals of the ultrasonic propagation signal under each stress state, including amplitude, frequency, and phase parameters. A training dataset is constructed, with stress values ​​as input and sinusoidal signal feature values ​​[amplitude and phase] as output. The extracted amplitude can be normalized by taking a logarithmic form. For the target stress prediction model in this embodiment, the initial neural network model framework is trained using training data. The initial neural network model framework includes an input layer, hidden layers, activation functions, and an output layer. The number of nodes in the input layer is equal to the stress parameter dimension, with 1 for single-axis stress and n for multi-axis stress. The hidden layers are fully connected layers, with 3-5 layers and 64-256 neurons per layer. The hidden layers use the ReLU activation function, and the output layer uses linear activation, with amplitude and phase branches set in the output layer. The Adam adaptive learning rate optimization algorithm is applied, with an initial learning rate of 1e-4. A cosine annealing strategy is used to monitor the validation set loss, and training is terminated if the loss does not decrease for 10 consecutive rounds. Once the neural network model has been trained for multiple rounds and its amplitude output error and output phase meet the accuracy requirements, it can be used to predict the amplitude and phase corresponding to zero stress.

[0052] S20: Using zero stress as input, the characteristics of the ultrasonic propagation signal under zero stress are output by the target stress prediction model to obtain a zero-stress test block model.

[0053] In the specific implementation process, the aforementioned embodiments established a stress prediction model, which can output the characteristics of the corresponding ultrasonic propagation signal based on the input stress value, thus achieving a truly zero-stress test block through prediction. The amplitude and phase of the characteristic parameters output by the zero-stress prediction are plotted as a sine wave signal according to the detection frequency and recorded in the detection instrument. When measuring residual stress subsequently, it is only necessary to compare the detected signal with the plotted signal using a cross-correlation / peak value algorithm to obtain the propagation time difference value and solve for the residual stress. That is: using zero stress as input, the characteristics of the ultrasonic propagation signal under zero stress are output using the target stress prediction model to obtain a zero-stress test block model, including: Using zero stress as input, the amplitude and phase of the ultrasonic propagation signal under zero stress are output using the target stress prediction model; Based on the amplitude and phase of the ultrasonic propagation signal under zero stress, the target sinusoidal signal is plotted to obtain a zero-stress test block model.

[0054] By embedding the traditional physical zero-stress test block calibration process into the testing instrument, steps such as high-altitude operation testing and confined space operation testing avoid the need for measurement and calibration procedures and carrying physical zero-stress test blocks, thus improving testing efficiency and on-site testing adaptability. The physical electronic zero-stress test block also requires no maintenance and will not experience stress changes due to static storage, preventing zero-stress benchmark shifts. Measurements are performed using a simple cantilever beam structure, combined with machine learning algorithms to predict the zero-stress acoustic time, replacing the process of preparing zero-stress test blocks through stress-relief annealing or other methods, saving both economic and time costs.

[0055] In this embodiment, ultrasonic measurements of a physical test block under different stress levels are used to collect ultrasonic propagation signals. Training data based on the corresponding residual stress is then used to construct a stress prediction model. This model can output the characteristics of the ultrasonic propagation signal corresponding to the input stress value. Finally, using zero stress as input, it outputs the characteristics of the ultrasonic propagation signal at that stress level, thus constructing a zero-stress test block model without a physical test block. This avoids the cumbersome process of preparing physical test blocks. Subsequent testing only requires calling the corresponding electronic test block data for comparison, improving testing efficiency. Furthermore, the electronic test block, through a neural network model, restores the zero-stress state as closely as possible, avoiding detection errors introduced by the preparation of physical test blocks, improving detection accuracy, and enhancing the effect of ultrasonic measurement of residual stress.

[0056] See attached document Figure 7 Based on the same inventive concept as in the foregoing embodiments, this application also provides a zero-stress test block model acquisition device for ultrasonic residual stress measurement, comprising: The determination module is used to determine the target stress prediction model based on the target measurement object. The target stress prediction model is trained based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value. The module is used to obtain the characteristics of the ultrasonic propagation signal under zero stress by using the target stress prediction model as input and outputting the model.

[0057] Those skilled in the art should understand that the division of the various modules in the embodiments is merely a logical functional division. In actual applications, they can be fully or partially integrated into one or more actual carriers. These modules can be implemented entirely in software through processing unit calls, entirely in hardware, or a combination of software and hardware. It should be noted that each module in the zero-stress test block model acquisition device in this embodiment corresponds one-to-one with each step in the zero-stress test block model acquisition method in the aforementioned embodiments. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned zero-stress test block model acquisition method, which will not be repeated here.

[0058] Based on the same inventive concept as in the foregoing embodiments, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when loaded and executed by a processor, implements the method for obtaining a zero-stress test block model as provided in the embodiments of this application.

[0059] Based on the same inventive concept as in the foregoing embodiments, embodiments of this application also provide an electronic device, including a processor and a memory, wherein, Memory is used to store computer programs; The processor is used to load and execute computer programs to enable electronic devices to perform the zero-stress test block model acquisition method provided in the embodiments of this application.

[0060] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a device including one or any combination of the above-mentioned memories. The computer may be a variety of computing devices, including smart terminals and servers.

[0061] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0062] As an example, executable instructions may, but do not necessarily, correspond to files in the file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple collaborative files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0063] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0064] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0065] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0066] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a multimedia terminal device (which may be a mobile phone, computer, television receiver, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0067] In summary, the embodiments of this application provide a method, apparatus, medium, and device for obtaining a zero-stress test block model. The method includes: determining a target stress prediction model based on the target measurement object; wherein the target stress prediction model is trained based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical test block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value; taking zero stress as input, using the target stress prediction model to output the ultrasonic propagation signal characteristics under zero stress, thereby obtaining a zero-stress test block model. This application utilizes ultrasonic measurements of physical test blocks under different stress levels to collect ultrasonic propagation signals. Based on the corresponding residual stress, training data is generated to construct a stress prediction model. This model can output the characteristics of the ultrasonic propagation signal corresponding to the input stress value. Finally, using zero stress as input, it outputs the characteristics of the ultrasonic propagation signal at that stress level, constructing a zero-stress test block model without a physical test block. This avoids the cumbersome process of preparing physical test blocks. Subsequent testing only requires calling the corresponding electronic test block data for comparison, improving testing efficiency. Furthermore, the electronic test block, through a neural network model, restores the zero-stress state as closely as possible, avoiding detection errors introduced by physical test block preparation, improving detection accuracy, and enhancing the effect of ultrasonic measurement of residual stress.

[0068] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for obtaining a zero-stress test block model, characterized in that, The application to ultrasonic residual stress measurement includes the following steps: Based on the target measurement object, a target stress prediction model is determined; wherein, the target stress prediction model is obtained by training based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value. Using zero stress as input, the characteristics of the ultrasonic propagation signal under zero stress are output by the target stress prediction model to obtain a zero-stress test block model.

2. The method for obtaining a zero-stress test block model according to claim 1, characterized in that, Before determining the target stress prediction model based on the target measurement object, the method further includes: Ultrasonic propagation signals were collected from the state measurement area of ​​the target physical specimen under different stress levels; wherein, the different stress levels included the initial zero stress level; Based on the ultrasonic propagation signal, the residual stress in the state measurement area under different stress levels is obtained; The target stress prediction model is trained by taking the residual stress in the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output.

3. The method for obtaining a zero-stress test block model according to claim 2, characterized in that, Before training the target stress prediction model using the residual stress of the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the measurement results of the blind hole measurement method on the target entity test block, the residual stress under the initial zero stress level is corrected to obtain the initial calibration stress of the state measurement area; The process of training the target stress prediction model by taking the residual stress in the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output includes: The target stress prediction model is trained by taking the residual stress in the state measurement area under different stress levels and the initial calibration stress as inputs and the characteristics of the ultrasonic propagation signal as outputs.

4. The method for obtaining a zero-stress test block model according to claim 2, characterized in that, The step of obtaining the residual stress in the state measurement area under different stress levels based on the ultrasonic propagation signal includes: Based on the ultrasonic propagation signal, the propagation sound time of the state measurement area under different stress levels is obtained respectively; wherein, the propagation sound time includes the propagation sound time under the initial zero stress level; The residual stress of the state measurement region under different stress levels is obtained by comparing the propagation time of the sound in the state measurement region under different stress levels with the propagation time of the sound in the initial zero stress level.

5. The method for obtaining a zero-stress test block model according to claim 4, characterized in that, Before obtaining the residual stress of the state measurement region at different stress levels based on the propagation sound time of the state measurement region at different stress levels and the propagation sound time at the initial zero stress level, the method further includes: Based on the different grain size states of the target entity test block, the propagation time of sound under the initial zero stress level is compensated to obtain the zero stress propagation time of sound. The step of obtaining the residual stress of the state measurement region at different stress levels based on the propagation sound time of the state measurement region at different stress levels and the propagation sound time at the initial zero stress level includes: The residual stress of the state measurement region under different stress levels is obtained by comparing the propagation sound time of the state measurement region under different stress levels with the propagation sound time of the zero stress at the initial zero stress level.

6. The method for obtaining a zero-stress test block model according to claim 2, characterized in that, Before training the target stress prediction model using the residual stress of the state measurement area under different stress levels as input and the characteristics of the ultrasonic propagation signal as output, the method further includes: Based on the sinusoidal wave signal of the ultrasonic propagation signal, the amplitude, frequency, and phase parameters are extracted to obtain the characteristics of the ultrasonic propagation signal.

7. The method for obtaining a zero-stress test block model according to claim 6, characterized in that, The process of obtaining a zero-stress test block model by using zero stress as input and utilizing the characteristics of the ultrasonic propagation signal under zero stress output by the target stress prediction model includes: Using zero stress as input, the amplitude and phase of the ultrasonic propagation signal under zero stress are output using the target stress prediction model. Based on the amplitude and phase of the ultrasonic propagation signal under zero stress, the target sinusoidal signal is plotted to obtain a zero-stress test block model.

8. A device for obtaining a zero-stress test block model, characterized in that, Applications include ultrasonic residual stress measurement, including: The determination module is used to determine the target stress prediction model based on the target measurement object; wherein, the target stress prediction model is obtained by training based on the residual stress and corresponding ultrasonic propagation signal characteristics of the state measurement area of ​​the target physical block under different stress levels, so that the target stress prediction model can output the corresponding ultrasonic propagation signal characteristics based on the input stress value. The module is used to obtain a zero-stress test block model by taking zero stress as input and utilizing the characteristics of the ultrasonic propagation signal under zero stress output by the target stress prediction model.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it implements the method for obtaining a zero-stress test block model as described in any one of claims 1-7.

10. An electronic device, characterized in that, Including processor and memory, among which, The memory is used to store computer programs; The processor is used to load and execute the computer program to cause the electronic device to perform the method for obtaining a zero-stress test block model as described in any one of claims 1-7.

Images