A method and system for internal stress separation of a boundary-weighted photoelastic model

Abstract

The application discloses a boundary-weighted photoelastic model internal stress separation method and a system thereof, and the method comprises the following steps: collecting model isoclinal line parameters and isochromatic line parameters; calculating a full-field principal stress, and specifying a calculation path direction and a solution region; performing boundary identification and data preprocessing; calculating a full-field shear stress, performing average pooling on the full-field shear stress; judging whether abnormal values and pixelated random errors are effectively eliminated, and whether gradient information in original data is reserved in the preprocessed data; performing boundary initial normal stress calculation; determining internal normal stress in one direction; calculating internal normal stress in another direction; respectively smoothing the internal normal stress in the two directions to obtain final full-field normal stress distribution. The application fully considers abnormal values and pixelated random errors in collected data, and the method proposed by setting a reasonable threshold and a correction method can effectively avoid the influence of collection errors on full-field stress solution.

Description

Technical Field

[0001] This application relates to the field of photoelastic experiments, and more specifically, to a method and system for separating internal stresses in a boundary-weighted photoelastic model. Background Technology

[0002] Photoelasticity is an optical experimental technique used to analyze the internal stress distribution of loaded complex geometries. Photoelasticity relies on the photoelastic effect (also known as temporary birefringence) of transparent materials. The refractive index of a photoelastic material changes under stress, causing the transmitted light to form different interference fringes due to the difference in refractive index. These fringes can be used to determine the stress distribution within the material. The internal information directly obtained by photoelasticity includes the principal stress directions, the principal stress differences, and the shear stress derived from them. Photoelasticity cannot directly obtain the normal stress components within the model. Stress separation techniques utilize existing information to determine the normal stress components in the model. The shear difference method, a commonly used stress separation technique, approximates the stress balance equation using a finite difference, calculating the normal stress increment along the path using the shear stress difference on both sides of the calculation path, thus transferring the stress state from the free boundary to the interior of the model. Current research focuses on stress separation techniques in photoelasticity, for example, using the Simpson integral form of continuous pixel shear stress to replace the traditional normal stress increment calculation mode, thereby reducing the calculation error of the shear stress difference. Alternatively, based on the normal stress components at the initial point on the target path, the centroid interpolation method can be used to calculate the normal stress components at each point on the target path, reducing the calculation error of the normal stress components. However, the shearing method, as the most commonly used stress separation method, is affected by the cumulative error in its calculation process and the experimental acquisition error. Regarding experimental acquisition error, as an experimental method based on image acquisition and processing, photoelasticity experiments inevitably have random errors due to pixelation. At the same time, factors such as bad light sources and optical path obstruction can bring about coarse acquisition errors. The calculation error of the shearing method originates from the finite difference approximation of the plane stress state. Since the shearing method calculates the internal stress components along a certain path starting from the boundary point, the calculation error will accumulate continuously as the calculation path extends. Since the normal stress components perpendicular to the calculation path are calculated through the stress components along the path, the accumulated error is also transmitted to the perpendicular stress components. Fully considering the sources of stress separation error and taking corresponding error suppression measures is the key to obtaining a high-precision stress field. Existing stress separation methods do not first consider how to suppress the effects of pixelation random errors and gross errors in experimental data acquisition, and secondly, they do not propose effective control methods for the cumulative errors in the shearing method calculation.

[0003] Therefore, how to provide a method that can achieve closed-loop control of the calculation results of internal normal stress has become an urgent problem to be solved in this field. Summary of the Invention

[0004] To address the aforementioned issues, this application proposes a boundary-weighted photoelastic model internal stress separation method, comprising the following steps: collecting isoclinal and isochromatic line parameters within the photoelastic model; calculating the principal stresses across the entire field based on the collected isoclinal and isochromatic line parameters, and specifying the calculation path direction and solution region; after completing the calculation of the principal stresses across the entire field and specifying the calculation path direction and solution region, performing boundary identification and data preprocessing; after completing boundary identification and data preprocessing, calculating the shear stress across the entire field, and performing average pooling on the shear stress across the entire field; determining whether outliers and pixelation random errors are effectively eliminated, and whether the preprocessed data retains the gradient information in the original data; calculating the initial normal stress at the boundary; determining the internal normal stress in one direction of the photoelastic model based on the initial normal stress and the overall shear stress; calculating the internal normal stress in another direction within the photoelastic model based on the internal normal stress in one direction; and smoothing the internal normal stresses in the two directions within the photoelastic model to obtain the final overall normal stress distribution.

[0005] As described above, the calculation of the principal stresses across the entire field based on the collected isoclinal and isochromatic parameters, as well as the specification of the calculation path direction and solution region, includes the following sub-steps: calculating the direction of the principal stresses across the entire field based on the isoclinal parameters; calculating the difference in principal stresses across the entire field based on the isochromatic parameters combined with the photoelastic constants of the material; specifying the calculation path direction and defining the solution region based on the geometric configuration of the photoelastic model.

[0006] As shown above, the principal stress difference across the entire field Represented as: , For the thickness of the photoelastic model, The stress stripe value of the material. This indicates the parameters for the isochromatic lines.

[0007] As shown above, the boundary identification and data preprocessing includes the following sub-steps: outlier identification and correction; pixel random error gradient smoothing.

[0008] As described above, pixel random error gradient smoothing includes: using a sliding window with a window width of a specified number of data points; calculating the linear regression slope of the boundary data within each window; and setting a relative deviation threshold for the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is If the relative deviation of the linear regression slopes of adjacent windows exceeds a threshold... If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

[0009] A boundary-weighted photoelastic model internal stress separation system includes: a parameter collection unit, a principal stress calculation unit, a boundary identification and processing unit, a full-field shear stress calculation unit, a judgment unit, a boundary initial normal stress calculation unit, a unit for determining internal normal stress in one direction, a unit for determining internal normal stress in another direction, and a unit for determining the full-field normal stress distribution; the parameter collection unit is used to collect isoclinal and isochromatic line parameters within the photoelastic model; the principal stress calculation unit is used to calculate the full-field principal stress based on the collected isoclinal and isochromatic line parameters, and to specify the calculation path direction and solution region; the boundary identification and processing unit is used for boundary identification and data preprocessing; and the full-field shear stress calculation unit is used to calculate the full-field shear stress. The system employs a stress calculation unit, which performs average pooling on the full-field shear stress. A judgment unit determines whether outliers and pixelation random errors are effectively eliminated, and whether the preprocessed data retains the gradient information from the original data. A boundary initial normal stress calculation unit calculates the boundary initial normal stress. A unit determines the internal normal stress in one direction based on the initial normal stress and the full-field shear stress. A unit determines the internal normal stress in another direction based on the internal normal stress in one direction, and a unit determines the full-field normal stress distribution by smoothing the internal normal stresses in the two directions within the photoelastic model.

[0010] As shown above, the principal stress calculation unit calculates the principal stress across the entire field based on the collected isoclinal and isochromatic parameters, and specifies the calculation path direction and solution region. This includes the following sub-steps: calculating the direction of the principal stress across the entire field based on the isoclinal parameters; calculating the difference of the principal stress across the entire field based on the isochromatic parameters combined with the photoelastic constant of the material; specifying the calculation path direction and defining the solution region based on the geometric configuration of the photoelastic model.

[0011] As shown above, in the principal stress calculation unit, the total principal stress difference... Represented as: , For the thickness of the photoelastic model, The stress stripe value of the material. This indicates the parameters for the isochromatic lines.

[0012] As shown above, the boundary recognition processing unit performs boundary recognition and data preprocessing, including the following sub-steps: outlier recognition and correction; pixel random error gradient smoothing.

[0013] As described above, the pixel random error gradient smoothing process performed by the boundary recognition processing unit includes: using a sliding window with a window width of a specified number of data points; calculating the linear regression slope of the boundary data within each window; and setting a relative deviation threshold for the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is If the relative deviation of the linear regression slopes of adjacent windows exceeds a threshold... If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

[0014] This application has the following beneficial effects:

[0015] (1) The photoelastic model stress separation method proposed in this application includes data preprocessing, which fully considers outliers and pixelation random errors in the collected data. By setting reasonable thresholds and correction methods, the proposed method can effectively avoid the impact of collection errors on the solution of stress in the whole field.

[0016] (2) This application proposes stress state identification and initial stress calculation based on pixelated boundary direction and principal stress direction, which makes the proposed model widely applicable to different complex geometric configurations.

[0017] (3) This application introduces additional boundary points to perform closed-loop control on the cumulative error in the continuous finite difference calculation of the shear stress difference method, so that the length of the calculation path, i.e. the cumulative error length, is not greater than the feature size of the model.

[0018] (4) This application calculates the normal stress in one direction at a time and obtains the principal stress in another direction using only one stress coordinate transformation, thus avoiding the accumulation of absolute errors caused by stress coordinate transformation in the model. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0020] Figure 1 This is a flowchart illustrating the stress separation method within a boundary-weighted photoelastic model provided in the embodiments of this application.

[0021] Figure 2 This is a schematic diagram of the internal structure of the stress separation system within the boundary-weighted photoelastic model provided in the embodiments of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0023] Example 1

[0024] like Figure 1 As shown, this embodiment provides a method for separating the internal stress of a boundary-weighted photoelastic model, specifically including the following steps:

[0025] Step S1: Collect the isoclinal and isochromatic line parameters inside the photoelastic model.

[0026] The phase-shifting method was used to determine the isoclinal and isochromatic parameters within the photoelastic model.

[0027] In a specific embodiment of the present invention, white light is used as the light source, and a planar orthogonally polarized light field is arranged. The polarizer and analyzer are rotated counterclockwise synchronously by angles of 0 degrees, 22.5 degrees, 45 degrees, and 67.5 degrees, respectively, to obtain four full-field light intensity images. A four-step phase-shifting method was used to obtain the phase wrapping diagram of isoclins across the entire field, and the isoclin parameters across the entire field were obtained by unwrapping the wrapping diagram. That is, the direction of the principal stress. Using monochromatic light as the light source, a double orthogonal circularly polarized light field is arranged, and the full-field light intensity image is obtained using a six-step phase-shifting method. Obtain the isochromatic line parameters for the entire field by unpacking. .

[0028] Step S2: Based on the collected isoclinal parameters With isochromatic line parameters Perform the calculation of principal stresses across the entire field, and specify the calculation path direction and solution region.

[0029] Step S2 includes the following sub-steps:

[0030] Step S21: Calculate the direction of principal stresses across the entire field based on isoclinal parameters. .

[0031] The direction of the principal stress throughout the field .

[0032] Step S22: Based on the isochromatic line parameters Calculate the principal stress difference across the entire field using the photoelastic constants of the material. .

[0033] Among them, the principal stress difference across the field ,

[0034] in For the thickness of the photoelastic model, This represents the stress stripe value of the material.

[0035] Among the parameters of the isochromatic lines Use the following formula to calculate:

[0036] ;

[0037] Step S23: Specify the calculation path direction according to the geometry of the photoelastic model.

[0038] The computation path direction is specified according to the geometric configuration of the photoelastic model, ensuring that the computation paths are in the same direction and that each computation path is as short as possible. The solution region is defined, and the planar region formed by all paths is the solution region.

[0039] The path direction is calculated, including the vertical Y direction and the horizontal X direction.

[0040] Step S3: After completing the calculation of the principal stresses across the entire field and specifying the calculation path direction and solution region, perform boundary identification and data preprocessing.

[0041] Based on images acquired through photoelastic experiments, the model boundaries are identified using edge detection algorithms (such as the Canny algorithm), and the principal stress directions and principal stress differences on the boundaries are extracted and stored as a one-dimensional boundary dataset. Simultaneously, it is necessary to suppress the influence of outliers and pixelation random errors in the photoelastic experimental data on stress separation.

[0042] Based on the above, step S3 includes the following sub-steps:

[0043] Step S31: Identify and correct outliers.

[0044] Calculate the average deviation of each boundary data point from its specified neighboring points (e.g., 5 points). Fitting average deviation The logarithmic distribution characteristics are used, and the upper confidence limit of the fitted logarithmic distribution confidence interval () is used as the outlier identification threshold. (For example, using the 94.45% confidence upper limit of the fitted logarithmic distribution as an outlier criterion) ), will exceed the deviation The data points (gross errors) are replaced with the mean of the first non-outlier points before and after, thus completing the outlier correction.

[0045] Step S32: Perform pixel random error gradient smoothing.

[0046] Using a sliding window with a specified number of data points (e.g., 20), calculate the linear regression slope of the boundary data within each window, and set a threshold for the relative deviation of the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is If the relative deviation of the linear regression slopes of adjacent windows exceeds the slope threshold... (Right now If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

[0047] Preferably, It is 0.01.

[0048] Step S4: After completing boundary identification and data preprocessing, calculate the full-field shear stress and perform average pooling on the full-field shear stress.

[0049] The direction of the principal stress in the whole field is used. Difference between principal stresses Calculate the full-field shear stress Specifically, it is expressed as:

[0050] ;

[0051] The average pooling process includes performing a 3×3 pooling kernel on the shear stress across the entire field with a step size of 1, automatically removing blank points outside the model boundary, and further unifying random errors through pixelation.

[0052] Step S5: Determine whether outliers and pixelation random errors have been effectively eliminated, and whether the preprocessed data retains the gradient information in the original data.

[0053] If outliers and pixelation random errors are not eliminated, and the preprocessed data does not retain the gradient information from the original data, adjust... , Then repeat steps S3-S4.

[0054] If outliers and pixelation random errors are eliminated, and the preprocessed data retains the gradient information from the original data, then proceed to step S6.

[0055] Step S6: Perform initial normal stress calculation at the boundary.

[0056] In a specific embodiment of the present invention, the nearest neighboring data point B that is closest to each data point A and whose position changes along the calculation path is found. The direction of the line connecting the two data points is taken as the pixel boundary direction of the data point A. .

[0057] Based on the free boundary stress state, combined with the pixel boundary direction The initial value of the normal stress is calculated according to the following discriminant, with respect to the principal stress direction θ:

[0058] (Formula 1);

[0059] in These are the initial normal stress components on the boundary. and These are the first and second principal stresses, respectively. Principal stress difference The preferred value for the direction discrimination threshold is [value to be filled in]. .

[0060] Step S7: Determine the internal normal stress in one direction of the photoelastic model based on the initial normal stress and the overall shear stress.

[0061] Based on the initial normal stress and the full-field shear stress, the bidirectional normal stress on the calculation path is calculated, and the inverse distance weighting of the bidirectional normal stress is calculated. The weighted normal stress is then used as the internal normal stress of the photoelastic model.

[0062] Step S7 includes the following sub-steps:

[0063] Step S71: Starting from the two parallel boundaries of the photoelastic model (such as the upper and lower boundaries of the curved beam), calculate the internal normal stress of the model using the shear stress difference method along the two directions between the boundary points, according to the calculation path direction.

[0064] Formula 2;

[0065] Formula 3;

[0066] in, and These are the results of normal stress separation using the shear stress difference method, starting from the upper and lower boundaries. and These are the initial values ​​of the normal stress at the upper and lower boundaries, respectively. To calculate the shear stress difference along the path, and These are the width and height of a pixel, respectively.

[0067] Step S72: Calculate the inverse distance weighted result of the bidirectional normal stress as an internal normal stress in one direction in the photoelastic model.

[0068] The inverse distance weighted average of the bidirectional normal stress is calculated as the stress separation result in the model. Taking the calculation path along the Y direction as an example, the internal normal stress... Represented as:

[0069] Formula 4;

[0070] in and These are the weighting coefficients for the shearing method along the two directions, specifically expressed as follows:

[0071] ;

[0072] in It calculates the path length. This is the distance between the current calculation point and the upper boundary.

[0073] Step S8: Based on the internal normal stress in one direction, calculate the internal normal stress in the other direction within the photoelastic model.

[0074] Among them, the normal stress in the Y direction Taking the calculation of normal stress in the X direction as an example, normal stress in the X direction Represented as:

[0075] ;

[0076] Step S9: Smooth the internal normal stresses in the two directions within the model to obtain the final full-field normal stress distribution.

[0077] As a specific embodiment of the present invention, the full-field normal stress is divided along the orthogonal direction of the calculation path (e.g., when the calculation path is vertical, the orthogonal direction is horizontal). For each divided region, a cubic polynomial fitting smoothing is performed to eliminate the result fluctuations in the orthogonal direction of the path, and the final full-field normal stress distribution is obtained.

[0078] Y-normal stress sequence along the X direction Represented as:

[0079] ;

[0080] in To calculate the point number, Calculate the number of points for the sequence.

[0081] Replace the function value of the fitted function at the calculated point index with the actual function value. The normal stress value, the cubic polynomial fitting function Represented as:

[0082] ;

[0083] Where i is the calculation point number. The parameters are to be determined, and can be obtained by solving the following formula:

[0084] .

[0085] Similarly, a similar sequence of X-normal stresses along the Y-direction... cubic fitting function With undetermined parameters The calculation formula is as follows:

[0086] ;

[0087] ;

[0088] .

[0089] The following are specific examples illustrating the application of this invention:

[0090] This invention is used to measure the internal stress of a complex boundary curved beam under a three-point bending load. The epoxy resin curved beam model has a length and height of 240 mm and 45 mm, respectively, and a thickness of 5.73 mm. Rollers are symmetrically arranged on both sides for support, and a 12 N vertical load is applied by a roller in the middle. The internal stress measurement process of the curved beam model is as follows:

[0091] A four-step phase-shifting method is used to obtain the isoclinal parameters of the entire field, and a six-step phase-shifting method is used to obtain the isochromatic parameters of the entire field. The isochromatic lines of the curved beam model are then obtained. Based on the isoclinal and isochromatic parameters of the curved beam model, the principal stress difference and principal stress direction are calculated. The calculation path is specified along the vertical direction. Data on the upper and lower boundary lines of the curved beam model are extracted, and outlier identification and gradient smoothing are performed sequentially. The shear stress of the entire field is calculated and average pooling is applied. The bidirectional normal stress is calculated using the shear stress difference method, and the inverse distance weighted average is used as the result of normal stress separation within the curved beam model.

[0092] Example 2

[0093] like Figure 2 As shown in the embodiment of this application, a boundary weighted photoelastic model internal stress separation system is provided, which specifically includes: a parameter collection unit 210, a principal stress calculation unit 220, a boundary identification and processing unit 230, a full-field shear stress calculation unit 240, a judgment unit 250, a boundary initial normal stress calculation unit 260, an internal normal stress determination unit in one direction 270, an internal normal stress determination unit in another direction 280, and a full-field normal stress distribution determination unit 290.

[0094] The parameter collection unit 210 is used to collect the isoclinal and isochromatic line parameters inside the photoelastic model.

[0095] The phase-shifting method was used to determine the isoclinal and isochromatic parameters within the photoelastic model.

[0096] In a specific embodiment of the present invention, white light is used as the light source, and a planar orthogonally polarized light field is arranged. The polarizer and analyzer are rotated counterclockwise synchronously by angles of 0 degrees, 22.5 degrees, 45 degrees, and 67.5 degrees, respectively, to obtain four full-field light intensity images. A four-step phase-shifting method was used to obtain the phase wrapping diagram of isoclins across the entire field, and the isoclin parameters across the entire field were obtained by unwrapping the wrapping diagram. That is, the direction of the principal stress. Using monochromatic light as the light source, a double orthogonal circularly polarized light field is arranged, and the full-field light intensity image is obtained using a six-step phase-shifting method. Obtain the isochromatic line parameters for the entire field by unpacking. .

[0097] Principal stress calculation unit 220 is used to calculate the principal stress based on the collected isoclinal parameters. With isochromatic line parameters Perform the calculation of principal stresses across the entire field, and specify the calculation path direction and solution region.

[0098] The principal stress calculation unit 220 performs the following sub-steps:

[0099] Step T1: Calculate the direction of principal stresses across the entire field based on isoclinal parameters. .

[0100] The direction of the principal stress throughout the field .

[0101] Step T2: Calculate the principal stress difference across the entire field based on the isochromatic line parameters and the material's photoelastic constant. .

[0102] Among them, the principal stress difference across the field ;

[0103] in For model thickness, This represents the stress stripe value of the material.

[0104] Among the isoclin parameters Use the following formula to calculate:

[0105] .

[0106] Step T3: Specify the calculation path direction according to the geometric configuration of the photoelastic model and define the solution region.

[0107] The computation path direction is specified according to the geometric configuration of the photoelastic model, ensuring that the computation paths are in the same direction and that each computation path is as short as possible. The solution region is defined, and the planar region formed by all paths is the solution region.

[0108] The path direction is calculated, including the vertical Y direction and the horizontal X direction.

[0109] The boundary recognition processing unit 230 is used for boundary recognition and data preprocessing.

[0110] Based on images acquired through photoelastic experiments, the model boundaries are identified using edge detection algorithms (such as the Canny algorithm), and the principal stress directions and principal stress differences on the boundaries are extracted and stored as a one-dimensional boundary dataset. Simultaneously, it is necessary to suppress the influence of outliers and pixelation random errors in the photoelastic experimental data on stress separation.

[0111] Based on the above, the boundary recognition processing unit 230 performs the following sub-steps:

[0112] Step Q1: Identify and correct outliers.

[0113] Calculate the average deviation of each boundary data point from its specified neighboring points (e.g., 5 points). , average deviation of fit The logarithmic distribution characteristics are used, and the upper confidence limit of the fitted logarithmic distribution confidence interval () is used as the outlier identification threshold. (For example, using the 94.45% confidence upper limit of the fitted logarithmic distribution as an outlier criterion) ), will exceed the deviation The data points (gross errors) are replaced with the mean of the first non-outlier points before and after, thus completing the outlier correction.

[0114] Step Q2: Perform pixel random error gradient smoothing.

[0115] Using a sliding window with a specified number of data points (e.g., 20), calculate the linear regression slope of the boundary data within each window, and set a threshold for the relative deviation of the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is If the relative deviation of the linear regression slopes of adjacent windows exceeds the threshold (Right now If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

[0116] Preferably, It is 0.01.

[0117] The full-field shear stress calculation unit 240 is used to calculate the full-field shear stress and perform average pooling on the full-field shear stress.

[0118] The direction of the principal stress in the whole field is used. Difference between principal stresses Calculate the full-field shear stress Specifically, it is expressed as:

[0119] ;

[0120] The average pooling process includes performing a 3×3 pooling kernel on the shear stress across the entire field with a step size of 1, automatically removing blank points outside the model boundary, and further unifying random errors through pixelation.

[0121] The judgment unit 250 is used to determine whether outliers and pixelation random errors have been effectively eliminated, and whether the preprocessed data retains the gradient information in the original data.

[0122] If outliers and pixelation random errors are not eliminated, and the preprocessed data does not retain the gradient information from the original data, adjust... , The boundary identification processing unit 230 and the full-field shear stress calculation unit 240 are then repeatedly executed.

[0123] If outliers and pixelation random errors are eliminated, and the preprocessed data retains the gradient information from the original data, then the boundary initial normal stress calculation unit 260 is executed.

[0124] The initial normal stress calculation unit 260 is used to calculate the initial normal stress of the boundary.

[0125] In a specific embodiment of the present invention, the nearest neighboring data point B that is closest to each data point A and whose position changes along the calculation path is found. The direction of the line connecting the two data points is taken as the pixel boundary direction of the data point A. .

[0126] Based on the free boundary stress state, combined with the pixel boundary direction The initial value of the normal stress is calculated according to the following discriminant, with respect to the principal stress direction θ:

[0127] (Formula 1);

[0128] in These are the initial normal stress components on the boundary. and These are the first and second principal stresses, respectively. Principal stress difference The preferred value for the direction discrimination threshold is [value to be filled in]. .

[0129] An internal normal stress determination element 270 is used to determine the internal normal stress in one direction of the photoelastic model based on the initial normal stress and the overall shear stress.

[0130] Based on the initial normal stress and the full-field shear stress, the bidirectional normal stress on the calculation path is calculated, and the inverse distance weighting of the bidirectional normal stress is calculated. The weighted normal stress is then used as the internal normal stress of the photoelastic model.

[0131] One of the internal normal stress determination elements 270 performs the following sub-steps:

[0132] Step F1: Starting from the two parallel boundaries of the model (such as the upper and lower boundaries of the curved beam), calculate the internal normal stress of the model using the shear stress difference method along the two directions between the boundary points, according to the calculation path direction.

[0133] Formula 2;

[0134] Formula 3;

[0135] in, and These are the results of normal stress separation using the shear stress difference method, starting from the upper and lower boundaries. and These are the initial values ​​of the normal stress at the upper and lower boundaries, respectively. To calculate the shear stress difference along the path, and These are the width and height of a pixel, respectively.

[0136] Step F2: Calculate the inverse distance weighted result of the bidirectional normal stress as the internal normal stress in one direction of the model.

[0137] Another internal normal stress determination element 280 is used to calculate the inverse distance weighted average of the bidirectional normal stress as the stress separation result in the model. Taking the calculation path along the Y direction as an example, the internal normal stress... Represented as:

[0138] Formula 4;

[0139] in and These are the weighting coefficients for the shearing method along the two directions, specifically expressed as follows:

[0140] ;

[0141] in It calculates the path length. This is the distance between the current calculation point and the upper boundary.

[0142] Calculate the internal normal stress in another direction within the photoelastic model based on the internal normal stress in one direction.

[0143] Among them, the normal stress in the Y direction Taking the calculation of normal stress in the X direction as an example, normal stress in the X direction Represented as:

[0144] ;

[0145] The full-field normal stress distribution determination element 290 is used to smooth the internal normal stress in two directions within the model to obtain the final full-field normal stress distribution.

[0146] As a specific embodiment of the present invention, the full-field normal stress is divided along the orthogonal direction of the calculation path (e.g., when the calculation path is vertical, the orthogonal direction is horizontal). For each divided region, a cubic polynomial fitting smoothing is performed to eliminate the result fluctuations in the orthogonal direction of the path, and the final full-field normal stress distribution is obtained.

[0147] Y-normal stress sequence along the X direction Represented as:

[0148] ;

[0149] in To calculate the point number, Calculate the number of points for the sequence.

[0150] Replace the function value of the fitted function at the calculated point index with the actual function value. The normal stress value, the cubic polynomial fitting function Represented as:

[0151] ;

[0152] Where i is the calculation point number. The parameters are to be determined, and can be obtained by solving the following formula:

[0153] .

[0154] Similarly, a similar sequence of X-normal stresses along the Y-direction... cubic fitting function With undetermined parameters The calculation formula is as follows:

[0155] ;

[0156] ;

[0157] .

[0158] The following are specific examples illustrating the application of this invention:

[0159] This invention is used to measure the internal stress of a complex boundary curved beam under a three-point bending load. The epoxy resin curved beam model has a length and height of 240 mm and 45 mm, respectively, and a thickness of 5.73 mm. Rollers are symmetrically arranged on both sides for support, and a 12 N vertical load is applied by a roller in the middle. The internal stress measurement process of the curved beam model is as follows:

[0160] A four-step phase-shifting method is used to obtain the isoclinal parameters of the entire field, and a six-step phase-shifting method is used to obtain the isochromatic parameters of the entire field. The isochromatic lines of the curved beam model are then obtained. Based on the isoclinal and isochromatic parameters of the curved beam model, the principal stress difference and principal stress direction are calculated. The calculation path is specified along the vertical direction. Data on the upper and lower boundary lines of the curved beam model are extracted, and outlier identification and gradient smoothing are performed sequentially. The shear stress of the entire field is calculated and average pooling is applied. The bidirectional normal stress is calculated using the shear stress difference method, and the inverse distance weighted average is used as the result of normal stress separation within the curved beam model.

[0161] This application also provides a computer storage medium storing computer instructions, which, when invoked, are used to execute the stress separation method within the boundary-weighted photoelastic model.

[0162] The embodiments disclosed in this invention provide a computer-readable storage medium storing computer program instructions that, when executed on a computer, cause the computer to perform the aforementioned method for separating internal stresses in a boundary-weighted photoelastic model.

[0163] This invention provides a processor for processing the above-described method for separating internal stresses in a boundary-weighted photoelastic model.

[0164] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0165] The various methods, steps, and logic diagrams disclosed in the embodiments of this invention can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods.

[0166] The storage medium can be memory, such as volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0167] Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).

[0168] This application has the following beneficial effects:

[0169] (1) The photoelastic model stress separation method proposed in this application includes data preprocessing, which fully considers outliers and pixelation random errors in the collected data. By setting reasonable thresholds and correction methods, the proposed method can effectively avoid the impact of collection errors on the solution of stress in the whole field.

[0170] (2) This application proposes stress state identification and initial stress calculation based on pixelated boundary direction and principal stress direction, which makes the proposed model widely applicable to different complex geometric configurations.

[0171] (3) This application introduces additional boundary points to perform closed-loop control on the cumulative error in the continuous finite difference calculation of the shear stress difference method, so that the length of the calculation path, i.e. the cumulative error length, is not greater than the feature size of the model.

[0172] (4) This application calculates the normal stress in one direction at a time and obtains the principal stress in another direction using only one stress coordinate transformation, thus avoiding the accumulation of absolute errors caused by stress coordinate transformation in the model.

[0173] Although the examples referenced in this application are described for illustrative purposes only and not for limiting the scope of this application, changes, additions and / or deletions to the implementation may be made without departing from the scope of this application.

[0174] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for internal stress separation of a boundary-weighted photoelastic model, characterized by, Includes the following steps: Collect isoclinal and isochromatic line parameters within the photoelastic model; The principal stresses across the entire field are calculated based on the collected isoclinal and isochromatic line parameters, and the calculation path direction and solution region are specified. After completing the calculation of the principal stresses across the entire field, and specifying the calculation path direction and solution region, boundary identification and data preprocessing are performed. After completing boundary identification and data preprocessing, the full-field shear stress is calculated and averaged pooled. Determine whether outliers and pixelation random errors are effectively eliminated, and whether the preprocessed data retains the gradient information in the original data; If outliers and pixelation random errors are effectively eliminated, and the preprocessed data retains the gradient information in the original data, the initial normal stress of the boundary can be calculated. Based on the initial normal stress and the overall shear stress, determine the internal normal stress in one direction of the photoelastic model; Calculate the internal normal stress in another direction within the photoelastic model based on the internal normal stress in one direction. The internal normal stresses in the two directions within the photoelastic model are smoothed to obtain the final full-field normal stress distribution.

2. The boundary-weighted photoelastic model internal stress separation method of claim 1, wherein, The calculation of principal stresses across the entire field is performed based on the collected isoclinal and isochromatic line parameters, along with the specified calculation path direction and solution domain. This includes the following sub-steps: Calculate the direction of principal stresses across the entire field based on isochoric parameters; The principal stress difference across the entire field is calculated based on the isochromatic line parameters and the photoelastic constant of the material. The calculation path direction is specified based on the geometry of the photoelastic model.

3. The method for separating the internal stress of a boundary-weighted photoelastic model as described in claim 2, characterized in that, Total principal stress difference Represented as: ； For the thickness of the photoelastic model, The stress stripe value of the material. This indicates the parameters for the contour lines.

4. The method for separating the internal stress of a boundary-weighted photoelastic model as described in claim 1, characterized in that, Boundary identification and data preprocessing include the following sub-steps: Perform outlier identification and correction; Perform pixel random error gradient smoothing.

5. The method for separating the internal stress of a boundary-weighted photoelastic model as described in claim 4, characterized in that, Performing pixel random error gradient smoothing includes: Use a sliding window with a width of a specified number of data points; Calculate the linear regression slope of the boundary data within each window, and set a threshold for the relative deviation of the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is ; If the relative deviation of the linear regression slopes of adjacent windows exceeds a threshold If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

6. A stress separation system for a boundary-weighted photoelastic model, characterized in that, include: The system includes a parameter collection unit, a principal stress calculation unit, a boundary identification and processing unit, a full-field shear stress calculation unit, a judgment unit, a boundary initial normal stress calculation unit, a unit for determining internal normal stress in one direction, a unit for determining internal normal stress in another direction, and a unit for determining the distribution of normal stress in the entire field. The parameter collection unit is used to collect the isoclinal and isochromatic line parameters inside the photoelastic model. The principal stress calculation unit is used to calculate the principal stresses across the entire field based on the collected isoclinal and isochromatic line parameters, and to specify the calculation path direction and solution area. The boundary recognition processing unit is used for boundary recognition and data preprocessing. The full-field shear stress calculation unit is used to calculate the full-field shear stress and perform average pooling on the full-field shear stress. The judgment unit is used to determine whether outliers and pixelation random errors have been effectively eliminated, and whether the preprocessed data retains the gradient information in the original data. The boundary initial normal stress calculation unit is used to calculate the boundary initial normal stress. An internal normal stress determination element is used to determine the internal normal stress in one direction of the photoelastic model based on the initial normal stress and the overall shear stress. Another internal normal stress determination element is used to calculate the internal normal stress in another direction within the photoelastic model based on the internal normal stress in one direction. The full-field normal stress distribution determination element is used to smooth the internal normal stress in two directions within the photoelastic model to obtain the final full-field normal stress distribution.

7. The stress separation system within the boundary-weighted photoelastic model as described in claim 6, characterized in that, The principal stress calculation unit calculates the principal stresses across the entire field based on the collected isoclinal and isochromatic line parameters, and specifies the calculation path direction and solution domain, including the following sub-steps: Calculate the direction of principal stresses across the entire field based on isochoric parameters; The principal stress difference across the entire field is calculated based on the isochromatic line parameters and the photoelastic constant of the material. The calculation path direction is specified based on the geometric configuration of the photoelastic model, and the solution region is defined.

8. The stress separation system within the boundary-weighted photoelastic model as described in claim 7, characterized in that, In the principal stress calculation unit, the total principal stress difference Represented as: ； For the thickness of the photoelastic model, The stress stripe value of the material. This indicates the parameters for the contour lines.

9. The stress separation system within the boundary-weighted photoelastic model as described in claim 6, characterized in that, The boundary recognition processing unit performs boundary recognition and data preprocessing, including the following sub-steps: Perform outlier identification and correction; Perform pixel random error gradient smoothing.

10. The stress separation system within the boundary-weighted photoelastic model as described in claim 9, characterized in that, The boundary recognition processing unit performs pixel random error gradient smoothing, including: Use a sliding window with a width of a specified number of data points; Calculate the linear regression slope of the boundary data within each window, and set a threshold for the relative deviation of the slope. The slope of the linear fit of the boundary data within the window is... The slope of the linear fit in the previous window is ; If the relative deviation of the linear regression slopes of adjacent windows exceeds a threshold If the deviation exceeds the standard in the next window, then the data points in the next window will be corrected to the fitted slope line in the previous window.

Images