Method for predicting relation between residual stress and deformation of crankshaft of diesel engine

Abstract

A diesel engine crankshaft residual stress and deformation relation prediction method belongs to the technical field of internal combustion engine key component manufacturing. The method comprises the steps of dynamically correcting a finite element model through multi-stage measured data, performing high-precision reconstruction of a crankshaft residual stress field, establishing a quantitative relation between stress distribution and deformation, and combining numerical simulation and response surface method optimization to realize high-precision residual stress distribution prediction. And further analyzing the release and redistribution rules of the residual stress in the material removal process. Residual stress distribution is actively controlled to prolong fatigue life, and stable production of high-precision and long-life crankshafts is ensured. The grinding feeding rate is reduced through active control over residual stress in the grinding process so that the tensile stress peak value in the layer depth direction can be reduced, and meanwhile it is guaranteed that the surface roughness is within a proper interval. A gradient cooling strategy is adopted, the cooling liquid flow is increased in the initial grinding stage, and surface layer thermal stress accumulation is restrained. And the grinding wheel linear speed is controlled, and grinding efficiency and residual stress inhibition requirements are balanced.

Description

TECHNICAL FIELD

[0001] The application relates to a diesel engine crankshaft residual stress and deformation relationship prediction method and belongs to the technical field of key components of internal combustion engines. BACKGROUND

[0002] As a core component of a diesel engine, the residual stress distribution of a crankshaft directly affects the fatigue strength and service life. The traditional method has the following defects:

[0003] Experimental measurement limitations: X-ray diffraction, drilling method and the like need to be measured point by point, and the efficiency is low and the full surface cannot be covered, so it is difficult to reflect the stress gradient change law in the depth direction and the circumferential direction.

[0004] Simulation model deviation: the existing finite element model does not dynamically fuse the measured data of the machining process, ignores the heat-force coupling, heat treatment phase change and the like, and the like, and the like, resulting in insufficient prediction accuracy.

[0005] Process optimization blind area: lacking quantitative mapping relationship between residual stress and deformation, process parameter adjustment depends on experience, and deformation out-of-tolerance or stress concentration is easily caused. SUMMARY

[0006] The application aims to provide a diesel engine crankshaft residual stress and deformation relationship prediction method, which dynamically corrects a finite element model through multi-stage measured data, reconstructs a high-precision residual stress field of the crankshaft, establishes a quantitative relationship between stress distribution and deformation, optimizes the numerical simulation and the response surface method, realizes high-precision residual stress distribution prediction, and is helpful to the machining process optimization and fatigue life evaluation of the crankshaft.

[0007] The application aims to realize the following technical solutions:

[0008] The application discloses a diesel engine crankshaft residual stress and deformation relationship prediction method, which comprises the following steps:

[0009] Step one: after the first process of the crankshaft, the residual stress discrete tensor values of the main journal, the connecting rod journal and the fillet along the depth and the cross-sectional circumferential direction are measured.

[0010] Step two: a three-dimensional geometric model is constructed based on the size after machining, and a finite element model of the crankshaft is generated according to the three-dimensional geometric model.

[0011] Step three: the residual stress discrete tensor values are further analyzed by using data analysis, two-dimensional interpolation calculation is performed between the known data points in the depth and cross-sectional circumferential directions of the crankshaft by using a bilinear interpolation method, and a residual stress distribution model of the crankshaft is established in combination with a local column coordinate system.

[0012] Step four: With the help of Abaqus finite element software, combined with SIGINI user subroutine, the residual stress discrete tensor value obtained in step one is assigned to the crankshaft finite element model obtained in step two, and the displacement distribution data of the. odb file is simulated and output.

[0013] Step five: Based on Python script, read the Abaqus. odb result file and convert the calculated displacement results into circular runout data, establish the mapping function relationship between the circular runout data and the axial coordinate. The mapping function relationship quantifies the micro-deformation caused by residual stress release as the circular runout data distributed along the axial direction, provides the basis for model correction and verification, and outputs the prediction results of circular runout deformation under the processing technology.

[0014] Step six: Repeat steps one to five to obtain the circular runout data under each process, complete the prediction of the circular runout deformation according to the residual stress distribution law of the crankshaft, and obtain the prediction results of the crankshaft circular runout deformation.

[0015] Further, it further includes step seven, which tests the residual stress value by using non-model establishment source data detection point, and compares and analyzes the simulation prediction results of the crankshaft circular runout value obtained in step six. Calculate the error between the predicted value and the actual value, and evaluate the prediction accuracy of the crankshaft circular runout deformation prediction model.

[0016] Further, it further includes step eight, which, under the premise that the verification result of step seven meets the preset accuracy requirement, optimizes the diesel engine crankshaft processing technology according to the crankshaft circular runout deformation prediction results obtained in step six and the deformation amount obtained in step five.

[0017] Further, the construction method of the mapping function relationship between the circular runout data and the axial coordinate in step five is as follows:

[0018] ① Use python-odb library to parse. odb file and extract node coordinates and residual stress tensor data.

[0019] ② In the circumferential-layer depth coordinate system, use cubic spline interpolation method to generate continuous stress distribution surface.

[0020] ③ Fit the stress gradient function by multivariate nonlinear regression. Regression analysis: establish the mapping function relationship between the circular runout data and the axial coordinate by multivariate nonlinear regression fitting, which is as follows:

[0021] σ(r,θ)=α·e -βr +γ·sin(θ+φ)+δ·rθ

[0022] Where, α, β, γ, δ, are fitting parameters, reflecting the attenuation and distribution law of stress with layer depth and circumference.

[0023] Further, in step five, the crankshaft roundness runout deformation is obtained according to the mapping function relationship between the roundness runout data and the axial coordinate, and the implementation steps are as follows:

[0024] Step 5.1: The residual stress distribution is taken as the initial condition and is introduced into the finite element model of the crankshaft. By using the finite element calculation software and combining the stress-strain property data of the material, the displacement distribution of the crankshaft under the action of the residual stress field is calculated. Considering the elasticity, plasticity and possible nonlinear behavior of the material under different stress levels, the overall displacement field distribution of the crankshaft under the action of the residual stress is obtained. The displacement field distribution of the crankshaft covers the displacement of each part of the crankshaft, providing a comprehensive data basis for subsequent displacement extraction and analysis.

[0025] Step 5.2: After obtaining the displacement distribution of the crankshaft, the displacement data of several key parts are further extracted, i.e. the radial displacement of the main journal, the tangential displacement of the connecting rod journal and the normal displacement at the fillet.

[0026] Step 5.3: Based on the key part displacement data extracted in step 5.2, the key indicators representing the overall and local deformation degree of the crankshaft are calculated:

[0027] Main journal bending deformation: linear fitting is performed on the radial displacement of the main journal along its axial distribution, and the maximum offset of the fitted straight line or curve or the radial runout relative to the ideal axis is calculated to quantify the bending or eccentric deformation degree of the main journal. Connecting rod journal torsional deformation: analyze the distribution pattern of the tangential displacement of the connecting rod journal in the circumferential direction. Calculate the difference of the tangential displacement at different phase angles, especially the displacement difference at the relative 180-degree position, or calculate the gradient change rate of the tangential displacement, to quantify the torsional deformation of the connecting rod journal relative to the main journal axis caused by residual stress. Fillet deformation: directly use the extracted normal displacement data at the fillet to calculate the deviation of the displacement relative to the normal displacement of the adjacent non-fillet area. This deviation directly reflects the local geometric shape change of the fillet caused by residual stress, and is a key indicator for evaluating the micro-deformation and fatigue sensitivity of the stress concentration area.

[0028] Further, the method for evaluating the prediction accuracy of the crankshaft roundness runout deformation prediction model in step seven is as follows:

[0029] Compare the stress gradient predicted by the crankshaft residual stress distribution model evaluation model with the X-ray diffraction depth scanning results, and the allowable error is ≤10%.

[0030] Compare the deformation calculated by the finite element method with the measured value of the ultrasonic critical refraction longitudinal wave detection, and the allowable error is ≤10%.

[0031] Further, in step eight, the implementation method for optimizing the diesel engine crankshaft machining process is as follows:

[0032] According to the circumferential stress concentration area identified according to the crankshaft residual stress distribution model, the rolling strengthening process parameters are recommended to be adjusted, and the specific implementation method is:

[0033] Increase the rolling pressure to introduce residual compressive stress in the surface layer, offset the tensile stress concentration under the working load, and make the fatigue cracks in the subsurface during use, reduce the crack propagation speed.

[0034] Control the rolling pressure value to avoid excessive plastic deformation and cause surface microcracks. Optimize the rolling path to cover the circumferential stress concentration area and ensure the uniformity of the compressive stress distribution.

[0035] Further, in step eight, the diesel engine crankshaft machining process optimization method is as follows:

[0036] According to the tensile stress distribution law in the depth direction, the grinding process parameters are optimized, and the specific implementation method is:

[0037] Through the residual stress active control in the grinding process, the grinding feed rate is reduced to reduce the tensile stress peak value in the depth direction, while ensuring that the surface roughness is in the appropriate interval. Gradient cooling strategy is adopted to increase the cooling liquid flow in the initial stage of grinding to suppress the accumulation of surface thermal stress. Control the wheel linear speed to balance the grinding efficiency and the demand for residual stress suppression.

[0038] Beneficial effects:

[0039] 1. The diesel engine crankshaft residual stress and deformation relationship prediction method disclosed in the present application, through multi-stage measured data dynamic correction of the finite element model, reconstructs the high-precision crankshaft residual stress field, establishes the quantitative relationship between stress distribution and deformation, optimizes the numerical simulation and response surface method, realizes high-precision prediction of residual stress distribution, analyzes the residual stress release and redistribution law in the material removal process, actively controls the residual stress distribution, improves the fatigue life, identifies the stress concentration risk points, and guarantees the stable production of high-precision and long-life crankshafts.

[0040] 2. The diesel engine crankshaft residual stress and deformation relationship prediction method disclosed in the present application, through the residual stress active control in the grinding process, reduces the grinding feed rate to reduce the tensile stress peak value in the depth direction, while ensuring that the surface roughness is in the appropriate interval; gradient cooling strategy is adopted to increase the cooling liquid flow in the initial stage to suppress the accumulation of surface thermal stress; control the wheel linear speed to balance the grinding efficiency and the demand for residual stress suppression.

[0041] 3, The diesel engine crankshaft residual stress and deformation relationship prediction method disclosed by the application, after reconstructing the residual stress field and establishing the stress and deformation quantitative relationship, the stress value is given to the model by means of Abaqus and SIGINI subroutine, the data is read and converted into circular runout data by combining Python script, the mapping relationship is established to quantify the microscopic deformation, the basis is provided for model correction, and the deformation prediction result is output, batch prediction is realized through collaborative simulation optimization, the precision and efficiency are improved, and process optimization and life assessment are facilitated. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The diesel engine crankshaft residual stress and deformation relationship prediction method of the application.

[0043] Figure 2 The abaqus subroutine data mapping principle diagram in the embodiment of the application.

[0044] Figure 3 The finite element model divided by the 16V crankshaft in the embodiment of the application.

[0045] Figure 4 The crankshaft residual stress measured data crankshaft measuring point arrangement schematic diagram in the embodiment of the application.

[0046] Figure 5 The residual stress field distribution model of the first machining process of the crankshaft established in the embodiment of the application.

[0047] Figure 6 The displacement distribution model of the crankshaft under the first machining process in the embodiment of the application.

[0048] Figure 7 The prediction value and actual value data comparison graph in the embodiment of the application. DETAILED DESCRIPTION

[0049] In order to better illustrate the purpose and advantages of the application, the content of the application will be further described below in combination with the drawings and examples.

[0050] Embodiment 1:

[0051] As Figure 1 shown, the diesel engine crankshaft residual stress and deformation relationship prediction method disclosed by the embodiment comprises the following steps:

[0052] Step one: after the first process of the crankshaft, the residual stress discrete tensor value of the main journal, connecting rod journal and fillet along the depth and cross-sectional circumferential direction is measured.

[0053] Step two: based on the size after machining, a three-dimensional geometric model is constructed, and a finite element model of the crankshaft is generated according to the three-dimensional geometric model.

[0054] Step three: further analyze the residual stress discrete tensor value by data analysis, use bilinear interpolation method to calculate two-dimensional interpolation between the known data points in the depth and cross-sectional circumferential direction of the crankshaft, and establish a residual stress distribution model of the crankshaft in combination with the local column coordinate system.

[0055] Step four: assign the residual stress discrete tensor value obtained in step one to the finite element model of the crankshaft obtained in step two by means of Abaqus finite element software in combination with SIGINI user subroutine, and simulate the displacement distribution data of the.odb file.

[0056] Step five: read the.odb result file of Abaqus based on Python script and convert the calculated displacement result into circular runout data, and establish a mapping function relationship between the circular runout data and the axial coordinate. The mapping function relationship quantifies the micro-deformation caused by the release of residual stress into the axial distribution of circular runout data, provides a basis for correcting and verifying the model, and outputs the prediction result of the circular runout deformation under the machining process.

[0057] Step six: repeat steps one to five to obtain the circular runout data under each process, complete the prediction of the circular runout deformation according to the residual stress distribution law of the crankshaft, and obtain the prediction result of the circular runout deformation of the crankshaft.

[0058] Further, it further includes step seven, which tests the residual stress value by using non-model establishment source data detection point, compares and analyzes the simulation prediction result of the crankshaft circular runout value obtained in step six, calculates the error between the predicted value and the actual value, and evaluates the prediction accuracy of the crankshaft circular runout deformation prediction model.

[0059] Further, it further includes step eight, which optimizes the diesel engine crankshaft machining process according to the crankshaft circular runout deformation prediction result obtained in step six and the deformation amount obtained in step five on the premise that the verification result in step seven meets the preset accuracy requirement.

[0060] Further, the construction method of the mapping function relationship between the circular runout data and the axial coordinate in step five is as follows:

[0061] ① Use python-odb library to parse.odb file and extract node coordinates and residual stress tensor data.

[0062] ② In the circumferential-depth coordinate system, use cubic spline interpolation method to generate continuous stress distribution surface.

[0063] ③ Fit the stress gradient function by multivariate nonlinear regression. Regression analysis: establish the mapping function relationship between the circular runout data and the axial coordinate by multivariate nonlinear regression fitting, which is as follows:

[0064] σ(r,θ)=α·e-βr + γ · sin (θ + φ) + δ · rθ

[0065] wherein a, b, g, d, are fitting parameters, reflecting the stress decay and distribution with depth and circumferential direction.

[0066] Further, in step five, the crankshaft roundness runout deformation is obtained according to the mapping function relationship between the roundness runout data and the axial coordinate, and the implementation steps are as follows:

[0067] Step 5.1: The residual stress distribution is taken as the initial condition and is introduced into the finite element model of the crankshaft. By using the finite element calculation software and combining with the stress-strain attribute data of the material, the displacement distribution of the crankshaft under the action of the residual stress field is calculated. Considering the elasticity, plasticity and possible nonlinear behavior of the material under different stress levels, the overall displacement field distribution of the crankshaft under the action of the residual stress is obtained. The displacement field distribution of the crankshaft covers the displacement of each part of the crankshaft, providing a comprehensive data basis for subsequent displacement extraction and analysis.

[0068] Step 5.2: After obtaining the displacement distribution of the crankshaft, the displacement data of several key parts are further extracted, i.e. the radial displacement of the main journal, the tangential displacement of the connecting rod journal and the normal displacement at the fillet.

[0069] Step 5.3: Based on the key part displacement data extracted in step 5.2, the key indicators representing the overall and local deformation degree of the crankshaft are calculated:

[0070] Main journal bending deformation: linear fitting is performed on the axial distribution of the radial displacement of the main journal, and the maximum offset or the radial runout relative to the ideal axis of the fitting straight line or curve is calculated to quantify the bending or eccentric deformation degree of the main journal. Connecting rod journal torsion deformation: analyze the distribution pattern of the tangential displacement of the connecting rod journal in the circumferential direction. Calculate the difference of the tangential displacement at different phase angles, especially the displacement difference at the position points of 180 degrees, or calculate the gradient change rate of the tangential displacement, to quantify the torsion deformation of the connecting rod journal relative to the main journal axis caused by the residual stress. Fillet deformation: directly use the extracted normal displacement data at the fillet to calculate the deviation amount of the displacement relative to the normal displacement of the adjacent non-fillet area. The deviation amount directly reflects the local geometric shape change of the fillet caused by the residual stress, and is a key indicator for evaluating the micro-deformation and fatigue sensitivity of the stress concentration area.

[0071] Further, the method for evaluating the prediction accuracy of the crankshaft roundness runout deformation prediction model in step seven is:

[0072] The stress gradient predicted by the crankshaft residual stress distribution model evaluation model is compared with the X-ray diffraction depth scanning result, and the allowable error is ≤10%.

[0073] The deformation amount of the comparative finite element calculation is compared with the measured value of the ultrasonic critical refraction longitudinal wave detection, and the allowable error is ≤10%.

[0074] Further, in step eight, the implementation method for optimizing the diesel engine crankshaft machining process is as follows:

[0075] According to the circumferential stress concentration area identified by the crankshaft residual stress distribution model, the rolling strengthening process parameters are recommended to be adjusted, and the specific implementation method is as follows:

[0076] Increase the rolling pressure to introduce residual compressive stress in the surface layer to offset the tensile stress concentration under working load, and make the fatigue crack in the subsurface during use, reducing the crack propagation speed.

[0077] Control the rolling pressure value to avoid excessive plastic deformation leading to surface micro-cracks. Optimize the rolling path to cover the circumferential stress concentration area to ensure the uniformity of compressive stress distribution.

[0078] Further, in step eight, the implementation method for optimizing the diesel engine crankshaft machining process is as follows:

[0079] According to the tensile stress distribution law in the depth direction, the grinding process parameters are optimized, and the specific implementation method is as follows:

[0080] Through active control of residual stress in the grinding process, reduce the grinding feed rate to reduce the tensile stress peak value in the depth direction, while ensuring that the surface roughness is in the appropriate interval. Use gradient cooling strategy to increase the cooling fluid flow rate at the initial stage of grinding to suppress the accumulation of surface layer thermal stress. Control the wheel linear speed to balance the grinding efficiency and residual stress suppression demand.

[0081] Taking a certain type of engine 16V crankshaft as the research object, the crankshaft after rough milling is selected as the first process to establish the distribution model. Based on the design drawings and in-depth communication with process personnel, the entity geometric model is accurately reconstructed, and ANSYS SpaceClaim 2024 is used for pre-processing of the crankshaft model: repair geometric defects, extract key areas such as main journal, connecting rod journal and transition fillet, and optimize the topological structure to adapt to the subsequent analysis requirements.

[0082] Subsequently, the pre-processed model is imported into HyperMesh 2021 to implement meshing: focus on refining the main journal, connecting rod journal and fillet area to better reflect the gradient distribution characteristics of residual stress and capture the small changes of residual stress. The main journal and connecting rod journal construct boundary layer mesh (minimum size 0.4mm), and apply C3D8R hexahedral elements to capture gradient characteristics. The fillet area is analyzed by radial encryption of stress attenuation. The total number of elements of the crankshaft finite element model after meshing is about 7.3 million. The crankshaft finite element model after meshing is shown in Figure 3 .

[0083] After the meshing is completed, the generated inp file is imported into Abaqus 2021 software, and the corresponding material parameters and boundary conditions are given to perform the first process (rough milling) to introduce residual stress. The implantation of residual stress is realized through the Abaqus subroutine function developed independently. The specific method is based on the measured data of the residual stress of the crankshaft after rough milling. The arrangement of the residual stress measuring points is shown in Figure 4 Figure 4 The orange font in the middle represents the connecting rod journal number, and the black font represents the main journal number. The code is written to use bilinear interpolation for interpolation operation, which is loaded as the initial stress field to the finite element model. To ensure the reliability of the model, the simulated and measured values at the main journal and connecting rod journal cross-section (0°, 90°, 180° and 270° phase angle) are compared, and the stress deviation of each depth point is required to be ≤10%. If local tolerance is exceeded, the interpolation smoothing factor is corrected in reverse or the boundary conditions are optimized until the accuracy requirements are met, thereby completing the residual stress analysis model of the key parts of the crankshaft under the first process with high fidelity. The residual stress gradient distribution and deformation data are extracted using python post-processing technology to study the residual stress distribution law under the first process and establish the residual stress distribution model under the first process, as shown in Figure 5 Figure 6

[0084] The simulated and measured values of the residual stress distribution model are compared, and the values taken during measurement are similar. Four measuring points are taken at each journal, with an interval of 90 degrees between each measuring point. Each measuring point corresponds to three depths.

[0085] During the manufacturing and service of the crankshaft, the residual stress distribution and structural deformation of the key parts have a crucial impact on the overall performance. Based on the established residual stress distribution model of the key parts of the crankshaft and the crankshaft round runout deformation prediction model, further research on the correlation model of the two has profound theoretical and practical significance.

[0086] To establish this correlation model, the residual stress distribution model and the crankshaft round runout deformation prediction model need to be coupled through the stress-strain relationship. The stress-strain relationship is one of the core relationships in material mechanics, which describes the corresponding law between stress and strain of materials under force action. Through this relationship, the change of residual stress can be linked to the deformation of the structure. When residual stress redistributes, the corresponding strain change can be deduced according to the stress-strain relationship, and then reflected to the structural deformation. Conversely, the deformation of the structure will also cause the redistribution of stress, forming a coupled system of mutual interaction. This coupling process can deeply reveal the essential correlation between stress redistribution and structural deformation in the processing process, i.e. they are not isolated, but a dynamic process of mutual influence and restriction.

[0087] ​​​On the basis of establishing the coupling relationship, the influence of residual stress redistribution on the crankshaft circular runout deformation can be further clarified by quantifying the relationship between the residual stress gradient and the circular runout deformation. The residual stress gradient refers to the rate of change of residual stress in a certain direction within the material, and its size and direction directly affect the stress state within the material.

[0088] After obtaining the preliminary correlation model of residual stress redistribution at key positions of the crankshaft and the crankshaft circular runout deformation, in order to further improve the accuracy and practicality of the model, parameter fitting work needs to be carried out in combination with the measured values of structural deformation in the process.

[0089] The process of parameter fitting is essentially to use the measured data to optimize and solve the unknown parameters in the correlation model, so that the model can more accurately describe the quantitative relationship between residual stress redistribution and structural deformation. Specifically, first, according to the theoretical framework of the correlation model, the parameters to be fitted in the model need to be determined, which may involve material properties, stress-strain conversion coefficients, boundary condition influence factors, etc. Then, using appropriate fitting algorithms (such as least squares method, neural network algorithm, etc.), the measured structural deformation data is substituted into the model, and the parameter values are adjusted through iterative calculation until the deviation between the predicted results of the model and the measured values reaches the minimum. Through this process, a fitting function of the stress-deformation correlation model can be established, which can clearly express the mathematical relationship between the residual stress distribution variables at each key position of the crankshaft and the circular runout deformation variables.

[0090] In order to verify the accuracy of the correlation model established, the results of finite element analysis, the measured results of structural deformation and the predicted results of the correlation model need to be systematically compared. The results of finite element analysis are theoretical data obtained by computer simulation, which can simulate the stress and deformation state of the crankshaft machining process in a virtual environment. The measured results are real data from actual production or experiments, which are the core standard for testing the authenticity of the model. The predicted results of the correlation model are theoretical values calculated based on the fitting function. In the comparison process, the errors between the three results, including absolute error, relative error, etc., need to be calculated for each key position of the crankshaft, and the overall error distribution is statistically compared. The comparison results are shown in Figure 7 .

[0091] The above specific description further illustrates the purpose, technical solution and beneficial effects of the application. It should be understood that the above description is only a specific embodiment of the application and does not limit the protection scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft, characterized by: Includes the following steps: Step 1: After the crankshaft undergoes the first process, the residual stress discrete tensor values ​​along the layer depth and cross-sectional circumference at the main journal, connecting rod journal, and fillet are measured. Step 2: Construct a three-dimensional geometric model based on the processed dimensions, and generate a finite element model of the crankshaft based on the three-dimensional geometric model; Step 3: Further analyze the discrete tensor value of residual stress using data analysis. Perform two-dimensional interpolation calculations between known data points in the crankshaft depth and cross-sectional circumferential directions using bilinear interpolation. Establish a crankshaft residual stress distribution model by combining the local cylindrical coordinate system. Step 4: Using Abaqus finite element software and the SIGINI user subroutine, assign the residual stress discrete tensor values ​​obtained in Step 1 to the crankshaft finite element model obtained in Step 2, and output the displacement distribution data in the .odb file. Step 5: Using a Python script, read the .odb result file from Abaqus and convert the calculated displacement results into circular runout data, establishing a mapping function relationship between the circular runout data and the axial coordinates. This established mapping function quantifies the micro-deformation caused by residual stress release into circular runout data distributed along the axial direction, providing a basis for model correction and verification, and outputting the predicted circular runout deformation under the machining process. Step 6: Repeat steps 1 to 5 to obtain the circular runout data for each process. Based on the distribution law of residual stress in the crankshaft, predict the amount of circular runout deformation and obtain the prediction result of the crankshaft circular runout deformation.

2. The method of claim 1, wherein: The process also includes step seven, which involves using non-model-based data to establish residual stress values ​​at detection points and comparing them with the crankshaft runout numerical simulation prediction results obtained in step six. The error between the predicted and actual values ​​is calculated to evaluate the prediction accuracy of the crankshaft runout deformation prediction model.

3. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 2, characterized in that: It also includes step eight, which, under the premise that the verification results in step seven meet the preset accuracy requirements, optimizes the diesel engine crankshaft machining process based on the crankshaft circular runout deformation prediction results obtained in step six and the deformation amount obtained in step five.

4. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 3, characterized in that: The method for constructing the mapping function relationship between the circular runout data and the axial coordinates in step five is as follows: ① Use the python-odb library to parse .odb files and extract nodal coordinates and residual stress tensor data; ② In the circumferential-depth coordinate system, a continuous stress distribution surface is generated using cubic spline interpolation; ③ Fit the stress gradient function using multivariate nonlinear regression; Regression analysis: Establish the mapping function relationship between circular runout data and axial coordinates using multivariate nonlinear regression, in the following form: σ(r, θ) = a · e -βr + γ · sin(θ + φ) + δ · rθ Among them, α, β, γ, δ, The fitting parameters reflect the attenuation and distribution of stress with layer depth and circumferential direction.

5. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 4, characterized in that: In step five, the crankshaft circular runout deformation is obtained based on the mapping function relationship between the circular runout data and the axial coordinates. The steps are as follows: Step 5.1: Import the residual stress distribution as the initial condition into the finite element model of the crankshaft; use finite element calculation software, combined with the stress-strain property data of the material, to calculate the displacement distribution of the crankshaft under the action of residual stress field; consider the elasticity, plasticity and possible nonlinear behavior of the material under different stress levels to obtain the overall displacement field distribution of the crankshaft under the action of residual stress; the displacement field distribution of the crankshaft covers the displacement of various parts of the crankshaft, providing a comprehensive data foundation for subsequent displacement extraction and analysis; Step 5.2: After obtaining the displacement distribution of the crankshaft, further extract the displacement data of several key parts, namely, extract the radial displacement of the main journal, the tangential displacement of the connecting rod journal, and the normal displacement at the fillet. Step 5.3: Based on the displacement data of key components extracted in Step 5.2, calculate the key indicators characterizing the overall and local deformation of the crankshaft: Main journal bending deformation: Linear fitting is performed on the radial displacement of the main journal along its axial direction, and the maximum offset of the fitted straight line or curve or the radial runout relative to the ideal axis is calculated to quantify the degree of bending or eccentric deformation of the main journal; Connecting rod journal torsional deformation: The distribution pattern of the tangential displacement of the connecting rod journal in the circumferential direction is analyzed. The difference in tangential displacement at different phase angles is calculated, especially the displacement difference relative to the 180-degree position point, or the gradient rate of change of its tangential displacement is calculated to quantify the torsional deformation of the connecting rod journal relative to the main journal axis caused by residual stress; Fillet deformation: The deviation of the normal displacement at the fillet is calculated directly using the extracted normal displacement data at the fillet relative to the normal displacement of the adjacent non-fillet region. This deviation directly reflects the local geometric change of the fillet caused by residual stress and is a key indicator for assessing the micro-deformation and fatigue sensitivity of the stress concentration area.

6. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 5, characterized in that: The method for evaluating the prediction accuracy of the crankshaft circular runout deformation prediction model described in step seven is as follows: The stress gradient predicted by the crankshaft residual stress distribution model was compared with the X-ray diffraction layer depth scan results, with an allowable error of ≤10%. The allowable error is ≤10% when comparing the deformation calculated by finite element method with the measured value of ultrasonic critical refraction longitudinal wave detection.

7. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 6, characterized in that: In step eight, the optimized diesel engine crankshaft machining process is implemented as follows: Based on the circumferential stress concentration areas identified by the crankshaft residual stress distribution model, it is recommended to adjust the rolling strengthening process parameters. The specific implementation method is as follows: Increasing the rolling pressure introduces residual compressive stress into the surface layer, counteracting the tensile stress concentration under working load, and causing fatigue cracks during use to occur on the subsurface, thus reducing the crack propagation rate. Control the rolling pressure value to avoid excessive plastic deformation that could lead to surface microcracks; optimize the rolling path to cover the circumferential stress concentration area to ensure uniform distribution of compressive stress.

8. The method for predicting the relationship between residual stress and deformation of a diesel engine crankshaft as described in claim 6, characterized in that: In step eight, the optimized diesel engine crankshaft machining process is implemented as follows: Based on the tensile stress distribution law in the depth direction, the grinding process parameters are optimized. The specific implementation method is as follows: By actively controlling residual stress during the grinding process, the grinding feed rate is reduced to decrease the peak tensile stress in the depth direction, while ensuring that the surface roughness is within a suitable range. A gradient cooling strategy is adopted to increase the coolant flow rate in the initial stage of grinding to suppress the accumulation of surface thermal stress. The grinding wheel linear speed is controlled to balance grinding efficiency and the need to suppress residual stress.

Images