A method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine.

Abstract

This invention discloses a method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine, comprising the following steps: establishing a J-C constitutive model of the cylinder head material; performing multi-process machining simulation analysis of the cylinder head using the "birth and death element" technique and the J-C constitutive model to obtain simulated residual stress values; obtaining experimental residual stress values ​​using the blind hole method, comparing the experimental values ​​with the simulated values, and controlling the error value within the allowable error range; establishing a nonlinear mapping relationship between cutting parameters and response targets based on a data regression prediction model using a BP neural network to obtain a prediction model; optimizing the cutting parameters of milling, turning, and boring processes of the cylinder head using the prediction model to predict the minimum residual stress value and the optimal cutting force; using the optimal cutting force as a boundary condition and performing machining process simulation analysis to obtain the optimized reduction range of residual stress values. This invention significantly reduces the residual stress in the cylinder head.

Description

Technical Field

[0001] This invention relates to the field of process parameter optimization technology for diesel engine cylinder heads, and more particularly to a method for optimizing process parameters to reduce residual stress in the cylinder heads of large-bore diesel engines. Background Technology

[0002] As a key component of diesel engines, the cylinder head's performance and reliability directly affect the engine's service life. However, during long-term service, the firing face of a large-diameter cylinder head has a larger area than that of a traditional cylinder head, making it more susceptible to temperature and pressure fluctuations, resulting in greater thermal and mechanical loads. This makes the firing face of a large-diameter cylinder head highly prone to fatigue damage such as cracks and fractures. During cylinder head production, significant residual stress is generated both internally and externally, which is a crucial factor affecting the cylinder head's service life and safety. While the influence of process parameters on residual stress can be determined through extensive experimentation by changing cutting tools and parameters, marine diesel engine cylinder heads are large and expensive, making it difficult to verify the relationship between processing parameters and residual stress through experimentation. Therefore, simulation analysis is used to predict the relationship between processing parameters and residual stress, which can significantly reduce manufacturing costs and production cycle. However, existing diesel engine cylinder head optimization analysis suffers from the following problems:

[0003] (1) Most studies use explicit dynamic simulation to simulate dynamic processes such as high-speed cutting and high impact that occur in a short time. The calculation has good stability and high efficiency, but it is only applicable to small simulation models and only focuses on a single processing procedure. However, due to factors such as large volume, complex structure and complicated processing procedures, large-diameter cylinder heads cannot obtain accurate simulation data.

[0004] (2) The method of using intelligent algorithms to optimize machining process parameters reduces research costs. Moreover, intelligent algorithms have strong global search capabilities and can quickly find the globally optimal combination of cutting parameters. However, most studies focus on optimizing process parameters for a single process and lack multi-objective optimization analysis of process parameters for multiple processes.

[0005] Therefore, the above problems urgently need to be solved. Summary of the Invention

[0006] Purpose of the invention: The purpose of this invention is to propose a method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine, which is beneficial for reducing residual stress and improving the fatigue life and overall performance of the cylinder head.

[0007] Technical Solution: To achieve the above objectives, this invention discloses a method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine, comprising the following steps:

[0008] (1) Establish the Johnson-Cook constitutive model of the cylinder head material;

[0009] (2) The cylinder head simulation model was constructed using Solidworks software, and a "birth and death element" layer was preset on the cylinder head surface. The "birth and death element" technology and the Johnson-Cook constitutive model of the cylinder head material were used to perform a multi-process machining simulation analysis of the cylinder head, and the magnitude and distribution of the maximum principal stress and minimum principal stress in the cylinder head cutting residual stress were obtained. The simulation values ​​of the maximum principal stress and minimum principal stress at several measuring points on the cylinder head fire surface were extracted.

[0010] (3) The residual stress of the cylinder head fire surface after actual processing is measured by the blind hole method, and the test values ​​of the maximum principal stress and minimum principal stress of several measuring points on the cylinder head fire surface are calculated. The test value of the maximum principal stress of each measuring point is compared with the simulation value, and the test value of the minimum principal stress of each measuring point is compared with the simulation value. The error value is controlled within the allowable error range. When the error value between the test value and the simulation value exceeds the allowable error, return to step (2) to re-perform the multi-process processing simulation analysis of the cylinder head until the error value is controlled within the allowable error range.

[0011] (4) Based on the cutting parameters of milling, turning and boring, several sets of simulation analysis were carried out respectively. The nonlinear mapping relationship between cutting parameters and response targets was established based on the data regression prediction model of BP neural network to obtain the prediction model. The cutting parameters include cutting speed, feed rate and depth of cut, and the response targets include cutting force and residual stress.

[0012] (5) The cutting parameters of the three machining processes of cylinder head milling, turning and boring are optimized by using a prediction model. The minimum residual stress value and the optimal cutting force corresponding to the optimal cutting parameters are predicted, and the average value of cutting force in the simulation machining is extracted.

[0013] (6) Using the predicted optimal cutting force as the boundary condition, the cylinder head is subjected to a multi-process machining process simulation analysis again. The simulation results after optimization of several measuring points are compared with the simulation results before optimization to obtain the optimization reduction value of the residual stress value of each measuring point.

[0014] In step (1), the diesel engine cylinder head will generate high temperature, high strain and complex flow stress during the cutting process. The Johnson-Cook constitutive model is selected to describe the dynamic behavior of the material from low strain rate to high strain rate. The material flow stress formula of the Johnson-Cook constitutive model is as follows:

[0015]

[0016] T * =(TT) r ) / (T m -T r )

[0017] In the formula: σ e For equivalent stress, ε e For equivalent plastic strain, A is the initial yield stress, B is the hardening constant, n is the hardening exponent, C is the strain rate constant, m is the thermal softening exponent, ε is the current strain rate, ε0 is the reference strain rate, T is the room temperature, and T r For reference temperature, T m Let f be the melting temperature of the material, f be a function, and ln be a logarithmic sign.

[0018] Preferably, in step (2), the cylinder head simulation model is first constructed using Solidworks software based on the cylinder head drawings. After the model is constructed, a "life and death unit" layer is preset on the cylinder head surface using Solidworks software to represent the external structure of the cylinder head in its blank state. The actual machining process is then performed based on the cutting forces F in three directions. x F y F z Take the actual cutting force F during the machining process. x F y F z average cutting force The average cutting force is synthesized into a complete cutting force and used as the boundary condition in the cutting simulation process;

[0019] Next, the Johnson-Cook constitutive model obtained in step (1) is imported into the Ansys simulation software as a material property. Then, the cylinder head simulation model is imported, and the cylinder head simulation model is divided into tetrahedral meshes. The mesh size and constraints of the cylinder head simulation model are set, and the cutting force is set as a boundary condition on the surface to be machined. Then, the "birth and death element" technology is used to perform a multi-process machining process simulation analysis of the cylinder head. In the machining process simulation analysis, the cutting force corresponding to different processes in the actual machining is applied to each surface of the cylinder head. After the entire cylinder head cutting process is completed, several measuring points are taken on the cylinder head fire surface. The number of measuring points is >10. The location range of the measuring points is selected in the area within 15mm of the nose bridge area and the intake and exhaust ports of the cylinder head fire surface. The simulation values ​​of the maximum principal stress and minimum principal stress of each measuring point are extracted.

[0020] Furthermore, the "life and death unit" layer preset on the cylinder head surface in step (2) refers to the preset roughing and finishing layers of the "life and death unit" layer on the top surface, four sides, outer circle, fire surface, valve hole and fuel injection hole of the cylinder head respectively.

[0021] Furthermore, in step (2), the multi-process machining process of the diesel engine cylinder head cutting simulation is planned in sequence as follows: rough turning of the top surface, rough milling of the four sides, finish milling of the four sides, rough turning of the fire surface, finish milling of the top surface, rough turning of the outer circle, finish turning of the outer circle, finish turning of the fire surface, finish turning of the fuel injection hole, rough boring of the valve hole, and finish boring of the valve hole.

[0022] Preferably, in step (3), during the blind hole method measurement, a blind hole is first drilled on a cylinder head with residual stress. After the stress around the blind hole is released, strain is released. The released strain is measured using a strain gauge and substituted into formula (1) to calculate the magnitude of the residual stress at the measuring point:

[0023] In the formula: σ1 and σ2 are the minimum and maximum principal stresses, respectively; ε1, ε2, and ε3 are the released strains measured by strain gauges No. 1, No. 2, and No. 3, respectively; k1 and k2 are the strain release coefficients; and θ is the angle between σ2 and strain gauge No. 1.

[0024] The strain release coefficients k1 and k2 of the material are related to the size of the blind hole and the mechanical properties of the material. In the case of drilling a blind hole, the strain release coefficients k1 and k2 are obtained through Kirsch theory:

[0025] In the formula: v is the Poisson's ratio of the material, E is the elastic modulus of the material, R is the drilling radius, and r1 and r2 are the inner and outer radii of the strain gauge;

[0026] Data was collected from several measuring points on the cylinder head fire surface using the blind hole method. The residual stress values ​​of the measuring points were calculated using formula (1). Several sets of residual stress values ​​were arranged in order of magnitude. The maximum principal stress and minimum principal stress of at least three sets of values ​​with larger values ​​were compared. The experimental values ​​and simulation values ​​of the residual stress of the measuring points on the cylinder head fire surface were compared. The error value was controlled within the allowable error range. When the error value between the experimental value and the simulation value exceeded the allowable error, the process was returned to step (2) to adjust the size of the mesh in the cylinder head simulation model mesh generation, the constraint conditions of the cylinder head simulation model and / or increase the number of measuring points. The multi-process machining process simulation analysis of the cylinder head was re-performed until the error value was controlled within the allowable error range.

[0027] Furthermore, in step (4), the cutting force and residual stress values ​​of the cylinder head under different cutting speeds, feed rates and cutting depths in the three machining processes of milling, turning and boring are first calculated using simulation. The point values ​​for simulation calculation are set according to the range of cutting parameters. Several sets of simulation analysis are performed according to the cutting parameters of the three processes to obtain the corresponding simulation data. Then, the simulation data is input into the data regression prediction model based on BP neural network to establish the nonlinear mapping relationship between cutting parameters and cutting force and residual stress to obtain the prediction model.

[0028] Furthermore, the point value setting for simulation calculation based on the range of cutting parameters in step (4) refers to:

[0029] In milling cutter machining: cutting speed v = 120 / 180 / 240 / 300 m / min, feed rate a = 0.05 / 0.1 / 0.15 / 0.2 mm, depth of cut d = 1 / 3 mm; in turning tool machining: v = 100 / 200 / 300 / 400 m / min, a = 0.05 / 0.1 / 0.15 / 0.2 mm, d = 1 / 3 mm; in boring tool machining: v = 30 / 50 / 70 / 90 m / min, a = 0.05 / 0.1 / 0.15 / 0.2 mm, d = 1 / 3 mm.

[0030] Preferably, when optimizing the cutting parameters in step (5), the optimization variables include cutting speed v, feed rate a and cutting depth d. Milling, turning and boring correspond to different ranges of cutting parameter values. The optimization targets are cutting force and residual stress. The minimum residual stress and optimal cutting force are predicted by the prediction model constructed by MATLAB software.

[0031] Furthermore, the different ranges of cutting parameters in step (5) refer to:

[0032] In milling cutter machining: cutting speed v = 120-300 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm; In turning tool machining: constraint conditions: cutting speed v = 100-400 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm; In boring tool machining: constraint conditions: cutting speed v = 30-90 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm.

[0033] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: The present invention comprehensively considers the optimization of the magnitude and distribution of the residual stress in the cylinder head by considering the cutting parameters of multiple processes, uses the "birth and death unit" technology to perform full-process cutting simulation analysis of the cylinder head, uses the blind hole method to verify and correct the simulation results of the cutting simulation analysis, and then performs multiple sets of simulation analysis to obtain the mapping relationship between cutting parameters and residual stress. Based on the BP neural network, a nonlinear mapping relationship between cutting parameters and response target is established, and a high-precision prediction model is used to predict the optimal cutting parameters, which is beneficial to reduce residual stress and improve the fatigue life and overall performance of the cylinder head. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the process in this invention;

[0035] Figure 2This is an experimental schematic diagram of the 105A WANCE mechanical testing machine used in this invention;

[0036] Figure 3 This is a diagram of the split Hopkinson bar experimental platform in this invention;

[0037] Figure 4 This is a schematic diagram of the multi-process machining flow of the cylinder head in this invention;

[0038] Figure 5 This is a platform diagram of the blind hole method experiment in this invention;

[0039] Figure 6 This is a schematic diagram of the strain rose structure used in the blind hole method of this invention. Detailed Implementation

[0040] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0041] like Figure 1 As shown, the present invention discloses a method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine, comprising the following steps:

[0042] (1) The Johnson-Cook constitutive model of QT400 material for cylinder head was obtained by room temperature quasi-static tensile and split Hopkinson bar tests.

[0043] In the study of the constitutive model of the cylinder head material, a 105A WANCE mechanical testing machine was used to conduct quasi-static tensile tests at room temperature to investigate the influence of plastic strain on the material's flow stress behavior. Figure 2 As shown; to obtain the strain rate effect of the material at different temperatures, a split Hopkinson pressure bar was used to conduct impact tests on the cylinder head material at different strain rates, such as... Figure 3 As shown; the Johnson-Cook constitutive model of the QT400 material for the cylinder head was obtained by fitting room temperature quasi-static tensile test and Hopkinson bar test.

[0044] During the machining of diesel engine cylinder heads, the material experiences high temperatures, high strains, and complex flow stresses. The Johnson-Cook constitutive model is used to describe the dynamic behavior of QT400 material from low strain rate to high strain rate. The material flow stress formula of the Johnson-Cook constitutive model is as follows:

[0045]

[0046] T * =(TT) r ) / (T m -T r )

[0047] In the formula: σe For equivalent stress, ε e For equivalent plastic strain, A is the initial yield stress, B is the hardening constant, n is the hardening exponent, C is the strain rate constant, m is the thermal softening exponent, ε is the current strain rate, ε0 is the reference strain rate, T is the room temperature, and T r For reference temperature, T m Let f be the melting temperature of the material, f be a function, and ln be a logarithmic sign.

[0048] (2) The cylinder head simulation model was constructed using Solidworks software, and a "birth and death element" layer was preset on the cylinder head surface. The "birth and death element" technology and the Johnson-Cook constitutive model of the cylinder head QT400 material were used to perform a multi-process machining process simulation analysis of the cylinder head, and the magnitude and distribution of the maximum principal stress and minimum principal stress in the cylinder head cutting residual stress were obtained. The simulation values ​​of the maximum principal stress and minimum principal stress at several measuring points on the cylinder head fire surface were extracted.

[0049] First, based on the cylinder head drawings, a simulation model of the cylinder head is constructed using Solidworks software. After construction, "dead and alive" element layers are preset on the cylinder head surface using Solidworks software to represent the shape and structure of the cylinder head in its raw state. Then, the cutting forces F in three directions during actual machining are considered. x F y F z Take the actual cutting force F during the machining process. x F y F z average cutting force The average cutting force is synthesized into a complete cutting force and used as the boundary condition in the cutting simulation process;

[0050] The "birth and death element" technique employed involves assembling a series of discrete analysis steps into a dynamic cutting process during simulation analysis. This simplifies the dynamic analysis process into numerous static analysis processes. The material actually removed is sequentially eliminated from the cutting elements through birth and death element simulation and tool path mapping. The volume of the element removed in each static analysis is calculated using the following formula:

[0051] V max =a p ×a e ×f d ×t

[0052] In the formula: V max a is the volume of material removed per unit analysis step. p For the depth of cut, a e f is the cutting width. d The feed rate is given by t, which represents the time per unit analysis step.

[0053] Pre-setting "life and death unit" layers on the cylinder head surface refers to pre-setting rough machining layers and fine machining layers on the top surface, four sides, outer circle, fire surface, valve hole and fuel injection hole of the cylinder head.

[0054] The multi-process machining flow of the diesel engine cylinder head cutting simulation is planned sequentially as follows: rough turning of the top surface, rough milling of the four sides, finish milling of the four sides, rough turning of the firing face, finish milling of the top surface, rough turning of the outer diameter, finish turning of the outer diameter, finish turning of the firing face, finish turning of the fuel injection hole, rough boring of the valve hole, and finish boring of the valve hole. The entire production process is as follows: Figure 4 As shown;

[0055] Next, the Johnson-Cook constitutive model measured in step (1) is imported into the Ansys simulation software as a material property. Then, the cylinder head simulation model is imported, and the cylinder head simulation model is divided into tetrahedral meshes with a mesh size of 10mm. The constraints of the cylinder head simulation model are set, and the cutting force is set as a boundary condition on the surface to be machined. Then, the "birth and death element" technology is used to perform a multi-process machining process simulation analysis on the cylinder head. Finally, the magnitude and distribution of the maximum and minimum principal stresses in the cylinder head cutting residual stress are obtained. In the machining process simulation analysis, the cutting forces corresponding to different processes in the actual machining are applied to each surface of the cylinder head. The entire cylinder head cutting process can be divided into three parts: cutting force application, machining completion, and cooling. After the entire cylinder head cutting process is completed, 17 measuring points are taken on the cylinder head fire surface. The location range of the measuring points is selected in the area within 15mm of the nose bridge area and the intake and exhaust ports of the cylinder head fire surface. 17 measuring points are evenly distributed and the simulation values ​​of the maximum and minimum principal stresses of each measuring point are extracted.

[0056] (3) The residual stress of the cylinder head fire surface after actual processing is measured by the blind hole method, and the test values ​​of the maximum principal stress and minimum principal stress at several measuring points on the cylinder head fire surface are calculated. The test value of the maximum principal stress at each measuring point is compared with the simulation value, and the test value of the minimum principal stress at each measuring point is compared with the simulation value. The error value is controlled within the allowable error range of 20% to prove the accuracy of the simulation. When the error value between the test value and the simulation value exceeds the allowable error of 20%, the process is returned to step (2) to re-analyze the multi-process processing flow of the cylinder head until the error value is controlled within the allowable error range of 20%.

[0057] The stress on the cylinder head firing surface was measured using the blind hole method. The accuracy of the simulation calculation results was verified through experiments, and it was also verified whether the simulation model could truly reflect the actual physical phenomena. This is beneficial for later use of simulation calculations to explore the impact of different cutting parameters on machining quality. Based on the stress testing standards, the distance between each measuring point should be more than 6 times the diameter of the blind hole, i.e., more than 12 mm. During the test, the test positions were reasonably arranged according to the specific dimensions of the machine body, ensuring that the measuring points were at least 15 mm away from the edges of the measured surface and at least 12 mm apart from each measuring point.

[0058] The experimental platform for the blind hole method is as follows: Figure 5 As shown, the main steps for measuring residual stress at the fire face using the blind hole method are: sanding and cleaning the measuring point with sandpaper; attaching the strain gauge (TJ120-1.5 model) while wearing clean gloves and allowing it to cool for 10-15 minutes; connecting the data cable using solderless butt-connected terminals; drilling holes in the target area using a drill bit holder (GB / T6135.2-2008 model); recording data using software (using the matching USB-1608G dynamic signal acquisition and analysis system); and finally processing and analyzing the data.

[0059] In the blind hole method of measurement, a blind hole is first drilled in a part with residual stress. After the stress around the blind hole is released, strain is released, and a method such as... Figure 6 The strain gauge shown is used to measure the released strain. Substituting this into formula (1), the magnitude of the residual stress at the measuring point is calculated:

[0060] In the formula: σ1 and σ2 are the minimum and maximum stresses, respectively; ε1, ε2, and ε3 are the released strains measured by strain gauges 1, 2, and 3, respectively; k1 and k2 are the strain release coefficients; and θ is the angle between σ2 and strain gauge 1. Figure 6 As shown;

[0061] The strain release coefficients k1 and k2 of the material are related to the size of the blind hole and the mechanical properties of the material. In the case of drilling, the strain release coefficients k1 and k2 are obtained through Kirsch theory:

[0062] In the formula: v is Poisson's ratio, E is elastic modulus, R is the perforation radius, and r1 and r2 are the inner and outer radii of the strain gauge, respectively. Figure 6 As shown;

[0063] Data was collected from 17 measuring points on the cylinder head fire surface using the blind hole method. The residual stress values ​​of the measuring points were calculated using formula (1). The 17 sets of residual stress values ​​were arranged in order of magnitude. The maximum principal stress and minimum principal stress of the four sets with larger values ​​were compared to verify the accuracy of the simulation. The residual stress values ​​of the test data and simulation results of the measuring points on the cylinder head fire surface were compared. The error value was controlled within 20%. The error of the maximum stress value in the area prone to fatigue failure was within an acceptable range, indicating that the simulation calculation can accurately reflect the real residual stress of the machined parts.

[0064] When the error between the experimental data and the simulation results exceeds 20%, the error can be reduced by the following methods: return to step (2) to adjust the size of the mesh in the cylinder head simulation model mesh generation, the constraint conditions of the cylinder head simulation model and / or increase the number of measuring points; re-perform the multi-process machining process simulation analysis of the cylinder head until the error value is controlled within the allowable error range.

[0065] (4) Based on the cutting parameter settings of milling, turning and boring, 32 sets of simulation analysis were performed for each of the three machining processes, for a total of 96 sets of simulation analysis. The nonlinear mapping relationship between cutting parameters and response targets was established based on the data regression prediction model of BP neural network to obtain a high-precision prediction model. The cutting parameters include cutting speed, feed rate and depth of cut, and the response targets include cutting force and residual stress.

[0066] First, simulation calculations were used to determine the cutting force and residual stress values ​​of the cylinder head under different cutting speeds, feed rates, and depths of cut in milling, turning, and boring processes. The simulation calculation point values ​​were set according to the range of cutting parameters. In milling: cutting speed v = 120 / 180 / 240 / 300 m / min, feed rate a = 0.05 / 0.1 / 0.15 / 0.2 mm / r, depth of cut d = 1 / 3 mm; in turning: v = 100 / 200 / 300 / 400 m / min, a = 0. 0.05 / 0.1 / 0.15 / 0.2mm / r, d=1 / 3mm; In boring tool machining: v=30 / 50 / 70 / 90m / min, a=0.05 / 0.1 / 0.15 / 0.2mm / r, d=1 / 3mm, 32 sets of simulation analysis were performed for each of the three processes to obtain the corresponding simulation data; then the simulation data were input into a data regression prediction model based on BP neural network to establish a nonlinear mapping relationship between cutting parameters and response target, and obtain a high-precision prediction model;

[0067] A backpropagation (BP) neural network is a multi-layer feedforward neural network that calculates the error between the output layer and the true value, and then backpropagates this error to each layer of the network, thereby updating the network's weights and biases, so that the network's output gradually approximates the true value. In the BP neural network model, the number of nodes in the input and output layers is determined based on actual research, and the number of nodes in the hidden layers is determined according to the following formula:

[0068]

[0069] In the formula: h is the number of nodes in the hidden layer, i is the number of nodes in the input layer, o is the number of nodes in the output layer, and c is a random constant with a value range of [1, 10].

[0070] During the forward propagation of the network model, the input training set is first normalized using the following formula:

[0071]

[0072] In the formula: x is the normalized value, x1 is the original data, x max x min These represent the maximum and minimum values ​​of the original data, respectively.

[0073] The output of each neuron in the hidden layer is calculated using the following formula:

[0074]

[0075] In the formula: H i Here, f is the hidden layer output, f is the hidden layer activation function, ω is the weight of the two nodes, and h is the output of the hidden layer. i is the hidden layer threshold, i is the number of hidden layer nodes, i = 1, 2, ..., q, j is the number of output layer nodes, j = 1, 2, ..., m;

[0076] The output of each neuron in the output layer is calculated using the following formula:

[0077]

[0078] In the formula: O k For output layer output, a k The threshold value is k, where k is the number of nodes in the output layer, k = 1, 2, ..., l;

[0079] In backpropagation, the output error formula is as follows:

[0080] e k =d k -O k

[0081] In the formula: e k For the output error, d kThe expected output;

[0082] Adjust the network connection weights and thresholds according to the following formula:

[0083]

[0084] a k =a k +e k

[0085] ω ij =ω ij +ηH i e k

[0086] In the formula: η is the learning rate;

[0087] A backpropagation (BP) neural network model was constructed using MATLAB software, and the accuracy of the model was verified. First, 28 sets of simulation results from 32 sets of milling cutter machining simulations were extracted as the training set, and the remaining 4 sets were used as the test set. Then, the data was normalized with a scaling interval of [0,1] to improve the model's stability and convergence speed. Next, the feedforward neural network was called using the neural network toolbox in MATLAB. Finally, error calculation and data inverse normalization were performed. The iteration stopped after the 9th iteration when the error accuracy (RMSE < 10) was met. The results included regression plots for training, verification, testing, and the overall model. The RMSE of the cutting force in the network model calculation was 4.4 for the training set and 7.3 for the test set. The RMSE of the residual stress in the network model calculation was 2.6 for the training set and 6.2 for the test set. The accuracy of the prediction model was determined based on the calculated error values. This model can be used to solve for the optimal machining process scheme in the later stages.

[0088] (5) The cutting parameters of the three machining processes of cylinder head milling, turning and boring are optimized by using a prediction model. The minimum residual stress value and the optimal cutting force corresponding to the optimal machining parameters are predicted, and the average value of cutting force in the simulation machining is extracted.

[0089] When optimizing cutting parameters, the optimization variables include cutting speed v, feed rate a, and depth of cut d. Milling, turning, and boring correspond to different ranges of cutting parameter values. In milling: cutting speed v = 120-300 m / min, feed rate a = 0.05-0.2 mm, depth of cut d = 1-3 mm; in turning: constraints include cutting speed v = 100-400 m / min, feed rate a = 0.05-0.2 mm, depth of cut d = 1-3 mm; in boring: constraints include cutting speed v = 30-90 m / min, feed rate a = 0.05-0.2 mm, depth of cut d = 1-3 mm. The optimization objectives are cutting force and residual stress. The minimum residual stress and the corresponding optimal cutting force are predicted using a prediction model built with MATLAB software.

[0090] Taking cutting force and residual stress as optimization objectives, the prediction formula for the objective function is as follows:

[0091] y = sim(net, A)

[0092] In the formula: y is the predicted value, net is the residual stress prediction model of the trained BP surface, and A is the input variable matrix.

[0093] (6) Using the predicted optimal cutting force as the boundary condition, the cylinder head is subjected to a multi-process machining process simulation analysis again. The simulation results after optimization of several measuring points are compared with the simulation results before optimization to obtain the residual stress value of each measuring point. The maximum optimization effect is that the residual stress value is reduced by 20.5%.

[0094] This invention comprehensively considers the optimization of the magnitude and distribution of residual stress in cylinder head cutting by considering the cutting parameters of multiple processes. It uses the "birth and death unit" technology to perform full-process cutting simulation analysis of the cylinder head, uses the blind hole method to verify and correct the simulation results of the cutting simulation analysis, and then performs multiple sets of simulation analysis to obtain the mapping relationship between cutting parameters and residual stress. Based on the BP neural network, a nonlinear mapping relationship between cutting parameters and response target is established, and a high-precision prediction model is used to predict the optimal cutting parameters, which is beneficial to reduce residual stress and improve the fatigue life and overall performance of the cylinder head.

Claims

1. A method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine, characterized in that, Includes the following steps: (1) Establish the Johnson-Cook constitutive model of the cylinder head material; (2) The cylinder head simulation model was constructed using Solidworks software, and a "birth and death element" layer was preset on the cylinder head surface. The "birth and death element" technology and the Johnson-Cook constitutive model of the cylinder head material were used to perform a multi-process machining process simulation analysis of the cylinder head, and the magnitude and distribution of the maximum principal stress and minimum principal stress in the cylinder head cutting residual stress were obtained. The simulation values ​​of the maximum principal stress and minimum principal stress at several measuring points on the cylinder head fire surface were extracted. (3) The residual stress of the cylinder head fire surface after actual processing is measured by the blind hole method, and the test values ​​of the maximum principal stress and minimum principal stress of several measuring points on the cylinder head fire surface are calculated. The test value of the maximum principal stress of each measuring point is compared with the simulation value, and the test value of the minimum principal stress of each measuring point is compared with the simulation value. The error value is controlled within the allowable error range. When the error value between the test value and the simulation value exceeds the allowable error, return to step (2) to re-perform the multi-process processing simulation analysis of the cylinder head until the error value is controlled within the allowable error range. (4) Based on the cutting parameters of milling, turning and boring, several sets of simulation analysis were carried out respectively. The nonlinear mapping relationship between cutting parameters and response targets was established based on the data regression prediction model of BP neural network to obtain the prediction model. The cutting parameters include cutting speed, feed rate and depth of cut, and the response targets include cutting force and residual stress. (5) The cutting parameters of the three machining processes of cylinder head milling, turning and boring are optimized by using a prediction model. The minimum residual stress value and the optimal cutting force corresponding to the optimal cutting parameters are predicted, and the average value of cutting force in the simulation machining is extracted. (6) Using the predicted optimal cutting force as the boundary condition, the cylinder head is subjected to a multi-process machining process simulation analysis again. The simulation results after optimization of several measuring points are compared with the simulation results before optimization to obtain the optimization reduction value of the residual stress value of each measuring point.

2. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 1, characterized in that, In step (1), the diesel engine cylinder head will generate high temperature, high strain and complex flow stress during the cutting process. The Johnson-Cook constitutive model is selected to describe the dynamic behavior of the material from low strain rate to high strain rate. The material flow stress formula of the Johnson-Cook constitutive model is as follows: T * =(T-T r ) / (T m -T r ) In the formula: σ e For equivalent stress, ε e For equivalent plastic strain, A is the initial yield stress, B is the hardening constant, n is the hardening exponent, C is the strain rate constant, m is the thermal softening exponent, ε is the current strain rate, ε0 is the reference strain rate, T is the room temperature, and T r For reference temperature, T m Let f be the melting temperature of the material, f be a function, and ln be a logarithmic sign.

3. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 1, characterized in that, In step (2), the cylinder head simulation model is first constructed using Solidworks software based on the cylinder head drawings. After the model is constructed, a "life and death unit" layer is preset on the cylinder head surface using Solidworks software to represent the shape and structure of the cylinder head in its blank state. The actual machining process is then performed based on the cutting forces F in three directions. x F y F z Take the actual cutting force F during the machining process. x F y F z average cutting force The average cutting force is synthesized into a complete cutting force and used as the boundary condition in the cutting simulation process; Next, the Johnson-Cook constitutive model obtained in step (1) is imported into the Ansys simulation software as a material property. Then, the cylinder head simulation model is imported, and the cylinder head simulation model is divided into tetrahedral meshes. The mesh size and constraints of the cylinder head simulation model are set, and the cutting force is set as a boundary condition on the surface to be machined. Then, the "birth and death element" technology is used to perform a multi-process machining process simulation analysis of the cylinder head. In the machining process simulation analysis, the cutting force corresponding to different processes in the actual machining is applied to each surface of the cylinder head. After the entire cylinder head cutting process is completed, several measuring points are taken on the cylinder head fire surface. The number of measuring points is >10. The location range of the measuring points is selected in the area within 15mm of the nose bridge area and the intake and exhaust ports of the cylinder head fire surface. The simulation values ​​of the maximum principal stress and minimum principal stress of each measuring point are extracted.

4. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 3, characterized in that, In step (2), the "life and death unit" layer on the cylinder head surface refers to the "life and death unit" layer with rough machining layer and fine machining layer respectively on the top surface, four sides, outer circle, fire surface, valve hole and fuel injection hole of the cylinder head.

5. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 3, characterized in that, In step (2), the multi-process machining process of the diesel engine cylinder head cutting simulation is planned as follows: rough turning of the top surface, rough milling of the four sides, finish milling of the four sides, rough turning of the fire surface, finish milling of the top surface, rough turning of the outer circle, finish turning of the outer circle, finish turning of the fire surface, finish turning of the fuel injection hole, rough boring of the valve hole and finish boring of the valve hole.

6. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 3, characterized in that, In step (3), during the blind hole method measurement, a blind hole is first drilled on a cylinder head with residual stress. After the stress around the blind hole is released, strain is released. The released strain is measured using a strain gauge and substituted into formula (1) to calculate the magnitude of the residual stress at the measuring point: In the formula: σ1 and σ2 are the minimum and maximum principal stresses, respectively; ε1, ε2, and ε3 are the released strains measured by strain gauges No. 1, No. 2, and No. 3, respectively; k1 and k2 are the strain release coefficients; and θ is the angle between σ2 and strain gauge No.

1. The strain release coefficients k1 and k2 of the material are related to the size of the blind hole and the mechanical properties of the material. In the case of drilling a blind hole, the strain release coefficients k1 and k2 are obtained through Kirsch theory: In the formula: v is the Poisson's ratio of the material, E is the elastic modulus of the material, R is the drilling radius, and r1 and r2 are the inner and outer radii of the strain gauge; Data was collected from several measuring points on the cylinder head fire surface using the blind hole method. The residual stress values ​​of the measuring points were calculated using formula (1). Several sets of residual stress values ​​were arranged in order of magnitude. The maximum principal stress and minimum principal stress of at least three sets of values ​​with larger values ​​were compared. The experimental values ​​and simulation values ​​of the residual stress of the measuring points on the cylinder head fire surface were compared. The error value was controlled within the allowable error range. When the error value between the experimental value and the simulation value exceeded the allowable error, the process was returned to step (2) to adjust the size of the mesh in the cylinder head simulation model mesh generation, the constraint conditions of the cylinder head simulation model and / or increase the number of measuring points. The multi-process machining process simulation analysis of the cylinder head was re-performed until the error value was controlled within the allowable error range.

7. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 6, characterized in that, In step (4), the cutting force and residual stress values ​​of the cylinder head under different cutting speeds, feed rates and cutting depths in the three machining processes of milling, turning and boring are first calculated using simulation. The point values ​​for simulation calculation are set according to the range of cutting parameters. Several sets of simulation analysis are performed according to the cutting parameters of the three processes to obtain the corresponding simulation data. Then, the simulation data is input into a data regression prediction model based on a BP neural network to establish a nonlinear mapping relationship between cutting parameters, cutting force, and residual stress, thus obtaining the prediction model.

8. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 7, characterized in that, The point value setting in step (4) for simulation calculation based on the range of cutting parameters refers to: In milling cutter machining: cutting speed v = 120 / 180 / 240 / 300 m / min, feed rate a = 0.05 / 0.1 / 0.15 / 0.2 mm / r, depth of cut d = 1 / 3 mm; In lathe tool machining: v = 100 / 200 / 300 / 400 m / min, a = 0.05 / 0.1 / 0.15 / 0.2 mm / r, d = 1 / 3 mm; In boring tool machining: v = 30 / 50 / 70 / 90 m / min, a = 0.05 / 0.1 / 0.15 / 0.2 mm / r, d = 1 / 3 mm.

9. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 8, characterized in that, In step (5), when optimizing the cutting parameters, the optimization variables include cutting speed v, feed rate a, and depth of cut d. Milling, turning, and boring correspond to different ranges of cutting parameter values. The optimization targets are cutting force and residual stress. The minimum residual stress and optimal cutting force are predicted by the prediction model built using MATLAB software.

10. The method for optimizing process parameters to reduce residual stress in the cylinder head of a large-bore diesel engine according to claim 9, characterized in that, The different ranges of cutting parameters in step (5) refer to: In milling cutter machining: cutting speed v = 120-300 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm; In turning tool machining: constraint conditions: cutting speed v = 100-400 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm; In boring tool machining: constraint conditions: cutting speed v = 30-90 m / min, feed rate a = 0.05-0.2 mm / r, depth of cut d = 1-3 mm.

Images