Adaptive adjustment method for crankshaft clamping center support position

Abstract

The application discloses a self-adaptive adjustment method of a center frame supporting position of a crankshaft, and comprises the following steps: S1, establishing a finite element model of crankshaft clamping, and obtaining crankshaft deformation data; S2, obtaining a crankshaft simulation data set; S3, establishing a center frame supporting position-crankshaft arm distance difference proxy model; and S4, optimizing and adjusting the center frame supporting position. The application solves the problems of deformation of the crankshaft in clamping and low adjustment efficiency in grinding processing, and realizes that the arm distance difference of the crankshaft meets the subsequent processing conditions under the adjustment of the center frame position for a small number of times. The stress field and deformation of the crankshaft affected by the center frame position in the clamping process can be analyzed by using the method, the stress concentration area of the crankshaft affected by the clamping and the change rule of the crankshaft crank deformation within one rotation of the crankshaft are obtained. The method provided by the application is suitable for the rapid adjustment of the center frame position of other models of crankshafts in the clamping process of grinding, and the control of the arm distance difference of the crankshaft is realized.

Description

Technical Field

[0001] This invention relates to an adaptive adjustment method for the support position of the crankshaft clamping center support, belonging to the field of machining. Background Technology

[0002] The crankshaft is a critical component of a diesel engine. Its overall structure is complex and asymmetrical, making its machining quality the most difficult to guarantee among the key components. In actual grinding, the crankshaft is prone to complex elastic deformation due to its own weight and the clamping force of the grinding machine fixture, severely affecting the grinding accuracy. To reduce the elastic deformation of the crankshaft and ensure the coaxiality of the main journals, a center rest is needed to support the crankshaft's main journals during grinding, and the position of the center rest support must be adjusted according to the crankshaft's deformation.

[0003] Regarding the issue of crankshaft deformation and adjustment during grinding, in actual grinding of large marine crankshafts, the deformation is usually measured by the difference in crankshaft arm distance. A crank rest gauge is used to measure this difference, and the center support position is repeatedly adjusted based on the worker's experience to keep the arm distance difference within the required machining range. However, existing methods require a large amount of manpower and have low adjustment efficiency, failing to achieve rapid adjustment of the center support position. Summary of the Invention

[0004] Purpose of the invention: To address the shortcomings of existing technologies, this invention provides an adaptive adjustment method for the support position of the crankshaft clamping center frame. By optimizing and iteratively adjusting the position of the center frame, this invention enables rapid adjustment of the support position of the crankshaft clamping center frame, controlling the deformation of the crankshaft during clamping within the required range, while improving production efficiency.

[0005] Technical solution: An adaptive adjustment method for the support position of the crankshaft clamping center support, comprising the following steps:

[0006] S1. Establish a crankshaft clamping finite element model: Based on the crankshaft size and structure and the center frame structure, establish a crankshaft clamping finite element model, analyze the crankshaft deformation during the clamping and adjustment process and the center frame support position adjustment process, and obtain crankshaft deformation data, that is, the deformation amount at each crank arm of the crankshaft.

[0007] S2. Obtain the crankshaft simulation dataset: Process the deformation data in S1 to obtain the arm distance difference data of each crankshaft crank and the corresponding center frame support position data, and establish the simulation dataset.

[0008] S3. Establish a surrogate model for the difference between the center frame support position and the crankshaft arm distance: Establish a prediction model for the arm distance difference based on a BP neural network, set the hyperparameters of the prediction model for the arm distance difference, and train the prediction model for the arm distance difference using the simulation dataset in S2. Take the center frame support position as the input and the crankshaft arm distance difference as the output to obtain a surrogate model for the prediction of the difference between the center frame support position and the crankshaft arm distance.

[0009] S4. Optimization and adjustment of the center frame support position: Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, the NSGA-II multi-objective genetic algorithm is used to establish a reverse optimization algorithm for the center frame support position, and the support position of the center frame is iteratively adjusted based on the measured crankshaft crank arm distance difference data using this reverse optimization algorithm.

[0010] Preferably, S1 includes:

[0011] S101. Analyze the structural features of the crankshaft, establish a finite element model of the crankshaft clamping based on the crankshaft size structure and center frame structure, and simplify the center frame structure into a V-block structure for easy simulation analysis.

[0012] S102. Set the material parameters of the crankshaft clamping finite element model: crankshaft density, Young's modulus, Poisson's ratio, and yield strength;

[0013] S103. Mesh the crankshaft clamping finite element model;

[0014] S104. Set the boundary conditions of the crankshaft according to the actual clamping situation to obtain the initial crankshaft clamping finite element model.

[0015] S105. Based on the initial crankshaft clamping finite element model established in S104, parameterize the model and solve it to obtain crankshaft deformation data under different center frame support positions.

[0016] In a preferred embodiment, S103 specifically comprises:

[0017] Different mesh density strategies were adopted for different regions. Tetrahedral element meshes were used for both the crankshaft and the center frame, and corresponding node sets were set for the extraction of crankshaft deformation data.

[0018] The areas requiring grid density division include the surface area where the center frame contacts the crankshaft, the area where the center frame does not contact the crankshaft, and the area where the crankshaft does not contact the center frame;

[0019] The grid size of the surface area where the center support contacts the crankshaft is 0.5-1mm, the grid size of the area where the center support does not contact the crankshaft is 10-20mm, and the grid size of the area where the crankshaft does not contact the center support is 20-30mm.

[0020] After the mesh is generated, each region contains several elements, and each element contains several nodes. The midpoints of the lowest edges of all crank arms in the nodes are used to form a node set. By processing the displacement changes of the nodes in the node set, the deformation difference at each crank arm of the crankshaft is obtained.

[0021] Preferably, in step S104, setting the boundary conditions of the crankshaft according to the actual clamping condition of the crankshaft specifically involves:

[0022] S1041. Apply a completely fixed constraint to the center hole at one end of the crankshaft, and apply a clamping force along the axial direction at the other end.

[0023] S1042. Move the center support to the set position, apply a completely fixed constraint to the center support, and adjust the clamping force applied to the crankshaft.

[0024] S1043. Constrains the movement and rotation of the center holes at both ends of the crankshaft, causing the crankshaft to rotate one revolution;

[0025] Gravity was applied during the actual crankshaft clamping process described above, with a gravitational acceleration of 9.8 m / s². 2 .

[0026] Preferably, S105 includes:

[0027] S1051. Determine the adjustment range of the center support position. Based on the settings of S101-S104, perform clamping simulation analysis on the crankshaft under unsupported conditions to obtain the crankshaft simulation deformation results.

[0028] S1052. Extract the deformation of the crankshaft at the required support position of the center frame from the simulation results, and use this as the lower limit H of the center frame support position. min The upper limit of the central frame support position is set to H. max =0;

[0029] S1053. Obtain different center support positions based on the actual crankshaft clamping conditions. Within the range of the center support positions, i.e., H... min ≤H≤H max Different combinations of center frame support positions within the specified range were selected to parameterize the initial crankshaft clamping finite element model, and the crankshaft deformation data under different center frame support positions were obtained by solving the problem.

[0030] In a preferred embodiment, S1053 specifically involves selecting a center frame support position adjustment amount δh < 0.1H. maxTo ensure that the simulation results reflect the influence of different center frame support height deviations on the crankshaft arm distance difference, the Latin hypercube sampling (LHS) method was used to select different center frame support position combinations. This was achieved by dividing the range of each variable into mutually exclusive intervals with equal probability, and then randomly selecting samples from each interval. The formula is as follows:

[0031]

[0032] In the formula, i = (1,2,…n) represents the number of variables, i.e., the number of central frames; j = (1,2,…m) represents the sampling dimension, i.e., the j-th sampling; P i -1 Let α be the probability distribution function; ij is a random number within the range [0,1]; N is the number of samples, i.e., the number of different combinations of central frame support positions;

[0033] Finally, N different combinations of center frame support positions were selected to complete the parameterization of the initial crankshaft clamping finite element model.

[0034] In a preferred embodiment, S2 specifically comprises:

[0035] The crankshaft deformation data obtained from the S1 simulation is extracted, and the axial displacement U3 of each node in the S1 node set is extracted. The arm distance difference of each crank of the crankshaft is obtained by processing the node data under different rotation angles. The arm distance difference data of each crank under different actual crankshaft clamping conditions and the corresponding center frame adjustment position data are organized into a simulation dataset.

[0036] In a preferred embodiment, S3 specifically includes:

[0037] A backpropagation (BP) neural network model was established, with the number of center supports and crankshaft cranks used as the input and output, respectively. Data from the simulation dataset was randomly selected in a 7:1:2 ratio as the training set, validation set, and test set for evaluating the center support position-crankshaft arm distance difference surrogate model. The hyperparameters of the BP neural network model were set, including the maximum number of iterations, learning rate, and minimum training error. The structure of the BP neural network model was also defined. The model was trained on the training set, and the trained model was used to predict the crankshaft arm distance difference, providing a basis for solving the optimal center support position. The trained surrogate model for the center support position-crankshaft arm distance difference is as follows:

[0038]

[0039] is the predicted crankshaft arm distance difference, net is the trained BP neural network model, and H is the center frame support position.

[0040] Preferably, S4 includes:

[0041] S401. Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, it is brought into the NSGA-Ⅱ optimization algorithm, and the condition of ensuring that the sum of the actual support position of the current center frame and the crankshaft rotation axis is minimized is added to complete the establishment of the reverse solution optimization algorithm for the center frame support position.

[0042] S402. The actual crankshaft arm distance difference is measured to determine whether the actual crankshaft arm distance difference meets the processing conditions. If it does not meet the conditions, the actual crankshaft arm distance difference is input into the reverse optimization algorithm for the center frame support position to obtain the current actual support position of the center frame. The current actual support position of the center frame is adjusted. After adjustment, the arm distance difference data of each crankshaft is measured again. If the processing conditions are not met, the data is input into the reverse optimization algorithm for iterative adjustment until the actual crankshaft arm distance difference meets the actual processing conditions.

[0043] In a preferred embodiment, S401 specifically includes:

[0044] First, we establish a mathematical problem for optimizing the center frame support position, minimizing the absolute error between the measured crankshaft arm distance difference and the predicted crankshaft arm distance difference, while simultaneously minimizing the sum of the absolute values ​​of the center frame positions corresponding to the predicted arm distance difference.

[0045] Next, constraints are set, namely, the upper and lower limits of the adjustment range of the central frame position are set, thus completing the establishment of the central frame support position optimization problem;

[0046] H min and H max These represent the lower and upper limits for center frame position adjustment, respectively, and D is the actual measured crankshaft arm distance difference. Let f1(D) be the predicted crankshaft arm distance difference, H be the center frame support position, f2(H) be the predicted crankshaft arm distance difference error, and f3(D) be the center frame support position.

[0047] Finally, the relevant parameters of the NSGA-II multi-objective genetic algorithm were set, including population size, maximum number of iterations, crossover probability and mutation probability. The optimization problem of the central support position was then incorporated into the NSGA-II multi-objective genetic algorithm, thus establishing the reverse optimization algorithm for the central support position.

[0048] Beneficial Effects: This invention solves the problems of crankshaft deformation and low adjustment efficiency during grinding clamping, achieving crankshaft arm distance difference that meets subsequent machining conditions with fewer adjustments to the center rest position. This method allows analysis of the stress field and deformation of the crankshaft under the influence of the center rest position during clamping, revealing the stress concentration areas and the variation of crankshaft crank deformation within one crankshaft rotation. The method proposed in this invention is applicable to the rapid adjustment of the center rest position during grinding clamping of other crankshaft models, enabling control of crankshaft arm distance difference. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0050] Figure 1 This is a flowchart of the method of the present invention;

[0051] Figure 2 (a)-(b) are the three-dimensional model and main dimension diagram of the crankshaft in this invention, respectively;

[0052] Figure 3 This is a diagram showing the crankshaft axis deviation under conditions without center frame support in this invention;

[0053] Figure 4 (a)-(i) are graphs showing the changes in arm distance values ​​of cranks 1 to 9 at different support positions of the crankshaft in this invention within one revolution;

[0054] Figure 5 This is a graph showing the mean square error training results of the crankshaft arm distance difference BP model in this invention;

[0055] Figure 6 This is a regression result graph of the crankshaft arm distance difference BP model in this invention;

[0056] Figure 7 This is the Pareto optimal solution set diagram of the central frame position adjustment scheme obtained by the NSGA-II non-dominated sorting genetic algorithm in this invention;

[0057] Figure 8 This is a graph showing the change in crankshaft arm distance difference during three adjustments to the center frame position in this invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0060] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0061] like Figure 1 As shown, an adaptive adjustment method for the crankshaft clamping center support position is used in applications such as... Figure 2 On the crankshaft model shown, the deformation of the crankshaft during grinding clamping is analyzed, and the deformation of the crankshaft is controlled within the machining range by adjusting the center rest. This includes the following steps:

[0062] S1. Establish a crankshaft clamping finite element model: Based on the crankshaft size and structure and the center frame structure, establish a crankshaft clamping finite element model, analyze the crankshaft deformation during the clamping and adjustment process and the center frame support position adjustment process, and obtain crankshaft deformation data, that is, the deformation amount at each crank arm of the crankshaft.

[0063] S101. Analyze the structural features of the crankshaft, establish a finite element model of the crankshaft clamping based on the crankshaft size structure and center frame structure, and simplify the center frame structure into a V-block structure for easy simulation analysis; in this embodiment, the small structural features of the crankshaft, such as chamfers, fillets, small holes and oil holes, are removed, thereby improving the efficiency of the simulation and the accuracy of the simulation results.

[0064] S102. Set the material parameters of the crankshaft clamping finite element model: crankshaft density, Young's modulus, yield strength, Poisson's ratio;

[0065] Table 1.42 CrMo alloy steel mechanical and physical properties parameters

[0066]

[0067] S103. Mesh the crankshaft clamping finite element model:

[0068] Different mesh density strategies were adopted for different regions. Tetrahedral element meshes were used for both the crankshaft and the center frame, with the element type set to C3D10. Corresponding node sets were also set for the extraction of crankshaft deformation data.

[0069] The areas requiring grid density division include the surface area where the center frame contacts the crankshaft, the area where the center frame does not contact the crankshaft, and the area where the crankshaft does not contact the center frame;

[0070] The grid size of the surface area where the center support contacts the crankshaft is 0.5-1mm, the grid size of the area where the center support does not contact the crankshaft is 10-20mm, and the grid size of the area where the crankshaft does not contact the center support is 20-30mm.

[0071] After the mesh is generated, each region contains several elements, and each element contains several nodes. The midpoints of the lowest edges of all crank arms in the nodes are used to form a node set. By processing the displacement changes of the nodes in the node set, the deformation difference at each crank arm of the crankshaft is obtained.

[0072] S104. Set the boundary conditions of the crankshaft according to the actual clamping situation to obtain the initial crankshaft clamping finite element model.

[0073] In step S104, the boundary conditions of the crankshaft are set according to the actual clamping situation of the crankshaft, specifically as follows:

[0074] S1041. Apply a completely fixed constraint to the center hole at one end of the crankshaft, and apply a clamping force of 20KN along the axial direction at the other end.

[0075] S1042. Move the center support to the set position, apply a completely fixed constraint to the center support, and adjust the clamping force applied to the crankshaft to 7.5KN.

[0076] S1043. Constrain the movement and rotation in the X, Y, and Z directions at the center holes at both ends of the crankshaft, so that the crankshaft rotates one revolution;

[0077] Gravity was applied during the actual crankshaft clamping process described above, with a gravitational acceleration of 9.8 m / s². 2Boundary conditions are fixed constraints, clamping forces, crankshaft rotation, and movement.

[0078] S105. Based on the initial crankshaft clamping finite element model established in S104, parameterize the model and solve it to obtain crankshaft deformation data under different center frame support positions.

[0079] S1051. Determine the adjustment range of the center support position. Based on the settings of S101-S104, perform clamping simulation analysis on the crankshaft under unsupported conditions to obtain the crankshaft simulation deformation results.

[0080] S1052. Extract the deformation of the crankshaft at the required support position of the center frame from the simulation results, and use this as the lower limit H of the center frame support position. min The upper limit of the central frame support position is set to H. max =0;

[0081] A simulation analysis of crankshaft stress and deformation was conducted without central support. The simulation results were processed, and the deformation at 24 nodes along the crankshaft axis in the direction of gravity was extracted at equal intervals. For example... Figure 3 As shown, without the support of a center frame, the crankshaft deforms most at the middle under gravity, and the axis bends in a downward arc. The deformation H at the support positions of the four center frames is shown. min The values ​​are -0.314, -0.668, -0.672, and -0.328 mm, respectively, with the deformation approximately symmetrically distributed around the midpoint of the crankshaft.

[0082] S1053. Obtain different center support positions based on the actual crankshaft clamping conditions. Within the range of the center support positions, i.e., H... min ≤H≤H max Different combinations of center frame support positions within a specified range were selected. Multiple simulation .inp files for different center frame support positions were generated using a Python script. This completed the parameterization of the initial crankshaft clamping finite element model, and the crankshaft deformation data under different center frame support positions were obtained.

[0083] The adjustment amount δh for the center frame support position is selected to be less than 0.1H. max To ensure that the simulation results reflect the influence of different center frame support height deviations on the crankshaft arm distance difference, the Latin hypercube sampling (LHS) method was used to select different center frame support position combinations. This was achieved by dividing the range of each variable into mutually exclusive intervals with equal probability, and then randomly selecting samples from each interval. The formula is as follows:

[0084]

[0085] In the formula, i = (1,2,…n) represents the number of variables, i.e., the number of central frames; j = (1,2,…m) represents the sampling dimension, i.e., the j-th sampling; P i -1 Let α be the probability distribution function; ij is a random number within the range [0,1]; N is the number of samples, i.e., the number of different combinations of central frame support positions;

[0086] Finally, N different combinations of center frame support positions were selected to complete the parameterization of the initial crankshaft clamping finite element model.

[0087] S2. Obtain the crankshaft simulation dataset: Process the deformation data in S1 using a Python script to obtain the arm distance difference data for each crankshaft crank and the corresponding center frame support position data, and establish the simulation dataset:

[0088] Using ABAQUS secondary development, a Python post-processing script was written to extract crankshaft deformation data from the ODB result file obtained from the S1 simulation. The axial displacement U3 of each node in the S1 node set was extracted. The arm distance difference of each crankshaft crank was obtained by subtracting the nodes on the corresponding two crank arms on the same crank. The arm distance difference data of each crank and the corresponding center frame adjustment position data of different actual crankshaft clamping conditions were organized into a simulation dataset.

[0089] In this embodiment, the crankshaft simulation deformation results under three different center frame support schemes were selected for processing, such as... Figure 4 As shown in Figures (a)-(i), the crankshaft arm distance changes during one revolution of cranks 1 to 9 under three different support schemes. The crankshaft arm distance changes most significantly when there is no center frame support. The crankshaft arm distance changes less when the four center frame support positions deviate from the crankshaft axis by 0.25, 0.5, 0.5, and 0.25 mm respectively. The crankshaft arm distance changes least when the deviations are 0.1, 0.2, 0.2, and 0.1 mm, indicating a positive correlation between crankshaft axis deviation and the deformation of each crank.

[0090] Among them, cranks 1 and 5 are most affected by the axis deviation, with a change in arm distance of 0.12 mm when there is no center frame support; the arm distance changes of cranks 6 and 7 are close to 0.09 mm; the arm distance changes of cranks 3, 4, 8, and 9 are within 0.03-0.08 mm; and the arm distance change of crank 2 is the smallest, within 0.03 mm. As can be seen from the figure, the change in crank distance during one revolution is similar to a sine wave. Due to the angle between each crank, there is a certain phase angle deviation in the change of arm distance for different cranks.

[0091] The difference between the maximum and minimum crankshaft arm distance values ​​within one revolution is the arm distance difference. The crankshaft arm distance values ​​under 200 different center frame support schemes were processed to calculate the corresponding crankshaft arm distance difference data. This data was then compiled into a dataset for subsequent prediction of the crankshaft arm distance difference and adaptive adjustment of the center frame.

[0092] S3. Establish a surrogate model for the center frame support position-crankshaft arm distance difference: Establish an arm distance difference prediction model based on a BP neural network, set the hyperparameters of the arm distance difference prediction model, and train the arm distance difference prediction model using the simulation dataset in S2. Take the center frame support position as input and the crankshaft arm distance difference as output to obtain a surrogate model for predicting the center frame support position-crankshaft arm distance difference:

[0093] A backpropagation (BP) neural network model was established, with the four central frames and nine crankshaft cranks used as the input and output layer neurons, respectively.

[0094] The 200 sets of simulation data were divided into three parts: training, validation, and testing. The training samples comprised 70% of the simulation dataset, with 140 sets used for initial training and tuning of the network model. The validation samples comprised 20 sets (10%), used for model parameter optimization and hyperparameter tuning during training. The testing samples comprised 40 sets, used to evaluate the model's predictive performance. The maximum number of iterations for the BP neural network model was 1000, the learning rate was 0.001, and the minimum training error was set to 1e-6 to ensure convergence and stability during training.

[0095] Meanwhile, the structure of the BP neural network model is set. In this embodiment, the BP neural network model is set to have 4 layers. The first layer is the input layer, which is used to receive data on the support position of the central frame. The number of neurons in this layer is equal to the number of central frames. The second and third layers are hidden layers, and the number of neurons is set to be between 13 and 21. Within this range, the combination of the number of neurons with the smallest training error is found, such as 15-19. The fourth layer is the output layer, which is used to output the predicted crank arm distance difference data. The number of neurons in this layer is equal to the number of crankshaft cranks.

[0096] The training set is used to train the model, which is then used to predict crankshaft arm distance differences and provides a basis for solving the optimal position of the center frame. A test set is used, and mean squared error is selected as the main metric to measure the deviation between predicted and observed values. Figure 5The training results are shown, including error changes at different iteration stages and validation and testing performance. As the number of model iterations increases, the MSE corresponding to the validation curve continuously decreases. After training, the validation set MSE reaches its minimum value of 0.013213 when the number of training iterations reaches 29, at which point training is stopped. Regression analysis is used to evaluate the correlation between the BP neural network training results and the target value. The correlation coefficient of the BP neural network model on the training samples is 0.9963, on the validation samples it is 0.98371, on the test samples it is 0.98705, and the overall correlation coefficient is 0.99334, indicating that the training model performs very well. The results are shown in [Figure number missing]. Figure 6 .

[0097] The surrogate model for the difference between the center frame support position and the crankshaft arm distance after training is as follows:

[0098]

[0099] is the predicted crankshaft arm distance difference, net is the trained BP neural network model, and H is the center frame support position.

[0100] S4. Optimization and adjustment of the center frame support position: Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, the NSGA-II multi-objective genetic algorithm is used to establish a reverse optimization algorithm for the center frame support position, and the support position of the center frame is iteratively adjusted based on the measured crankshaft crank arm distance difference data using this reverse optimization algorithm.

[0101] S401. Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, this model is incorporated into the NSGA-Ⅱ optimization algorithm. A condition is added to ensure that the sum of the actual support position of the current center frame and the distance to the crankshaft rotation axis is minimized to prevent over-adjustment of the center frame. This completes the establishment of the inverse solution optimization algorithm for the center frame support position.

[0102] First, we establish a mathematical problem for optimizing the center frame support position, minimizing the absolute error between the measured crankshaft arm distance difference and the predicted crankshaft arm distance difference, while simultaneously minimizing the sum of the absolute values ​​of the center frame positions corresponding to the predicted arm distance difference.

[0103] Next, constraints are set, namely, the upper and lower limits of the adjustment range of the central frame position are set, thus completing the establishment of the central frame support position optimization problem;

[0104] H min and H max These represent the lower and upper limits for center frame position adjustment, respectively, and D is the actual measured crankshaft arm distance difference. Let f1(D) be the predicted crankshaft arm distance difference, H be the center frame support position, f2(H) be the predicted crankshaft arm distance difference error, and f3(D) be the center frame support position.

[0105] Finally, the relevant parameters of the NSGA-II multi-objective genetic algorithm were set. The model parameters included a population size of 200, a maximum number of iterations of 300, a crossover probability of 0.9, and a mutation probability of 0.05. The optimization problem of the central support position was then incorporated into the NSGA-II multi-objective genetic algorithm to complete the establishment of the reverse optimization algorithm for the central support position.

[0106] The reverse optimization algorithm for the support position of the central frame was verified through simulation analysis:

[0107] First, a random combination of center frame support positions is initialized, and the crankshaft arm distance difference under the current center frame support position scheme is obtained through simulation analysis. Next, the obtained crankshaft arm distance difference is input into the optimization algorithm, and the algorithm is used to solve for the support position of the center frame under the current deformation of the crankshaft. Finally, a scheme is found in the Pareto solution set to adjust the center frame position. This scheme must meet the following conditions: (1) the error between the predicted crankshaft arm distance difference and the actual measured crankshaft arm distance difference under this set of center frame support schemes is minimized; (2) the absolute value of the sum of the distances from this set of center frame support positions to the crankshaft axis is minimized, that is, the adjustment increment required for the center frame position is minimized.

[0108] S402. The actual crankshaft arm distance difference is measured to determine whether the actual crankshaft arm distance difference meets the processing conditions. If it does not meet the conditions, the actual crankshaft arm distance difference is input into the reverse optimization algorithm for the center frame support position to obtain the current actual support position of the center frame. The current actual support position of the center frame is adjusted. After adjustment, the arm distance difference data of each crankshaft is measured again. If the processing conditions are not met, the data is input into the reverse optimization algorithm for iterative adjustment until the actual crankshaft arm distance difference meets the actual processing conditions.

[0109] The above algorithm was used to measure and adjust the deformation under multiple given center frame support schemes. After three adjustments, the crankshaft axis deviation was brought within 0.01mm. Figure 7 As shown, the Pareto solution set after three adjustments was obtained using the NSGA-II algorithm. As the number of adjustments increases, the Pareto solution set gradually decreases and tends to 0, reflecting that the deviation of the support position of the center frame from the crankshaft axis becomes smaller and smaller during the continuous adjustment process.

[0110] like Figure 8As shown, the adjustment process of one set of center support positions is demonstrated. After selecting the optimal adjustment scheme, the position of the center support is adjusted. Within three iterations, the arm distance difference of the nine crankshaft cranks is reduced to within 0.02mm, and remains stable in subsequent adjustments. The specific changes in the arm distance difference of each crank are shown in Table 2. After the first adjustment, only the arm distance difference of crank 8 does not meet the subsequent processing conditions. After the second adjustment, the arm distance differences of all cranks meet the requirements. Furthermore, the third adjustment shows that the crankshaft arm distance difference tends to stabilize, with a maximum change of 0.4µm, demonstrating the feasibility of the proposed method.

[0111] Table 2. Crankshaft arm distance difference during iterative adjustment

[0112]

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0114] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An adaptive adjustment method for the crankshaft clamping center support position, comprising the following steps: S1. Establish a crankshaft clamping finite element model: Based on the crankshaft size and structure and the center frame structure, establish a crankshaft clamping finite element model, analyze the crankshaft deformation during the clamping and adjustment process and the center frame support position adjustment process, and obtain crankshaft deformation data, that is, the deformation amount at each crank arm of the crankshaft. S2. Obtain the crankshaft simulation dataset: Process the deformation data in S1 to obtain the arm distance difference data of each crankshaft crank and the corresponding center frame support position data, and establish the simulation dataset. S3. Establish a surrogate model for the difference between the center frame support position and the crankshaft arm distance: Establish a prediction model for the arm distance difference based on a BP neural network, set the hyperparameters of the prediction model for the arm distance difference, and train the prediction model for the arm distance difference using the simulation dataset in S2. Take the center frame support position as the input and the crankshaft arm distance difference as the output to obtain a surrogate model for the prediction of the difference between the center frame support position and the crankshaft arm distance. S4. Optimization and adjustment of the center frame support position: Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, the NSGA-II multi-objective genetic algorithm is used to establish a reverse optimization algorithm for the center frame support position, and the support position of the center frame is iteratively adjusted based on the measured crankshaft crank arm distance difference data using this reverse optimization algorithm. The characteristic is that S4 includes: S401. Based on the surrogate model of the center frame support position-crankshaft arm distance difference trained in S3, it is brought into the NSGA-Ⅱ optimization algorithm, and the condition of ensuring that the sum of the actual support position of the current center frame and the crankshaft rotation axis is minimized is added to complete the establishment of the reverse solution optimization algorithm for the center frame support position. Specifically, S401 is: First, we establish a mathematical problem for optimizing the center frame support position, minimizing the absolute error between the measured crankshaft arm distance difference and the predicted crankshaft arm distance difference, while simultaneously minimizing the sum of the absolute values ​​of the center frame positions corresponding to the predicted arm distance difference. Next, constraints are set, namely, the upper and lower limits of the adjustment range of the central frame position are set, thus completing the establishment of the central frame support position optimization problem; , and These represent the lower and upper limits for center frame position adjustment, respectively, and D is the actual measured crankshaft arm distance difference. The predicted crankshaft arm distance difference is given by H, where H represents the center support position. The sum of the predicted crankshaft arm distance difference error, For the central frame support position and; Finally, the relevant parameters of the NSGA-II multi-objective genetic algorithm were set, including population size, maximum number of iterations, crossover probability and mutation probability. The optimization problem of the central frame support position was then incorporated into the NSGA-II multi-objective genetic algorithm to complete the establishment of the reverse optimization algorithm for the central frame support position. S402. The actual crankshaft arm distance difference is measured to determine whether the actual crankshaft arm distance difference meets the processing conditions. If not, the actual crankshaft arm distance difference is input into the reverse optimization algorithm for the center frame support position to obtain the current actual support position of the center frame. The current actual support position of the center frame is adjusted. After adjustment, the arm distance difference data of each crankshaft is measured again. If the processing conditions are not met, the data is input into the reverse optimization algorithm for iterative adjustment until the actual crankshaft arm distance difference meets the actual processing conditions.

2. The adaptive adjustment method for the crankshaft clamping center support position according to claim 1, characterized in that: S1 includes: S101. Analyze the structural features of the crankshaft, establish a finite element model of the crankshaft clamping based on the crankshaft size structure and center frame structure, and simplify the center frame structure into a V-block structure for easy simulation analysis. S102. Set the material parameters of the crankshaft clamping finite element model: crankshaft density, Young's modulus, Poisson's ratio, and yield strength; S103. Mesh the crankshaft clamping finite element model; S104. Set the boundary conditions of the crankshaft according to the actual clamping situation to obtain the initial crankshaft clamping finite element model; S105. Based on the initial crankshaft clamping finite element model established in S104, parameterize the model and solve it to obtain crankshaft deformation data under different center frame support positions.

3. The adaptive adjustment method for the crankshaft clamping center support position according to claim 2, characterized in that: Specifically, S103 is: Different mesh density strategies were adopted for different regions. Tetrahedral element meshes were used for both the crankshaft and the center frame, and corresponding node sets were set for the extraction of crankshaft deformation data. The areas requiring grid density division include the surface area where the center frame contacts the crankshaft, the area where the center frame does not contact the crankshaft, and the area where the crankshaft does not contact the center frame; The grid size of the surface area where the center support contacts the crankshaft is 0.5-1mm, the grid size of the area where the center support does not contact the crankshaft is 10-20mm, and the grid size of the area where the crankshaft does not contact the center support is 20-30mm. After the mesh is generated, each region contains several elements, and each element contains several nodes. The midpoints of the lowest edges of all crank arms in the nodes are used to form a node set. By processing the displacement changes of the nodes in the node set, the deformation difference at each crank arm of the crankshaft is obtained.

4. The adaptive adjustment method for the crankshaft clamping center support position according to claim 2, characterized in that: In step S104, the boundary conditions of the crankshaft are set according to the actual clamping situation of the crankshaft, specifically as follows: S1041. Apply a complete fixing constraint to the center hole at one end of the crankshaft, and apply a clamping force along the axial direction at the other end; S1042. Move the center support to the set position, apply a completely fixed constraint to the center support, and adjust the clamping force applied to the crankshaft. S1043. Constrains the movement and rotation of the center holes at both ends of the crankshaft, causing the crankshaft to rotate one revolution; Gravity conditions were applied during the actual clamping process of the crankshafts, with a gravitational acceleration of 9.8 m / s².

5. The adaptive adjustment method for the support position of the crankshaft clamping center frame according to claim 2, characterized in that: S105 includes: S1051. Determine the adjustment range of the center support position. Based on the settings of S101-S104, perform clamping simulation analysis on the crankshaft under unsupported conditions to obtain the crankshaft simulation deformation results. S1052. Extract the deformation of the crankshaft at the required support position of the center frame from the simulation results, and use this as the lower limit of the center frame support position. The upper limit of the central frame support position is set to ; S1053. Obtain different center support positions based on the actual crankshaft clamping conditions. Within the range of the center support positions, i.e. Different combinations of center frame support positions within the specified range were selected to parameterize the initial crankshaft clamping finite element model, and the crankshaft deformation data under different center frame support positions were obtained by solving the problem.

6. The adaptive adjustment method for the support position of the crankshaft clamping center frame according to claim 5, characterized in that: Specifically, S1053 involves selecting the adjustment amount of the center frame support position. <0.1 To ensure that the simulation results reflect the influence of different center frame support height deviations on the crankshaft arm distance difference, the Latin hypercube sampling (LHS) method was used to select different center frame support position combinations. This was achieved by dividing the range of each variable into mutually exclusive intervals with equal probability, and then randomly selecting samples from each interval. The formula is as follows: ； In the formula, i=(1,2,…n) is the number of variables, i.e. the number of central frames; j=(1,2,…m) is the sampling dimension, i.e. the j-th sampling; It is the probability distribution function; is a random number within the range [0,1]; N is the number of samples, i.e., the number of different combinations of central frame support positions; Finally, N different combinations of center frame support positions were selected to complete the parameterization of the initial crankshaft clamping finite element model.

7. The adaptive adjustment method for the crankshaft clamping center support position according to claim 3, characterized in that: Specifically, S2 is: The crankshaft deformation data obtained from the S1 simulation is extracted, and the axial displacement U3 of each node in the S1 node set is extracted. The arm distance difference of each crank of the crankshaft is obtained by processing the node data under different rotation angles. The arm distance difference data of each crank under different actual crankshaft clamping conditions and the corresponding center frame adjustment position data are organized into a simulation dataset.

8. The adaptive adjustment method for the support position of the crankshaft clamping center frame according to claim 1, characterized in that: Specifically, S3 is: A backpropagation (BP) neural network model was established, with the number of center supports and crankshaft cranks used as the input and output, respectively. Data from the simulation dataset was randomly selected in a 7:1:2 ratio as the training set, validation set, and test set for evaluating the center support position-crankshaft arm distance difference surrogate model. The hyperparameters of the BP neural network model were set, including the maximum number of iterations, learning rate, and minimum training error. The structure of the BP neural network model was also defined. The model was trained on the training set, and the trained model was used to predict the crankshaft arm distance difference, providing a basis for solving the optimal center support position. The trained surrogate model for the center support position-crankshaft arm distance difference is as follows: ； For the predicted crankshaft arm distance difference, Here is the trained BP neural network model, and H is the position of the central support frame.

Images