Multi-strategy improved tire model parameter identification method and system considering envelope constraint

Abstract

The present application relates to the technical field of tire mechanics, in particular to a multi-strategy improved tire model parameter identification method and system considering envelope constraints, a multi-strategy tire model parameter identification method considering envelope constraints is fused with tire data preprocessing, covers pure longitudinal working conditions, pure lateral working conditions, pure lateral return-to-normal working conditions, combined longitudinal working conditions and combined lateral working conditions, and relates to all working conditions of tire testing; the model parameter method is determined in a smaller parameter range, which can greatly reduce the search range and improve the convergence accuracy and speed; and the original AO algorithm is improved in a targeted multi-optimization strategy to improve the algorithm performance.

Description

Technical Field

[0001] This invention relates to the field of tire mechanics technology, and specifically to a multi-strategy improved tire model parameter identification method and system that considers envelope constraints. Background Technology

[0002] As the only component of a vehicle in direct contact with the ground, tires' dynamic characteristics have a crucial impact on vehicle safety, handling, comfort, and energy efficiency. However, during actual vehicle operation, tires undergo complex operating conditions such as acceleration, deceleration, steering, and camber, while the loads and air pressure they bear also change due to variations in vehicle motion, load, and environmental factors. Clearly, the force and torque data obtained from actual testing under tire operating conditions and test conditions cannot fully describe the tire's dynamic characteristics. Therefore, it is necessary to model the tire's mechanical properties, using model parameters to characterize the mechanical information under complex tire operating conditions and different test conditions, in order to dynamically guide vehicle tuning and design.

[0003] Although the Chinese invention patent application with publication number CN118114379A provides a method for identifying road six-component force parameters based on experimental data reconstruction, this method only discusses tire parameter identification under three specific working conditions: pure longitudinal, pure lateral, and combined lateral. It mainly involves cleaning and five-point weighted filtering of the original data, then generating a model from the processed data for parameter identification. This method relies heavily on prior six-component force data for training the original model, requiring high quality and representativeness of the prior data. If the prior data is biased or incomplete, it may affect the accuracy of the model. Furthermore, the neural network used in this method is a fully connected neural network with three hidden layers and 32 nodes in each layer. Fully connected neural networks are prone to overfitting when training data is limited, making them unsuitable for identifying tire data with relatively small amounts of data. Moreover, the performance of this network is highly sensitive to the selection of hyperparameters, requiring extensive experiments to find the optimal hyperparameter combination, without any specific improvements tailored to the characteristics of tire model parameters. Additionally, the working conditions covered by this method are not comprehensive, making it difficult to widely apply.

[0004] Therefore, most existing tire parameter identification methods use a single algorithm for parameter identification. Although they are easy to operate, they have few adjustable parameters, do not cover all working conditions, and lack targeted identification and optimization of tire mechanics, its models, and parameter characteristics. As a result, the identification accuracy is relatively low and cannot meet the needs of actual use. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-strategy improved tire model parameter identification method and system that considers envelope constraints, thereby solving the technical problems of limited adjustable parameters, incomplete working conditions, and low identification accuracy in current tire models.

[0006] The solution of the present invention to the above-mentioned technical problems is as follows:

[0007] A multi-strategy improved tire model parameter identification method considering envelope constraints, characterized by the following steps:

[0008] S1. Obtain tire data under different operating conditions;

[0009] S2. Construct tire models under different working conditions;

[0010] S3. Define the parameters to be identified for the tire model under different working conditions;

[0011] S4. Based on tire data under different working conditions, construct the envelope of the parameters to be identified under different working conditions;

[0012] S5. Based on the envelope of the parameters to be identified, determine the range of values ​​for the parameters to be identified in the tire model under different working conditions.

[0013] S6. Determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the value range after initialization.

[0014] Further defined, the operating conditions include pure longitudinal operating conditions, pure lateral operating conditions, and combined operating conditions;

[0015] The tire data includes:

[0016] Longitudinal slip ratio, longitudinal force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations;

[0017] The camber moment, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different combinations of camber angles and vertical loads;

[0018] Longitudinal slip ratio, longitudinal force, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations;

[0019] The longitudinal slip ratio and sideslip angle are the independent variables of the tire data, while the longitudinal force, lateral force, and self-aligning torque are the dependent variables of the tire data.

[0020] Further specifying, step S1 includes the following steps:

[0021] S1.1 Obtain the original data of the tire under pure longitudinal working conditions, pure lateral working conditions, and combined working conditions;

[0022] S1.2. The acquired longitudinal working condition raw data, pure lateral working condition raw data and combined working condition raw data are all preprocessed to obtain tire data under different working conditions.

[0023] Further specifying, step S1.2 specifically includes:

[0024] Determine the baseline distance between any two data points in the tire data;

[0025] Calculate the local distance coefficient of each data point based on the baseline distance between any two data points;

[0026] Calculate the standard distance index of each data point based on the local distance coefficient of each data point;

[0027] An anomaly threshold is set, and abnormal data points are removed by comparing the distance standard index of each data point with the anomaly threshold to obtain tire data.

[0028] Further specifying, step S2 includes the following steps:

[0029] Construct a tire longitudinal force formula model for pure longitudinal working conditions based on the PAC2002 tire model:

[0030]

[0031] in, For longitudinal force, For pure longitudinal slip peak factor, For pure longitudinal slip shape factor, B x For pure longitudinal slip stiffness factor, For pure longitudinal curvature factor, To correct the tire slip ratio, S Vx This represents the vertical drift value under pure lateral deviation conditions.

[0032] A tire lateral force formula model for pure lateral conditions is constructed based on the PAC2002 tire model:

[0033]

[0034] in, It is a lateral force; For pure lateral skewness peak factor, For pure lateral shape factor, For pure lateral stiffness factor, For pure lateral curvature factor, To correct the sideslip angle, This represents the vertical drift value under pure lateral deviation conditions.

[0035] A tire self-alignment torque formula model for pure lateral conditions is constructed based on the PAC2002 tire model:

[0036]

[0037]

[0038]

[0039] in, To correct the torque, For tire trail, It is a lateral force. This is the residual torque; The peak factor of the pure lateral residual moment. To characterize the rectangular factor of pure lateral deviation residual force, The stiffness factor is the residual moment of pure lateral deflection. To correct the residual moment sideslip angle, Side slip angle; As the positive peak factor, For the positive shape factor, To correct the lateral stiffness factor, The normal curvature factor, To correct tire skid angle;

[0040] Construct a tire longitudinal force formula model under combined working conditions based on the PAC2002 tire model:

[0041]

[0042] in, For combined operating conditions , For combined operating conditions, the longitudinal weighted peak factor is... For combined working conditions, longitudinal shape factor, For the combined working condition longitudinal stiffness factor, For combined working conditions, the longitudinal curvature factor, Correct the sideslip angle for combined operating conditions;

[0043] Construct a tire lateral force formula model under combined working conditions based on the PAC2002 tire model:

[0044]

[0045] in, For combined operating conditions , The lateral peak factor for combined operating conditions. For the combined operating conditions, the lateral shape factor, For the combined working condition lateral stiffness factor, For the combined operating conditions, the lateral curvature factor, This represents the lateral and vertical drift value under combined operating conditions. To correct the combined longitudinal slip ratio.

[0046] Further specifying, step S3 includes the following steps:

[0047] Definition of parameters to be identified in the tire longitudinal force formula model under pure longitudinal working conditions:

[0048] , For longitudinal friction, For tire load, It is a scaling factor;

[0049] , Outward tilt angle, The coefficient of friction for longitudinal force is... The coefficient of friction varies with tire load. The coefficient of friction varies with the outward tilt angle Changes, To standardize the vertical load increment, This is the proportionality coefficient. Tire pressure, Longitudinal friction force With changes in tire pressure, Longitudinal friction force With the change of the square of tire pressure, , The nominal camber angle is for pure longitudinal slip.

[0050] , For longitudinal force shape factor;

[0051] , , The longitudinal stiffness index is... This is the longitudinal stiffness coefficient. The variation of longitudinal slip stiffness coefficient with tire load, The longitudinal stiffness index varies with tire load. For the change of slip stiffness with tire pressure, This represents the variation of slip stiffness with the square of tire pressure.

[0052] , The longitudinal force curvature coefficient, The curvature coefficient varies with tire load. The curvature coefficient varies with the square of the tire load. The curvature coefficient during driving; To correct the longitudinal slip ratio, , For longitudinal slip ratio, This represents the horizontal drift value under pure longitudinal sliding conditions. , This is the horizontal offset. This represents the variation of horizontal offset with tire load.

[0053] , For vertical displacement, This represents the change in vertical displacement with tire load.

[0054] The , , , , , , , , , , , , , , , , , and All are parameters to be identified;

[0055] Definition of parameters to be identified in the tire lateral force formula model under pure lateral conditions:

[0056] , It is lateral friction force;

[0057] , The coefficient of lateral friction is... The variation of the lateral friction coefficient with tire load; The coefficient of lateral friction varies with the outward tilt angle. Changes; The nominal camber angle is for pure sideslip. ;

[0058] , The shape factor for lateral forces;

[0059] , , For lateral stiffness, Lateral stiffness varies with outward tilt angle Changes, The lateral stiffness index;

[0060] , This represents the maximum value of the lateral stiffness coefficient. The load at which the lateral stiffness coefficient reaches its maximum value. For nominal load, For maximum stiffness With changes in tire pressure, To the maximum Load changes with tire pressure;

[0061] , The lateral force curvature coefficient, The curvature coefficient of lateral force varies with tire load. The curvature coefficient of the lateral force is subordinate to the tilt angle. The lateral force curvature coefficient varies with the outward tilt angle. Changes; To correct the sideslip angle, , Side slip angle, This represents the horizontal drift value under pure sideslip conditions.

[0062] , This is the horizontal offset. This represents the variation of lateral offset with tire load. Horizontal offset with outward tilt angle Changes;

[0063] , This is the vertical offset. This represents the variation of vertical offset with tire load. Vertical offset as a function of outward tilt angle Changes, The vertical offset varies with tire load and camber angle;

[0064] The , , , , , , , , , , , , , , , , and All are parameters to be identified;

[0065] Definition of parameters to be identified in the tire self-centering torque formula model under pure lateral conditions:

[0066] , This represents the first horizontal drift value under pure sideslip correction conditions. ;

[0067] , This is the second horizontal drift value under pure sideslip correction conditions;

[0068] , This represents the horizontal displacement of the tailline. displacement With changes in tire load, displacement With outward tilt angle Changes, displacement With changes in tire load and camber angle;

[0069] , The slope coefficient is... This is the coefficient of slope variation with tire load. This is the coefficient of the square of the slope as a function of tire load. The slope varies with the outward angle The coefficient of change, This is the coefficient of slope variation with absolute inclination angle; The nominal camber angle is for pure sideslip correction. ;

[0070] , The shape factor for lateral forces;

[0071] , The radius of the unloaded tire. For nominal load, Peak coefficient, The coefficient representing the variation of peak value with load. This is the coefficient of variation of peak value with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle The coefficient of change;

[0072]

[0073] The lateral force curvature coefficient, This is the coefficient of lateral force curvature as a function of tire load. This is the coefficient of lateral force curvature that varies with the square of the tire load. Curvature as a function of tire trail Changes, Curvature as a function of tilt angle and tire trail Changes;

[0074] , The slope factor of the residual moment varies with the absolute tilt angle. The aerodynamic trajectory shape of the residual torque as a function of the slope factor;

[0075] ;

[0076] , This is the peak residual moment coefficient. The peak value varies with the load. The peak value varies with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle and changes in load;

[0077] The , , , , , , , , , , , , , , , , , , , , , , , , , and All are parameters to be identified;

[0078] Definition of parameters to be identified in the tire longitudinal force formula model under combined working conditions:

[0079] , This represents the longitudinal horizontal drift value under combined operating conditions.

[0080] , For combined operating conditions Reduced displacement factor;

[0081] , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the variation with longitudinal slip ratio;

[0082] , For combined operating conditions Reduced shape factor;

[0083]

[0084] , For combined operating conditions curvature factor, For combined operating conditions Curvature factor as a function of tire load;

[0085] The , , , , and All are parameters to be identified;

[0086] Definition of parameters to be identified in the tire lateral force formula model under combined working conditions:

[0087] , This represents the lateral horizontal drift value under combined operating conditions.

[0088] , For combined operating conditions Reduced curvature factor For combined operating conditions The reduced curvature factor changes with tire load;

[0089] , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the effect of sideslip angle Changes, For combined operating conditions Decrease the sideslip angle The displacement;

[0090] , For combined operating conditions Reduced shape factor;

[0091] , The peak factor is the lateral peak factor for the combined operating conditions. for With the side deflection angle Changes, for With the change of longitudinal slip ratio κ, It is lateral friction force;

[0092] , , for As tire load changes, for With outward tilt angle Changes, for With the side deflection angle Changes;

[0093]

[0094] , This represents the lateral horizontal drift value under combined operating conditions.

[0095] , For combined operating conditions curvature factor, For combined operating conditions The curvature coefficient in combination with the load;

[0096] The , , , , , , , , , , , , and All of these are parameters to be identified.

[0097] Further specifying, step S4 includes the following steps:

[0098] S4.1 Sort the independent variables in the tire data, and adjust the sorting of the corresponding dependent variables according to the sorting of the independent variables to obtain the sorted dependent variables;

[0099] S4.2 Divide the sorted independent variables into multiple groups, such that the number of tire data in each group is W, where W is a positive integer greater than 1.

[0100] S4.3. Group the sorted dependent variables according to the grouped independent variables;

[0101] S4.4 Obtain the maximum value yup and minimum value ydown in each dependent variable group, and determine the corresponding independent variables xup and xdown based on the maximum value yup and minimum value ydown; thus obtaining the basic coordinate points of the upper envelope and the lower envelope.

[0102] S4.5. Fit the basic coordinate points of the upper envelope and the lower envelope respectively using polynomials to obtain the upper envelope line and the lower envelope line.

[0103] Further specifying, step S5 includes the following steps:

[0104] S5.1 Randomly assign stiffness factor, shape factor, peak factor, curvature factor and vertical drift value under different working conditions to obtain the model to be fitted and the tire model under different working conditions;

[0105] S5.2 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the envelope. If yes, obtain the upper limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step.

[0106] S5.3 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the lower envelope. If yes, obtain the lower limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step.

[0107] S5.4 Determine the range of values ​​for the tire model parameters to be identified under different working conditions based on the adjustment of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value.

[0108] Further specifying, the improved artemisinin optimization algorithm includes the following steps:

[0109] In the initialization phase, the range of values ​​of the tire model parameters to be identified under different working conditions is set as the generation condition for parameter initialization.

[0110] In the complete elimination phase, a non-linearly decreasing inertial weight is added to the update mechanism:

[0111]

[0112]

[0113]

[0114] in, For the first j Vidi i The search agent before the iteration, For the first j Vidi i The search agent after one iteration The natural base, c The decay index, which indicates the concentration decay of a drug in the human body, This represents the current iteration number. The maximum number of iterations is given, and best is the current optimal value.

[0115] During the local elimination phase, a random value mutation operator is added to perturb the globally optimal population position:

[0116]

[0117]

[0118]

[0119]

[0120] in, This is the i-th search agent before the iteration in the current dimension. , and All of these are the optimal solutions from the previous iteration; Find the upper boundary of the space. To find the lower boundary of the space; For random value mutation operators; The normalized fitness value, and All are random numbers in the range [0, 1].

[0121] The current solution is obtained by constructing the solution sequence for this iteration:

[0122]

[0123] Based on the current solution, generate a reverse solution and compare it with the current solution. Keep the solution with the better fitness value and form a new population. Then, perform the next iteration to solve the objective function.

[0124] The objective function is:

[0125]

[0126] in, For tire data, denoted as the predicted value for tire data, and n represents the number of tire data points.

[0127] A multi-strategy improved tire model parameter identification system considering envelope constraints, used to perform the above-described multi-strategy improved tire model parameter identification method considering envelope constraints, including:

[0128] The tire data acquisition unit is used to acquire tire data under different working conditions;

[0129] Tire model building unit, used to build tire models under different working conditions;

[0130] The parameter definition unit is used to define the parameters to be identified for the tire model under different working conditions.

[0131] The envelope construction unit is used to construct the envelope of the parameters to be identified under different working conditions based on tire data under different working conditions.

[0132] The parameter value range determination unit is used to determine the value range of the tire model under different working conditions based on the envelope of the parameter to be identified.

[0133] The optimal value determination unit for the parameters to be identified is used to determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the initialized value range.

[0134] The beneficial effects of this invention are as follows:

[0135] This invention integrates a multi-strategy tire model parameter identification method that considers envelope constraints in tire data preprocessing, covering pure longitudinal conditions, pure lateral conditions, pure lateral return conditions, combined longitudinal conditions, and combined lateral conditions, encompassing all tire testing conditions. By determining the model parameters within a smaller parameter range, the search range can be significantly reduced, improving convergence accuracy and speed. Furthermore, targeted multi-optimization strategies are implemented to improve the existing AO algorithm, thereby enhancing its performance. Attached Figure Description

[0136] Figure 1 This diagram illustrates the steps of the multi-strategy improved tire model parameter identification method considering envelope constraints according to the present invention. Detailed Implementation

[0137] 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0138] Example 1

[0139] refer to Figure 1 This invention provides a multi-strategy improved tire model parameter identification method considering envelope constraints, comprising the following steps:

[0140] S1. Obtain tire data under different operating conditions;

[0141] S2. Construct tire models under different working conditions;

[0142] S3. Define the parameters to be identified for the tire model under different working conditions;

[0143] S4. Based on tire data under different working conditions, construct the envelope of the parameters to be identified under different working conditions;

[0144] S5. Based on the envelope of the parameters to be identified, determine the range of values ​​for the parameters to be identified in the tire model under different working conditions.

[0145] S6. Determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the value range after initialization.

[0146] Further details regarding tire data include:

[0147] Longitudinal slip ratio, longitudinal force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations;

[0148] The camber moment, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different combinations of camber angles and vertical loads;

[0149] Longitudinal slip ratio, longitudinal force, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations.

[0150] The longitudinal slip ratio and sideslip angle are the independent variables of the tire data, while the longitudinal force, lateral force, and self-aligning torque are the dependent variables.

[0151] The acquired tire data needs to be preprocessed to remove outliers. This includes the following steps:

[0152] The range of tire data is precisely determined by calculating the distance standard index for each tire data point.

[0153] Determine the nearest points to each data point in the tire data;

[0154] First, assuming the tire data includes n data points, then the tire dataset D = { l 1 、l 2 、l 3 、......、l n},in For the tire dataset, the first The feature vector of each data point.

[0155] Secondly, for any two data points in the tire dataset and Calculations yielded and Euclidean distance between two data points d（ , ):

[0156]

[0157] in, m For feature dimension, For data points The k 1 eigenvalue, For data points The k Each feature value.

[0158] Let Z be a positive integer, representing the data points in the tire dataset. l Z-nearest neighbor, denoted as That is, the tire data center, the distance from the data point l The set of the Z nearest neighbors.

[0159] Finally, for any two data points in the tire dataset and Calculations yielded and Baseline distance between two data points :

[0160]

[0161] in, To correlate each data point in the tire dataset with the data points in the dataset In Euclidean distance sorting, the first nearest neighbor distance d（ , )for and The Euclidean distance between two data points.

[0162] Calculate the local distance coefficient for each data point based on the baseline distance between any two data points.

[0163] For each data point in the tire dataset, calculate the local distance coefficient for each data point. :

[0164]

[0165] Wherein, the local distance coefficient is the data point l to the set The reciprocal of the average reachable distance of each neighboring point.

[0166] Calculate the standard distance index for each data point based on the local distance coefficient of each data point.

[0167] For each data point in the tire dataset, calculate the standard distance index for each data point. :

[0168]

[0169] Among them, the distance standard index is the data point and set The average local distance coefficient of each neighboring point in the data point The ratio of its own local distance coefficient.

[0170] Set an anomaly threshold and compare the distance standard index of each data point with the anomaly threshold to accurately obtain tire model data.

[0171] By distance from the standard index The point was measured The degree of difference in distance from its neighboring points, if A value close to 1 indicates that the data point The local distance coefficient is close to that of its neighboring points; if A value much greater than 1 indicates that the data point l The local distance coefficient of a point is lower than that of its neighboring points, which may indicate an outlier.

[0172] It is preferable to set a threshold, by comparing with Outliers are identified by comparing values; points that deviate from the standard index beyond a threshold are judged as outliers, thus refining the range of tire data.

[0173] Further specifying, step S2 includes the following steps:

[0174] Construct a tire longitudinal force formula model for pure longitudinal working conditions based on the PAC2002 tire model:

[0175] The PAC2002 (Pacejka2002) tire model is an optimized and improved version of the Pacejka89 and Pacejka96 tire models, providing a more accurate tire model description. Therefore, the PAC2002 tire model is selected.

[0176]

[0177] in, Represents the longitudinal force, lateral force, or self-aligning torque of the tire. As the peak factor, For shape factor, Stiffness factor For curvature factor, This refers to the tire slip ratio or sideslip angle.

[0178] From the PAC2002 tire model, the formula model for the longitudinal force of the tire under pure longitudinal working conditions can be obtained:

[0179]

[0180] in, For longitudinal force, For pure longitudinal slip peak factor, For pure longitudinal slip shape factor, B x For pure longitudinal slip stiffness factor, For pure longitudinal curvature factor, To correct the tire slip ratio, S Vx This represents the vertical drift value under pure lateral deviation conditions.

[0181] A tire lateral force formula model for pure lateral conditions is constructed based on the PAC2002 tire model:

[0182]

[0183] in, It is a lateral force; For pure lateral skewness peak factor, For pure lateral shape factor, For pure lateral stiffness factor, For pure lateral curvature factor, To correct the sideslip angle, This represents the vertical drift value under pure lateral deviation conditions.

[0184] A tire self-alignment torque formula model for pure lateral conditions is constructed based on the PAC2002 tire model:

[0185]

[0186]

[0187]

[0188] in, To correct the torque, For tire trail, It is a lateral force. This is the residual torque; The peak factor of the pure lateral residual moment. To characterize the rectangular factor of pure lateral deviation residual force, The stiffness factor is the residual moment of pure lateral deflection. To correct the residual moment sideslip angle, Side slip angle; As the positive peak factor, For the positive shape factor, To correct the lateral stiffness factor, The normal curvature factor, To correct tire skid angle;

[0189] Construct a tire longitudinal force formula model under combined working conditions based on the PAC2002 tire model:

[0190]

[0191] in, For combined operating conditions , For combined operating conditions, the longitudinal weighted peak factor is... For combined working conditions, longitudinal shape factor, For the combined working condition longitudinal stiffness factor, For combined working conditions, the longitudinal curvature factor, Correct the sideslip angle for combined operating conditions;

[0192] Construct a tire lateral force formula model under combined working conditions based on the PAC2002 tire model:

[0193]

[0194] in, For combined operating conditions , The lateral peak factor for combined operating conditions. For the combined operating conditions, the lateral shape factor, For the combined working condition lateral stiffness factor, For the combined operating conditions, the lateral curvature factor, This represents the lateral and vertical drift value under combined operating conditions. To correct the combined longitudinal slip ratio.

[0195] Further specifying, step S3 includes the following steps:

[0196] Definition of parameters to be identified in the tire longitudinal force formula model under pure longitudinal working conditions:

[0197] , For longitudinal friction, For tire load, This is the scaling factor, with a value of 1;

[0198] , Outward tilt angle, The coefficient of friction for longitudinal force is... The coefficient of friction varies with tire load. The coefficient of friction varies with the outward tilt angle Changes, To standardize the vertical load increment, This is the proportionality coefficient, with a value of 1; Tire pressure, Longitudinal friction force With changes in tire pressure, Longitudinal friction force With the change of the square of tire pressure, , The nominal camber angle is for pure longitudinal slip.

[0199] , For longitudinal force shape factor;

[0200] , , The longitudinal stiffness index is... This is the longitudinal stiffness coefficient. The variation of longitudinal slip stiffness coefficient with tire load, The longitudinal stiffness index varies with tire load. For the change of slip stiffness with tire pressure, This represents the variation of slip stiffness with the square of tire pressure.

[0201] , The longitudinal force curvature coefficient, The curvature coefficient varies with tire load. The curvature coefficient varies with the square of the tire load. The curvature coefficient during driving; To correct the longitudinal slip ratio, , For longitudinal slip ratio, This represents the horizontal drift value under pure longitudinal sliding conditions. , This is the horizontal offset. This represents the variation of horizontal offset with tire load.

[0202] , For vertical displacement, This represents the change in vertical displacement with tire load.

[0203] The , , , , , , , , , , , , , , , , , and All are parameters to be identified;

[0204] Definition of parameters to be identified in the tire lateral force formula model under pure lateral conditions:

[0205] , It is lateral friction force;

[0206] , The coefficient of lateral friction is... The variation of the lateral friction coefficient with tire load; The coefficient of lateral friction varies with the outward tilt angle. Changes; The nominal camber angle is for pure sideslip. ;

[0207] , The shape factor for lateral forces;

[0208] , , For lateral stiffness, Lateral stiffness varies with outward tilt angle Changes, The lateral stiffness index;

[0209] , This represents the maximum value of the lateral stiffness coefficient. The load at which the lateral stiffness coefficient reaches its maximum value. For nominal load, For maximum stiffness With changes in tire pressure, To the maximum The load changes with tire pressure.

[0210] , The lateral force curvature coefficient, The curvature coefficient of lateral force varies with tire load. The curvature coefficient of the lateral force is subordinate to the tilt angle. The lateral force curvature coefficient varies with the outward tilt angle. Changes; To correct the sideslip angle, , Side slip angle, This represents the horizontal drift value under pure sideslip conditions.

[0211] , This is the horizontal offset. This represents the variation of lateral offset with tire load. Horizontal offset with outward tilt angle The changes.

[0212] , This is the vertical offset. This represents the variation of vertical offset with tire load. Vertical offset as a function of outward tilt angle Changes, The vertical offset varies with tire load and camber angle;

[0213] The , , , , , , , , , , , , , , , , and All of these are parameters to be identified.

[0214] Definition of parameters to be identified in the tire self-centering torque formula model under pure lateral conditions:

[0215] , This represents the first horizontal drift value under pure sideslip correction conditions. ;

[0216] , This is the second horizontal drift value under pure sideslip correction conditions;

[0217] , This represents the horizontal displacement of the tailline. displacement With changes in tire load, displacement With outward tilt angle Changes, displacement With changes in tire load and camber angle;

[0218] , The slope coefficient is... This is the coefficient of slope variation with tire load. This is the coefficient of the square of the slope as a function of tire load. The slope varies with the outward angle The coefficient of change, This is the coefficient of slope variation with absolute inclination angle; The nominal camber angle is for pure sideslip correction. ;

[0219] , The shape factor for lateral forces;

[0220] , The radius of the unloaded tire. For nominal load, Peak coefficient, The coefficient representing the variation of peak value with load. This is the coefficient of variation of peak value with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle The coefficient of change;

[0221]

[0222]

[0223] The lateral force curvature coefficient, This is the coefficient of lateral force curvature as a function of tire load. This is the coefficient of lateral force curvature that varies with the square of the tire load. Curvature as a function of tire trail (Replace t with tire trail) changes, Curvature as a function of tilt angle and tire trail Changes;

[0224] , The slope factor of the residual moment varies with the absolute tilt angle. The aerodynamic trajectory shape of the residual torque as a function of the slope factor;

[0225] ;

[0226] , This is the peak residual moment coefficient. The peak value varies with the load. The peak value varies with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle and changes in load;

[0227] The , , , , , , , , , , , , , , , , , , , , , , , , , and All of these are parameters to be identified.

[0228] Definition of parameters to be identified in the tire longitudinal force formula model under combined working conditions:

[0229] , This represents the longitudinal horizontal drift value under combined operating conditions.

[0230] , For combined operating conditions Reduced displacement factor;

[0231] , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the variation with longitudinal slip ratio;

[0232] , For combined operating conditions Reduced shape factor;

[0233]

[0234] , For combined operating conditions curvature factor, For combined operating conditions Curvature factor as a function of tire load;

[0235] The , , , , and All of these are parameters to be identified.

[0236] Definition of parameters to be identified in the tire lateral force formula model under combined working conditions:

[0237] , This represents the lateral horizontal drift value under combined operating conditions.

[0238] , For combined operating conditions Reduced curvature factor For combined operating conditions The reduced curvature factor varies with tire load.

[0239] , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the effect of sideslip angle Changes, For combined operating conditions Decrease the sideslip angle The displacement.

[0240] , For combined operating conditions Reduced shape factor;

[0241] , The peak factor is the lateral peak factor for the combined operating conditions. for With the side deflection angle Changes, for With the change of longitudinal slip ratio κ, It is lateral friction force;

[0242] , , for As tire load changes, for With outward tilt angle Changes, for With the side deflection angle The changes.

[0243]

[0244] , This represents the lateral horizontal drift value under combined operating conditions.

[0245] , For combined operating conditions curvature factor, For combined operating conditions The curvature coefficient in combination with the load;

[0246] The , , , , , , , , , , , , and All of these are parameters to be identified.

[0247] Further specifying, step S4 includes the following steps:

[0248] S4.1 Sort the independent variables in the tire data, and adjust the sorting of the corresponding dependent variables according to the sorting of the independent variables to obtain the sorted dependent variables;

[0249] S4.2 Divide the sorted independent variables into multiple groups, such that the number of tire data in each group is W, where W is a positive integer greater than 1; preferably, W is 10.

[0250] S4.3. Group the sorted dependent variables according to the grouped independent variables;

[0251] S4.4 Obtain the maximum value yup and minimum value ydown in each dependent variable group, and determine the corresponding independent variables xup and xdown based on the maximum value yup and minimum value ydown; thus obtaining the basic coordinate points of the upper envelope and the lower envelope.

[0252] S4.5. Fit the basic coordinate points of the upper envelope and the lower envelope respectively using polynomials to obtain the upper envelope line and the lower envelope line.

[0253] Step S5 includes the following steps:

[0254] S5.1 Randomly assign stiffness factor, shape factor, peak factor, curvature factor and vertical drift value under different working conditions to obtain the model to be fitted and the tire model under different working conditions;

[0255] S5.2 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the envelope. If yes, obtain the upper limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step.

[0256] S5.3 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the lower envelope. If yes, obtain the lower limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step.

[0257] S5.4 Determine the range of values ​​for the tire model parameters to be identified under different working conditions based on the adjustment of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value.

[0258] Artemisinin Optimization (AO) is a metaheuristic algorithm inspired by the efficacy of artemisinin in treating malaria. It effectively addresses common optimization dilemmas, such as getting trapped in local optima. The AO algorithm consists of three optimization phases: a global elimination phase simulating global exploration, a local cleanup phase involving local development, and a post-consolidation phase enhancing the algorithm's ability to escape local optima.

[0259] The improved artemisinin optimization algorithm includes the following steps:

[0260] Initialization phase:

[0261] Patients receive artemisinin via oral administration or injection. The drug particles are conceptualized as search agents for the algorithm; the entire set of these agents constitutes the solution set of the algorithm. Initially, the entire population is initialized, denoted as A. The complete population consists of… It consists of several search agents, among which This represents a multidimensional component within the search agent. This abstraction reflects the breakdown and absorption of drugs within the body, distributing them throughout the body via the bloodstream.

[0262]

[0263] in, , R represents the boundary of the solution space, and R represents a set of random number sequences in the range (0, 1). The improved AO algorithm of this invention uses a random number sequence to generate the initial solution.

[0264] The initialization steps can be improved using the following strategies:

[0265] Based on the above steps, the approximate range of each calculation factor is initially determined by envelope calculation, and a series of nonlinear inequality constraints are added. Since the strong coupling between the parameters can cover all the parameters involved in the model, the initialization stage is changed to set the range of the corresponding parameters in the model obtained under the nonlinear inequality constraints as the generation condition for parameter initialization, thereby accurately determining the parameter range, reducing the number of iterations, and improving the convergence accuracy and convergence speed.

[0266] Complete elimination phase:

[0267] In the initial stages of malaria treatment, patients are given larger doses of the drug to quickly control the progression of the disease. Once absorbed, artemisinin is transported throughout the body via the bloodstream.

[0268] To better balance the global exploration capability and local exploitation capability of the improved AO algorithm during the exploration phase, and to improve convergence accuracy, we added a non-linearly decreasing inertial weight to the original position update mechanism. In the formula The maximum number of iterations is given. As can be seen from the above formula, the inertia weight value decreases as the number of iterations increases, and this decreases in a non-linear, initially rapid, then slowing manner. This aligns with the original AO algorithm's search characteristics of focusing on global exploration in the early stages and local exploration in the later stages, which is beneficial for improving the algorithm's convergence speed and accuracy.

[0269]

[0270] In this strategy, the search agent exhibits a large-scale decentralized characteristic, serving as a guide for exploring a complex solution space. Specifically, For the first j Vidi i The search agent before the iteration, For the first j Vidi i After several iterations, the best search agent is selected as the current optimal agent. Meanwhile, the diffusion of artemisinin-based drugs in the human body follows pharmacokinetic principles. This strategy takes into account the fact that drug concentration decreases over time. c represents the decay exponent of artemisinin drug concentration in the human body. The decay of artemisinin drug concentration can be described using a one-compartment model, as shown below:

[0271] ,

[0272]

[0273] variable Let K represent the drug concentration and K represent the rate constant. Find the differential equation. express In this model, the drug concentration at a given time point decays exponentially over time. Therefore, the decay exponent of artemisinin concentration in the human body can be calculated using the following formula:

[0274]

[0275] The natural base is used; in this strategy, the initial drug concentration is assumed to be 1, the drug decay rate is 4, and the time process in the model is simulated using the algorithm's evaluation process. and This represents the current iteration number and the maximum iteration number of the algorithm. Considering the differences in patient condition severity and physiological factors, leading to varying medication dosages and durations, patients may experience different durations of treatment at this stage. To account for this inherent variability, a probability coefficient is introduced. :

[0276]

[0277] In the formula, As a probability coefficient, combined with the algorithm's assessment progress, it simulates an objective scenario, where patients exhibit different responses and durations at this stage based on their individual circumstances.

[0278] The current solution is obtained by constructing the solution sequence for this iteration:

[0279]

[0280] The current solution is then optimized.

[0281] Local removal phase:

[0282] The goal is to eliminate any remaining malaria parasites in the body, preventing their reproduction and the recurrence of malaria symptoms. A random value mutation operator is added to the original strategy to perturb the globally optimal population position, expanding the search breadth and increasing population diversity. This helps the algorithm escape local optima and accelerates convergence.

[0283]

[0284]

[0285] ,

[0286] in, This is the i-th search agent before the iteration in the current dimension. , and All of these are the optimal solutions from the previous iteration. The normalized fitness value is transformed into a probability distribution, serving as the relative weight among individuals. This ensures that individuals with higher fitness have a higher probability. This helps to retain excellent individuals to some extent, while providing opportunities for poorly performing individuals, adjusting the algorithm's focus on different individuals. ω is a random value mutation operator, which can be randomly switched based on the random number rand[0,1] to diversify the mutation methods. and All are random numbers in the range [0, 1]. This represents the current iteration number. This represents the maximum number of iterations. This strategy simulates the process by which a small amount of artemisinin clears potential malaria parasites from the human body.

[0287] Post-consolidation phase:

[0288] Indifference to the severity of the disease and laxity during treatment are major factors leading to relapse. Even after the attack and maintenance phases, where most malaria parasites have been eradicated, a small percentage may gradually develop drug resistance and enter a dormant phase. This significantly reduces their biological activity, making them difficult for drugs to effectively kill. If treatment is discontinued, the malaria parasites, having passed through dormancy, may cause a relapse. This section represents the possibility of unforeseen circumstances during the post-consolidation phase and simulates this specific situation. The model for this strategy is represented as follows:

[0289]

[0290] In this equation, Indicates the first v Vidi w The column represents a subvector of the current optimal solution; it also represents malaria parasites that were not eliminated due to entering a dormant phase. This strategy enhances the search agent's ability to escape local optima.

[0291] For the solutions generated iteratively, a dynamic back-learning strategy is introduced to optimize them. Dynamic back-learning is a strategy derived from back-learning, and its purpose is to overcome the drawback that the back-learning solution is worse than the original solution. By generating new back-learning solutions based on the current solution and comparing them with the original solution, the set of solutions with better fitness values ​​is retained and formed into a new population for the next iteration to solve the objective function. This method can increase the diversity of the algorithm population and improve the convergence speed and accuracy of the algorithm.

[0292] The objective function is:

[0293]

[0294] in, For tire data, denoted as the predicted value for tire data, and n represents the number of tire data points.

[0295] Definition (Top-Level Backward Solution): The improved AO algorithm defines the self-extremum point of an individual in the current population as the top-level solution. , =1, 2, ... , =1, 2, ..., ， Its reverse solution It can be defined as:

[0296]

[0297]

[0298] In the formula, Let be the top-level reverse coefficient on (0, 1). The top-level reverse coefficient facilitates the algorithm to generate multiple different reverse prey individuals, forming a new reverse population, enhancing the diversity and quality of the population, and improving the algorithm's search capability.

[0299] ∈[ , ], and All of these are dynamic boundaries. Dynamic boundaries allow the population to preserve search experience, enabling it to find optimal solutions within a narrow search space and accelerating algorithm convergence.

[0300] During the dynamic boundary operations and iterations described above, an individual in the population may cross the search space. To address this, we employ a mirror reflection theory to handle outbound individuals. This method defines the upper and lower boundaries of the search space. and Considering two mirrors, and individual Y0 as the propagating beam within them, the magnitude of Y represents the light intensity. After multiple reflections within the mirror space, the beam will eventually be completely consumed at a point Y0 within the mirror due to energy loss in the intermediate medium. Y0 is the projection of the out-of-bounds individual Y into the search space. The formula for handling the mirror boundary is as follows:

[0301]

[0302] Here, ub and lb represent the upper and lower boundaries of the solution space, respectively. The method of handling boundary violations using the mirror theory ensures diversity in the regression of individuals that cross the boundary, allowing the algorithm to maintain population diversity and prevent these individuals from clustering at the boundary, thus addressing the problem of uneven distribution.

[0303] Example 2

[0304] This embodiment provides a multi-strategy improved tire model parameter identification system considering envelope constraints, used to implement the multi-strategy improved tire model parameter identification method considering envelope constraints described in Embodiment 1, including:

[0305] The tire data acquisition unit is used to acquire tire data under different working conditions;

[0306] Tire model building unit, used to build tire models under different working conditions;

[0307] The parameter definition unit is used to define the parameters to be identified for the tire model under different working conditions.

[0308] The envelope construction unit is used to construct the envelope of the parameters to be identified under different working conditions based on tire data under different working conditions.

[0309] The parameter value range determination unit is used to determine the value range of the tire model under different working conditions based on the envelope of the parameter to be identified.

[0310] The optimal value determination unit for the parameters to be identified is used to determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the initialized value range.

[0311] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A multi-strategy improved tire model parameter identification method considering envelope constraints, characterized in that, Includes the following steps: S1. Obtain tire data under different operating conditions; S2. Construct tire models under different working conditions; S3. Define the parameters to be identified for the tire model under different working conditions; S4. Based on tire data under different working conditions, construct the envelope of the parameters to be identified under different working conditions; S5. Based on the envelope of the parameters to be identified, determine the range of values ​​for the parameters to be identified in the tire model under different working conditions. S6. Determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the value range after initialization; Step S4 includes the following steps: S4.1 Sort the independent variables in the tire data, and adjust the sorting of the corresponding dependent variables according to the sorting of the independent variables to obtain the sorted dependent variables; S4.2 Divide the sorted independent variables into multiple groups, such that the number of tire data in each group is W, where W is a positive integer greater than 1. S4.

3. Group the sorted dependent variables according to the grouped independent variables; S4.4 Obtain the maximum value yup and minimum value ydown in each dependent variable group, and determine the corresponding independent variables xup and xdown based on the maximum value yup and minimum value ydown; thus obtaining the basic coordinate points of the upper envelope and the lower envelope. S4.

5. Fit the basic coordinate points of the upper envelope and the lower envelope respectively using polynomials to obtain the upper envelope line and the lower envelope line. Step S5 includes the following steps: S5.1 Randomly assign stiffness factor, shape factor, peak factor, curvature factor and vertical drift value under different working conditions to obtain the model to be fitted and the tire model under different working conditions; S5.2 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the envelope. If yes, obtain the upper limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step. S5.3 Input the independent and dependent variables from the tire data into the model to be fitted and the tire model for the corresponding working conditions, respectively. Determine whether the model to be fitted and the tire model fit the lower envelope. If yes, obtain the lower limits of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value of the tire model under different working conditions. If not, adjust the stiffness factor, shape factor, peak factor, curvature factor and vertical drift value, and re-execute the current step. S5.4 Determine the range of values ​​for the tire model parameters to be identified under different working conditions based on the adjustment of stiffness factor, shape factor, peak factor, curvature factor and vertical drift value.

2. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 1, characterized in that, The operating conditions include pure longitudinal operating conditions, pure lateral operating conditions, and combined operating conditions; The tire data includes: Longitudinal slip ratio, longitudinal force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations; The camber moment, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different combinations of camber angles and vertical loads; Longitudinal slip ratio, longitudinal force, sideslip angle, lateral force, tire pressure, vertical load, and camber angle under different camber angles and vertical load combinations; The longitudinal slip ratio and sideslip angle are the independent variables of the tire data, while the longitudinal force, lateral force, and self-aligning torque are the dependent variables of the tire data.

3. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 2, characterized in that, Step S1 includes the following steps: S1.1 Obtain the original data of the tire under pure longitudinal working conditions, pure lateral working conditions, and combined working conditions; S1.

2. The acquired longitudinal working condition raw data, pure lateral working condition raw data and combined working condition raw data are all preprocessed to obtain tire data under different working conditions.

4. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 3, characterized in that, Step S1.2 specifically includes: Determine the baseline distance between any two data points in the tire data; Calculate the local distance coefficient of each data point based on the baseline distance between any two data points; Calculate the standard distance index of each data point based on the local distance coefficient of each data point; An anomaly threshold is set, and abnormal data points are removed by comparing the distance standard index of each data point with the anomaly threshold to obtain tire data.

5. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 1, characterized in that, Step S2 includes the following steps: Construct a tire longitudinal force formula model for pure longitudinal working conditions based on the PAC2002 tire model: in, For longitudinal force, For pure longitudinal slip peak factor, For pure longitudinal slip shape factor, B x For pure longitudinal slip stiffness factor, For pure longitudinal curvature factor, To correct the tire slip ratio, S Vx This represents the vertical drift value under pure lateral deviation conditions. A tire lateral force formula model for pure lateral conditions is constructed based on the PAC2002 tire model: in, It is a lateral force; For pure lateral skewness peak factor, For pure lateral shape factor, For pure lateral stiffness factor, For pure lateral curvature factor, To correct the sideslip angle, This represents the vertical drift value under pure lateral deviation conditions. A tire self-alignment torque formula model for pure lateral conditions is constructed based on the PAC2002 tire model: in, To correct the torque, For tire trail, It is a lateral force. This is the residual torque; The peak factor of the pure lateral residual moment. To characterize the rectangular factor of pure lateral deviation residual force, The stiffness factor is the residual moment of pure lateral deflection. To correct the residual moment sideslip angle, Side slip angle; As the positive peak factor, For the positive shape factor, To correct the lateral stiffness factor, The normal curvature factor, To correct tire skid angle; Construct a tire longitudinal force formula model under combined working conditions based on the PAC2002 tire model: in, For combined operating conditions , For combined operating conditions, the longitudinal weighted peak factor is... For combined working conditions, longitudinal shape factor, For the combined working condition longitudinal stiffness factor, For combined working conditions, the longitudinal curvature factor, Correct the sideslip angle for combined operating conditions; Construct a tire lateral force formula model under combined working conditions based on the PAC2002 tire model: in, For combined operating conditions , The lateral peak factor for combined operating conditions. For the combined operating conditions, the lateral shape factor, For the combined working condition lateral stiffness factor, For the combined operating conditions, the lateral curvature factor, This represents the lateral and vertical drift value under combined operating conditions. To correct the combined longitudinal slip ratio.

6. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 5, characterized in that, Step S3 specifically involves: Definition of parameters to be identified in the tire longitudinal force formula model under pure longitudinal working conditions: , For longitudinal friction, For tire load, It is a scaling factor; , Outward tilt angle, The coefficient of friction for longitudinal force is... The coefficient of friction varies with tire load. The coefficient of friction varies with the outward tilt angle Changes, To standardize the vertical load increment, This is the proportionality coefficient. Tire pressure, Longitudinal friction force With changes in tire pressure, Longitudinal friction force With the change of the square of tire pressure, , The nominal camber angle is for pure longitudinal slip. , For longitudinal force shape factor; ； ； The longitudinal stiffness index is... This is the longitudinal stiffness coefficient. The variation of longitudinal slip stiffness coefficient with tire load, The longitudinal stiffness index varies with tire load. For the change of slip stiffness with tire pressure, This represents the variation of slip stiffness with the square of tire pressure. ； The longitudinal force curvature coefficient, The curvature coefficient varies with tire load. The curvature coefficient varies with the square of the tire load. The curvature coefficient during driving; To correct the longitudinal slip ratio, , For longitudinal slip ratio, This represents the horizontal drift value under pure longitudinal sliding conditions. , This is the horizontal offset. This represents the variation of horizontal offset with tire load. , For vertical displacement, This represents the change in vertical displacement with tire load. The , , , , , , , , , , , , , , , , , and All are parameters to be identified; Definition of parameters to be identified in the tire lateral force formula model under pure lateral conditions: , It is lateral friction force; , The coefficient of lateral friction is... The variation of the lateral friction coefficient with tire load; The coefficient of lateral friction varies with the outward tilt angle. Changes; The nominal camber angle is for pure sideslip. ; , The shape factor for lateral forces; , , For lateral stiffness, Lateral stiffness varies with outward tilt angle Changes, The lateral stiffness index; , This represents the maximum value of the lateral stiffness coefficient. The load at which the lateral stiffness coefficient reaches its maximum value. For nominal load, For maximum stiffness With changes in tire pressure, To the maximum Load changes with tire pressure; ； The lateral force curvature coefficient, The curvature coefficient of lateral force varies with tire load. The curvature coefficient of the lateral force is subordinate to the tilt angle. The lateral force curvature coefficient varies with the outward tilt angle. Changes; To correct the sideslip angle, , Side slip angle, This represents the horizontal drift value under pure sideslip conditions. , This is the horizontal offset. This represents the variation of lateral offset with tire load. Horizontal offset with outward tilt angle Changes; ； This is the vertical offset. This represents the variation of vertical offset with tire load. Vertical offset as a function of outward tilt angle Changes, The vertical offset varies with tire load and camber angle; The , , , , , , , , , , , , , , , , and All are parameters to be identified; Definition of parameters to be identified in the tire self-centering torque formula model under pure lateral conditions: , This represents the first horizontal drift value under pure sideslip correction conditions. ; , This is the second horizontal drift value under pure sideslip correction conditions; , This represents the horizontal displacement of the tailline. displacement With changes in tire load, displacement With outward tilt angle Changes, displacement With changes in tire load and camber angle; , The slope coefficient is... This is the coefficient of slope variation with tire load. This is the coefficient of the square of the slope as a function of tire load. The slope varies with the outward angle The coefficient of change, This is the coefficient of slope variation with absolute inclination angle; The nominal camber angle is for pure sideslip correction. ; , The shape factor for lateral forces; ； The radius of the unloaded tire. For nominal load, Peak coefficient, The coefficient representing the variation of peak value with load. This is the coefficient of variation of peak value with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle The coefficient of change; ； The lateral force curvature coefficient, This is the coefficient of lateral force curvature as a function of tire load. This is the coefficient of lateral force curvature that varies with the square of the tire load. Curvature as a function of tire trail Changes, Curvature as a function of tilt angle and tire trail Changes; , The slope factor of the residual moment varies with the absolute tilt angle. The aerodynamic trajectory shape of the residual torque as a function of the slope factor; ； ； This is the peak residual moment coefficient. The peak value varies with the load. The peak value varies with tire pressure. Peak value varies with outward tilt angle The coefficient of change, Peak value varies with outward tilt angle and changes in load; The , , , , , , , , , , , , , , , , , , , , , , , , , and All are parameters to be identified; Definition of parameters to be identified in the tire longitudinal force formula model under combined working conditions: , This represents the longitudinal horizontal drift value under combined operating conditions. , For combined operating conditions Reduced displacement factor; , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the variation with longitudinal slip ratio; , For combined operating conditions Reduced shape factor; ； , For combined operating conditions curvature factor, For combined operating conditions Curvature factor as a function of tire load; The , , , , and All are parameters to be identified; Definition of parameters to be identified in the tire lateral force formula model under combined working conditions: , This represents the lateral horizontal drift value under combined operating conditions. , For combined operating conditions Reduced curvature factor For combined operating conditions The reduced curvature factor changes with tire load; , For combined operating conditions Decreasing slope factor For combined operating conditions Reduce the effect of sideslip angle Changes, For combined operating conditions Decrease the sideslip angle The displacement; , For combined operating conditions Reduced shape factor; , The peak factor is the lateral peak factor for the combined operating conditions. for With the side deflection angle Changes, for With the change of longitudinal slip ratio κ, It is lateral friction force; , , for As tire load changes, for With outward tilt angle Changes, for With the side deflection angle Changes; , This represents the lateral horizontal drift value under combined operating conditions. , For combined operating conditions curvature factor, For combined operating conditions The curvature coefficient in combination with the load; The , , , , , , , , , , , , and All of these are parameters to be identified.

7. The multi-strategy improved tire model parameter identification method considering envelope constraints according to claim 6, characterized in that, The improved artemisinin optimization algorithm includes the following steps: In the initialization phase, the range of values ​​of the tire model parameters to be identified under different working conditions is set as the generation condition for parameter initialization. In the complete elimination phase, a non-linearly decreasing inertial weight is added to the update mechanism: in, For the first j Vidi i The search agent before the iteration, For the first j Vidi i The search agent after one iteration The natural base, c The decay index, which indicates the concentration decay of a drug in the human body, This represents the current iteration number. The maximum number of iterations is given, and best is the current optimal value. During the local elimination phase, a random value mutation operator is added to perturb the globally optimal population position: in, This is the i-th search agent before the iteration in the current dimension. , and All of these are the optimal solutions from the previous iteration; Find the upper boundary of the space. To find the lower boundary of the space; For random value mutation operators; The normalized fitness value, and All are random numbers in the range [0, 1]. The current solution is obtained by constructing the solution sequence for this iteration: Based on the current solution, generate a reverse solution and compare it with the current solution. Keep the solution with the better fitness value and form a new population. Then, perform the next iteration to solve the objective function. The objective function is: in, For tire data, denoted as the predicted value for tire data, and n represents the number of tire data points.

8. A multi-strategy improved tire model parameter identification system considering envelope constraints, characterized in that, The method for performing the multi-strategy improved tire model parameter identification method considering envelope constraints as described in any one of claims 1 to 7 includes: The tire data acquisition unit is used to acquire tire data under different working conditions; Tire model building unit, used to build tire models under different working conditions; The parameter definition unit is used to define the parameters to be identified for the tire model under different working conditions. The envelope construction unit is used to construct the envelope of the parameters to be identified under different working conditions based on tire data under different working conditions. The parameter value range determination unit is used to determine the value range of the tire model under different working conditions based on the envelope of the parameter to be identified. The optimal value determination unit for the parameters to be identified is used to determine the optimal values ​​of the tire model parameters to be identified under different working conditions based on the initialized value range.

Images