DESIGN METHOD FOR OUTER CONTOUR STRUCTURE OF TIRE FOR REDUCING TIRE WIND RESISTANCE

Abstract

A design method for an outer contour structure of a tire capable of reducing tire wind resistance, including the following steps: step 1, establishing an initial tire model; step 2, building an tire aerodynamic drag calculation model; step 3, designing a testing scheme using tire five contour parameters; step 4, building a function relationship between tire five outer contour parameters and an aerodynamic drag coefficient values, and verifying the accuracy of the function relationship; and step 5, obtaining tire five outer contour parameters when the aerodynamic drag coefficient value is smallest by using an optimization algorithm. The design method effectively avoids the blindness problem in the design process of the tire outer contour structure and can effectively reduce the design cycle number of the tire outer contour structure, thereby improving the improvement efficiency, meanwhile, it is benefit to reduce tire aerodynamic drag and improve vehicle economy and reduce harmful emissions.

Claims

1. A design method of a tire outer contour structure for reducing a tire aerodynamic drag, comprising the following steps: step 1: dividing, according to a selected tire model, outer contour parameters of a selected tire model into five outer contour parameters: an upper sidewall height, a running surface width, a shoulder transition radius, and two tread arc radiuses; drawing, according to initial values of the five outer contour parameters of the selected tire model, an initial tire three-dimensional model; step 2: building an aerodynamic drag calculation model for the initial tire three-dimensional model using the initial values of the five outer contour parameters of the selected tire model, and performing a numerical simulation analysis to obtain a tire aerodynamic drag coefficient value of the initial tire three-dimensional model; step 3: designing testing schemes, and taking the five outer contour parameters of the selected tire model as design variables, and setting value ranges of the design variables; taking the tire aerodynamic drag coefficient value of the initial tire three-dimensional model as a target response value; selecting a statistical sampling test design method to generate the testing schemes using the design variables, and conducting a tire aerodynamic drag simulation analysis of each of the testing schemes, and obtaining tire aerodynamic drag coefficient values of all the testing schemes; step 4: constructing, according to the testing schemes and the tire aerodynamic drag coefficient values in step 3, a functional relationship between the design variables of the five outer contour parameters of the selected tire model and the tire aerodynamic drag coefficient values of the testing schemes, and verifying an accuracy of the functional relationship; step 5: selecting an intelligent optimization method to optimize the functional relationship in step 4, so as to obtain optimized five outer contour parameters when the tire aerodynamic drag coefficient values are the smallest, and then determining an optimized tire outer contour structure using the optimized five outer contour parameters.

2. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein in the step 1, the five outer contour parameters of the selected tire model are mainly described for a half-tire model in a tire width direction.

3. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein in the step 2, a construction method for the tire aerodynamic drag coefficient value of the initial tire three-dimensional model is a computational fluid dynamics calculation method; building a three-dimensional virtual wind tunnel model including a tire and a road surface, and the three-dimensional virtual wind tunnel model is imported into a meshing software to generate a fluid domain grid and a boundary layer grid around the initial tire three-dimensional model, and then setting boundaries for a velocity inlet, a pressure outlet, and a tire wall movement mode, and the numerical simulation analysis of the tire aerodynamic drag coefficient value is obtained by using a computational fluid dynamics analysis software.

4. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein in the step 3, five outer contour parameters of the initial tire three-dimensional model are used as a reference basis, and value ranges of the five outer contour parameters of the initial tire three-dimensional model are that an upper sidewall height is 0.47 to 0.63 of a section height, a running surface width is 0.73 to 0.9 of a section width of a half tire model, a shoulder transition radius is 15 to 45 mm, and two tread arc radiuses are determined according to the initial tire three-dimensional model.

5. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein the statistical sampling test design method is a Latin hypercube test design method; through a statistical design and a data analysis of tire aerodynamic drag coefficients, key parameters of the five outer contour parameters having a significant influence on the tire aerodynamic drag coefficient values are screened out.

6. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein in the step 4, the functional relationship is to construct a regression function based on the five outer contour parameters of the selected tire model and the tire aerodynamic drag coefficient values of the testing schemes, and the regression function is a Kriging approximate model.

7. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 6, wherein in the step 4, a method for verifying the accuracy of the functional relationship is a determination coefficient R.sup.2 to check an accuracy of the Kriging approximate model; the closer to 1 the determination coefficient R.sup.2 is, the higher the accuracy of the Kriging approximate model is.

8. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 7, wherein an expression of the determination coefficient R.sup.2 is: $\begin{array}{cc}\frac{R^2=-\underset{i=1}{\overset{n}{.Math.}}{\left({\hat{y}}_i\stackrel{\_}{y}\right)}^2}{{\left(y_i-\stackrel{\_}{y}\right)}^2}& \left(1\right)\end{array}$ where n is a number of the testing schemes designed using the five outer contour parameters; ŷ.sub.i is a predictive value of an i-th response obtained using the Kriging approximate model; y.sub.i is a simulation value of an i-th testing scheme of the testing schemes obtained using the numerical simulation analysis; y is an average value of all predictive values.

9. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 1, wherein in the step 5, the intelligent optimization method is a simulated annealing optimization method, and the simulated annealing optimization method is used to conduct a nonlinear optimum analysis for a Kriging approximate model, and the optimized five outer contour parameters are proposed when the tire aerodynamic drag coefficient values are the smallest.

10. The design method of the tire outer contour structure for reducing the tire aerodynamic drag according to claim 9, a temperature update function of the simulated annealing optimization method is firstly designed as following:
t.sub.k=t.sub.0/ln(1+k)  (2) where t.sub.t is a temperature value at a step k, and t.sub.0 is an initial temperature value; a state acceptance function in the simulated annealing optimization method is designed as: $\begin{array}{cc}{p}_{\mathrm{ij}}=\{\begin{array}{cc}1& \Delta E\ge 0\\ \frac{1}{\left(1+\mathrm{esp}\frac{\Delta E}{t_k}\right)}& \Delta E<0\end{array}& \left(3\right)\end{array}$ where ΔE=E.sub.i−E.sub.j, E.sub.i is an energy function of a current solution state, E.sub.j is an energy function of a solution state to be accepted; the energy function ΔE is a value of an objective function; a solving process is to search in a neighborhood of the current solution state with a small step, and a new solution in the solution state to be accepted is obtained as the following equation (4):
X.sub.j=X.sub.i+lΔ  (4) where X.sub.i is a current solution, X.sub.j is the new solution, l is a number of a search cycle, and Δ is a search step size; a detail of the solving process of the simulated annealing optimization method are as follows: step 1: setting calculation parameters for the initial temperature value, a lowest temperature, the search step size, and an upper limit number of the search cycle; after then, randomly generating an initial solution X.sub.0, and setting the initial solution X.sub.0 as the current solution X.sub.i=X.sub.0; step 2: calculating the energy function E(X.sub.i) in the current solution state; and generating the new solution X.sub.j according to the equation (4); step 3: calculating the energy function E(X.sub.j) in the solution state to be accepted, and calculating the energy function ΔE; step 4: determining whether to accept the new solution X.sub.j according to ΔE using the equation (3); if ΔE<0, the new solution X.sub.j is accepted, and updating the current solution and setting X.sub.i=X.sub.j, and going to step 6; otherwise, if ΔE<0, going to step 5; step 5: determining whether the upper limit number of the search cycle is reached, if the upper limit number of the search cycle is satisfied, going to step 8; otherwise, updating the number of the search cycle, that is, l=l+1, and going to step 2; step 6: updating a temperature step k=k+1, and updating the temperature value according to the equation (2); step 7: determining whether the temperature value reaches a lowest temperature value, if not, going to step 2; otherwise, going to step 8; step 8: ending the solving process.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0037] FIG. 1 is the flow chart in this present invention;

[0038] FIG. 2 is the schematic diagram of the five outer contour parameters in this present invention;

[0039] FIG. 3 is the three-dimensional virtual wind tunnel model in this present invention;

[0040] FIG. 4 is the accuracy verification of the Kriging approximate model in the present invention;

[0041] FIG. 5 is the relationship between aerodynamic drag coefficient C.sub.d and shoulder transition radius R and tread radius R.sub.2 obtained using Kriging approximate model in this present invention;

[0042] FIG. 6 is the relationship between aerodynamic drag coefficient C.sub.d and the running surface width L.sub.2 and the upper sidewall height H obtained using Kriging approximate model in this present invention;

[0043] FIG. 7 is the relationship between aerodynamic drag coefficient C.sub.d and the tread radius R.sub.1 and the upper sidewall height H obtained using Kriging approximate model in this present invention;

[0044] FIG. 8 is the relationship between aerodynamic drag coefficient C.sub.d and shoulder transition radius R and upper sidewall height H obtained using Kriging approximate model in this present invention;

[0045] FIG. 9 is a comparison diagram of the layout of the tire outer contour structure before and after the optimization in the present invention.

[0046] FIG. 10 is a comparison diagram of the vortices distribution around tire before and after the optimization in the present invention.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The present invention will be further described below in conjunction with the drawings and specific embodiments.

[0048] As shown in FIG. 1, A design method for outer contour structure of tire for reducing tire wind resistance, comprising the following steps::

[0049] step 1: An initial tire model was selected, and its three-dimensional model was drawn using the initial values of the five outer contour parameters.

[0050] In example embodiment, a tire 185/65R15 was taken as an object, the section width is 185 mm, and half of the tread width is 97.5 mm, and the section height is 121 mm. According to the initial tire model, the five outer contour parameters are determined as the upper sidewall height H=57.06 mm, the running surface width L.sub.2=67.61 mm, the shoulder transition radius R=35.87 mm, the tread arc radius R.sub.1=817.391 mm and R.sub.2=327.83 mm, and the initial tire three-dimensional model as shown in FIG. 2 is established in CATIA software.

[0051] step 2: building an aerodynamic drag calculation model for the initial tire three-dimensional model using the initial values of the five outer contour parameters, and performing a numerical simulation analysis to obtain an aerodynamic drag coefficient value of the initial tire three-dimensional model.

[0052] Building a three-dimensional virtual wind tunnel model including tire and road surface, as shown in FIG. 3. The fluid domain grid and boundary layer grid around the tire three-dimensional model are generated using HYPERMESH software, and the aerodynamic drag coefficient value obtained using the computational fluid dynamics analysis software ANSYS Fluent

[0053] In order to simulate the tire rolling on the road at a speed of 54 km/h, an inlet airflow velocity is set to 54 km/h (15 m/s) in the direction of the tire rolling.

[0054] The boundary conditions of the three-dimensional virtual wind tunnel model are set as follows: the inlet velocity is set as 15 m/s; the outlet is a 0 Pa pressure outlet condition; a symmetrical model is adopted, and the center plane is set as a symmetrical surface; the road is a non-sliding wall surface with a moving speed 15 m/s along the air flow direction, the tire surface is a non-sliding wall surface with a rotational angular velocity 48.27 rad/s; all other surfaces are sliding wall surfaces with zero shear force.

[0055] After setting the boundary conditions, Firstly, a steady-state pressure base solver is used, the SST k-omega turbulence model is selected for flow steady calculations, a semi-implicit method for pressure linked equations-consistent (SIMPLEC) is used to resolve the coupling of pressure and velocity, and all discrete methods used the second-order upwind style to improve the calculation accuracy. The steady-state solution results are used in the unsteady calculation as initial conditions for transient analyses. The hybrid detached eddy simulations (DES) method is adopted to the unsteady calculation. In the unsteady calculation process, firstly, a time step 0.001 second is used to calculate 2000 steps, and its calculation time is 2 seconds, so that the instantaneous fluctuation of the flow field parameters is in a relative steady state. Then, the time step reduces to 0.0001 second and 10000 steps are continuously calculated, meanwhile, the aerodynamic drag coefficient of tire is monitored in the time domain; when the unsteady calculation is completed, the aerodynamic drag coefficient value is obtained by time-averaged method.

[0056] step 3: designing a testing scheme, and taking the five outer contour parameters of the selected tire as design variables, the value ranges of design variables are set, and the aerodynamic drag coefficient value of tire is taken as the target response value, a statistical sampling test design method is selected to generate testing schemes using the design variables, and a tire aerodynamic drag simulation analysis of each testing scheme is conducted, and the tire aerodynamic drag coefficient values of all the testing schemes are obtained.

[0057] The value ranges of design variables are set as follows: the section height of the initial tire is 121 mm, H is generally 0.47˜0.63 of the section height, where H∈[57.4, 76.5] mm, the tread arc radius R.sub.1∈[400, 1200] mm and R.sub.2∈[180, 350] mm, the running surface width L.sub.2 is generally 0.73 to 0.9 of the half of the tire section width, where L.sub.2∈[70, 90] mm, the shoulder transition radius is generally taken R∈[15, 45] mm. The Latin hypercube test design method is selected to generate testing schemes using the design variables, 24 groups of testing schemes are proposed. Simulation analysis is conducted using the three-dimensional virtual wind tunnel model, the tire aerodynamic drag coefficient values of all the testing schemes are shown in Table 1.

TABLE-US-00001 TABLE 1 Testing schemes and analysis results Testing schemes of Latin drag hyper- coefficient cube H/mm L.sub.2/mm R/mm R.sub.1/mm R.sub.2/mm value 1 57.06 67.61 35.87 817.391 327.83 0.402169833 2 58.72 71.96 39.78 747.826 194.78 0.4009206 3 62.87 80.65 45.00 539.13 268.7 0.346375822 4 70.35 71.09 42.39 782.609 320.43 0.430176384 5 62.04 81.52 33.26 817.391 350 0.430319355 6 69.52 69.35 37.17 1130.435 202.17 0.425949078 7 59.55 75.43 17.61 1095.652 305.65 0.396346664 8 67.86 65.87 28.04 1165.217 313.04 0.416845208 9 65.37 70.22 21.52 643.478 342.61 0.406634621 10 64.53 72.83 15.00 469.565 224.35 0.38621991 11 67.03 82.39 29.35 573.913 187.39 0.346136461 12 73.67 78.91 24.13 1026.087 335.22 0.439895011 13 61.21 65.00 22.83 886.957 231.74 0.423275194 14 68.69 78.04 20.22 1060.87 216.96 0.344710285 15 55.4 77.17 26.74 504.348 283.48 0.39798132 16 57.89 85.00 30.65 956.522 239.13 0.42155861 17 69.413 83.344 20.37 564.76 271.87 0.344919001 18 71.18 83.26 38.48 991.304 253.91 0.400896293 19 60.38 74.57 41.09 1200 268.7 0.40046475 20 56.23 75.43 18.91 921.739 180 0.346560972 21 63.7 66.74 34.57 400 261.3 0.405419515 22 74.5 68.48 25.43 713.043 246.52 0.421364689 23 65.37 84.13 16.3 678.261 290.87 0.341457935 24 72.84 73.7 43.7 608.696 209.57 0.408411731

[0058] step 4: The functional relationship between the tire five outer contour parameters and aerodynamic drag coefficient values of testing schemes is constructed according to the analysis results in Step 3, and its accuracy is verified using determination coefficient R.sup.2.

[0059] Kriging approximate model is used to construct the functional relationship between the tire five outer contour structure parameters and aerodynamic drag coefficient values under 24 testing schemes. Before the Kriging approximate model is used to analyze and calculate the relationship between the design variables and the target response values, the determination coefficient R.sup.2 was used to verify the accuracy of the Kriging approximate model. The expression of R.sup.2 was:

[00003]$\begin{array}{cc}\frac{R^2=-\underset{i=1}{\overset{n}{.Math.}}{\left({\hat{y}}_i\stackrel{\_}{y}\right)}^2}{{\left(y_i-\stackrel{\_}{y}\right)}^2}& \left(1\right)\end{array}$

[0060] where n is the number of testing schemes, n=24; ŷ.sub.i is the predictive value of the i-th response obtained using the Kriging approximate model; y.sub.i is the simulation value of the i-th testing scheme obtained using the numerical simulation analysis; y is the average value of all the predictive values.

[0061] The closer to 1 the determination coefficient R.sup.2 is, the higher the accuracy of the Kriging approximate model is. The determination coefficient R.sup.2 is 0.983. At the same time, the comparison between the predictive value using the Kriging approximate model and the simulation value using ANSYS Fluent is shown in FIG. 4. The comparison result of FIG. 4 further verifies the accuracy of the Kriging approximate model.

[0062] FIG. 5-8 shows the relationship between aerodynamic drag coefficient C.sub.d and two outer contour parameters obtained using Kriging approximate model. it can be seen from these figures that H and L.sub.2 has the greatest impact on C.sub.d, and the change trend of C.sub.d is complex, and a degree of nonlinearity between the response values and the variables is obvious, this means that the interaction of tire outer contour parameters is large. The tread arc radius R.sub.1 and R.sub.2 are positively correlated with C.sub.d, and C.sub.d increases with the increments of R.sub.1 and R.sub.2, and transition radius R has little influence on C.sub.d.

[0063] step 5: Simulated annealing optimization method is used to conduct nonlinear optimum analysis for the Kriging approximate model, and the optimized five outer contour parameters are proposed when the aerodynamic drag coefficient value is the smallest.

[0064] The optimization design problem of tire outer contour structure parameters is described by a mathematical optimization model, which is transformed into a problem of finding the smallest aerodynamic drag coefficient value. The simulated annealing optimization algorithm is used to perform nonlinear optimization for the Kriging approximate model to get the smallest aerodynamic drag coefficient value. The simulated annealing optimization algorithm firstly design the temperature update function as shown in equation (2):

t.sub.k=t.sub.0/ln(1+k)  (2)

[0065] where t.sub.k is the temperature value at step k, and t.sub.0 is the initial temperature value;

[0066] the state acceptance function in simulated annealing optimization method is designed as equation (3):

[00004]$\begin{array}{cc}{p}_{\mathrm{ij}}=\{\begin{array}{cc}1& \Delta E\ge 0\\ \frac{1}{\left(1+\mathrm{esp}\frac{\Delta E}{t_k}\right)}& \Delta E<0\end{array}& \left(3\right)\end{array}$

[0067] where ΔE=E.sub.i−E.sub.j, E.sub.i is the energy function of the current solution state, E.sub.j is the energy function of the solution state to be accepted; the energy function ΔE is the value of the objective function;

[0068] The solving process is to search in the neighborhood of the current solution state with a small step, and the new solution in the solution state to be accepted is obtained as the following equation (4):

X.sub.j=X.sub.i+lΔ  (4)

[0069] where X.sub.i is the current solution, X.sub.j is the new solution, l is the number of search cycle, and Δ is the search step size;

[0070] the detail solving process of the simulated annealing optimization method are as follows:

[0071] step 1: Setting calculation parameters for the initial temperature, the lowest temperature, the search step size and the upper limit number of the search cycle. After then, randomly generate an initial solution X.sub.0, and set it as the current solution X.sub.i=X.sub.0;

[0072] step 2: calculate the energy function E(X.sub.i) in the current solution state; and generate a new solution X.sub.j according to equation (4);

[0073] step 3: calculate the energy function E(X.sub.j) in the solution state to be accepted, and calculate the energy function ΔE;

[0074] step 4: determine whether to accept the new solution X.sub.j according to ΔE, using the equation (3); if ΔE<0, the new solution X.sub.j is accepted, and update the current solution and set X.sub.i=X.sub.j, and go to step 6; otherwise, if ΔE<0, go to step 5;

[0075] step 5: determine whether the upper limit number of the search cycle is reached, if it is satisfied, go to step 8; otherwise, update the number of search cycle, that is, l=l+1, and go to step 2;

[0076] step 6: update temperature step k=k+1, and update temperature value according to equation (2);

[0077] step 7: determine whether the temperature value has reached the lowest temperature value, if not, go to step 2; otherwise, go to step 8;

[0078] step 8: end the calculation process.

[0079] The initial temperature in the simulated annealing optimization method is set as 800°, and the lowest temperature is set as 0.05°. The search step size is Δ=0.01, the number of search cycle is 1000, and the energy function is the aerodynamic drag coefficient value. The simulated annealing optimization method is conduct to optimize the tire five outer contour parameters to get the smallest aerodynamic drag coefficient value. After optimization are completed, the optimized five outer contour parameters are determined when the aerodynamic drag coefficient value is smallest, and then the optimized tire outer contour structure is established. The optimized five outer contour parameters are H=72.01 mm, L.sub.2=79.78 mm, R=30.65 mm, R.sub.1=434.78 mm, R.sub.2=298.26 mm, and the optimized tire outer contour structure is shown in FIG. 9.

[0080] After the optimized tire outer contour structure established, Step 1 was used to simulate the aerodynamic drag coefficient value of the optimized tire is computed by using a three-dimensional virtual wind tunnel model presented in step 1. The comparison results before and after optimization are shown in Table 2. It can be seen from Table 2 that the aerodynamic drag coefficient value of optimized tire has decreased by 19.92% thus reducing tire wind resistance and improving tire aerodynamic characteristics.

TABLE-US-00002 TABLE 2 Comparison of aerodynamic drag coefficients before and after optimization Tire outer contour structure design Original Optimized parameters structure structure Upper sidewall 57.06 72.01 height H Running surface 67.61 79.78 width L.sub.2 Shoulder transition 35.87 30.65 radius R Tread arc radius R.sub.1 817.391 434.783 Tread arc radius R.sub.2 327.83 298.26 Aerodynamic drag 0.402169833 0.33389854 coefficient value C.sub.d

[0081] The example embodiments are the preferred embodiments in this present invention, but the present invention is not limited to the above-mentioned embodiments. Without departing from the essence of this present invention, any obvious improvements, substitutions or modifications that can be derived by those technicians in the field belong to the protection scope of the present invention.