Brake Disc Based on Bayesian Inference and Optimization Design Method

Abstract

The present invention discloses a brake disc and an optimization design method based on Bayesian inference. The brake disc includes an inner disc surface of the brake disc. The inner disc surface of the brake disc adopts a gradually changing curved surface structure with uneven convex and concave in the radial direction. The gradually changing curved surface structure includes an intermediate arc-shaped convex surface protruding outward, an outer arc-shaped concave surface arranged outside the intermediate arc-shaped convex surface and tangent to the intermediate arc-shaped convex surface, and an inner arc-shaped concave surface arranged inside the intermediate arc-shaped convex surface and tangent to the intermediate arc-shaped convex surface. The steps include: taking R1, R2, R3, L1, and L2 as design variables, selecting the average Mises stress of the satisfactory brake disc surface as the observed value, and reversely optimizing the design variables based on the Bayesian inference theory to realize the information update of the design variables, and finally determining the optimal solution of the design variables; the present invention increases the internal heat dissipation area, improves the heat dissipation capacity of the brake disc, the brake disc structure has good resistance to thermal cracking, can improve the thermal fatigue life of the brake disc, and improves the braking safety of the vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive braking components, and particularly to a brake disc based on Bayesian inference and an optimization design method. The present invention can improve the thermal fatigue life of the brake disc and enhance the braking safety of the vehicle. Background Art

[0002] The brake disc is a key component used for braking or decelerating in many heavy vehicles. Frequent braking will cause the brake disc surface to generate high temperatures, resulting in thermal fatigue failure. Especially under emergency braking conditions, the surface temperature may reach six or seven hundred degrees Celsius. Thermal fatigue failure is the most common failure form of the brake disc. How to improve the thermal fatigue life of the brake disc is a technical problem that the heavy truck industry urgently needs to solve. Under limited conditions, the usual design method is to improve the heat dissipation capacity or the resistance to thermal stress of the brake disc by improving its structure, thereby enhancing its thermal fatigue life.

[0003] The technical state of the inner surface of the brake disc in the prior art is usually a flat structure, such as Figure 3 、 Figure 6 shown. The defect of the inner surface structure of the brake disc in the prior art is that the corresponding heat dissipation space is small, the heat dissipation area is limited, and the thickness L between the disc surface near the journal and the inner surface remains uniform. However, the heat generated by braking at the disc surface near the journal is not uniform, but decreases radially from the average friction radius to both sides. At the places with relatively large heat and relatively small heat, the thickness L of the brake disc surface is equal, which will cause the temperature of the brake disc surface to also show a decreasing phenomenon from the average friction radius to both sides. The temperature difference will exacerbate the generation of internal thermal stress in the brake disc. Therefore, this structure is also very limited in improving the thermal stress of the brake disc. The present invention proposes a structure in which the thickness L changes continuously in the radial direction, with a larger thickness L at the places with relatively large heat and a smaller thickness L at the places with relatively small heat. This structure is dedicated to improving the thermal stress of the brake disc surface, making the temperature of the brake disc surface relatively uniform, and reducing the thermal stress generated due to the temperature difference. Summary of the Invention

[0004] The present invention is to overcome the above-mentioned drawbacks existing in the prior art and provides a brake disc based on Bayesian inference. The brake disc with a thermal crack-resistant structure has the inner surface 4a of the brake disc designed as a curved surface structure instead of the traditional flat surface.

[0005] The present invention also discloses an optimization design method for a brake disc based on Bayesian inference.

[0006] The technical solution adopted by the present invention to solve its technical problems is:

[0007] A brake disc based on Bayesian inference includes an inner brake disc surface, and the inner brake disc surface adopts a gradually changing curved surface structure with uneven convex and concave in the radial direction.

[0008] The gradient surface structure includes a middle arc convex surface protruding outward, an outer arc concave surface disposed outside the middle arc convex surface and tangent to the middle arc convex surface, and an inner arc concave surface disposed inside the middle arc convex surface and tangent to the middle arc convex surface.

[0009] The middle arc convex surface, the outer arc concave surface, and the inner arc concave surface are circular arcs.

[0010] The center positioning dimensions of the middle arc convex surface circular arc R2 are the horizontal positioning dimension L1 and the vertical positioning dimension L2. Among them, the horizontal positioning dimension L1 is based on the brake disc flange surface, and the vertical positioning dimension L2 is based on the brake disc rotation center line. The outer arc concave surface R1 is tangent to the middle arc convex surface R2 in the first quadrant, and the inner arc concave surface R3 is tangent in the fourth quadrant.

[0011] Reinforcing ribs are provided on the outer arc concave surface R1, the middle arc convex surface R2, and the inner arc concave surface R3.

[0012] A first reinforcing rib is provided at the position where the distance L between the middle arc convex surface R1 and the disc surface near the journal is the smallest, a second reinforcing rib is provided at the position where the distance L between the outer arc concave surface R2 and the disc surface near the journal is the largest, and a third reinforcing rib is provided at the position where the distance L between the inner arc concave surface R3 and the disc surface near the journal is the smallest.

[0013] The positions of the first reinforcing rib, the second reinforcing rib, and the third reinforcing rib are staggered.

[0014] An optimization design method for a brake disc based on Bayesian inference, with R1, R2, R3, L1, and L2 as design variables. By selecting the satisfactory average Mises stress of the brake disc surface as the observed value, the design variables are inversely optimized based on the Bayesian inference theory to update the information of the design variables, and finally the optimal solutions of the design variables are determined.

[0015] The specific steps are as follows:

[0016] 1) Calculate the total energy sum Q generated during the test braking process through the bench test conditions 总 :

[0017]

[0018] In the formula: Q 总 —— The total energy generated during the braking process, unit J;

[0019] M—— Braking torque, unit Nm;

[0020] v—— Vehicle speed, unit m / s;

[0021] R - wheel radius, unit: m;

[0022] 2) It is theoretically assumed that 95% of the heat will be transferred to the brake disc and 5% of the heat will be transferred to the friction pair. Calculate the heat flux density q of the disc surface near the journal and the outer disc surface:

[0023]

[0024] where, T - braking time, unit: s; S - friction surface area, unit: mm 2 ;

[0025] Calculate the heat flux density q of the disc surface near the journal 1 , and the heat flux density q of the outer disc surface 2 ;

[0026] 3) According to experience, the distribution types of the five design variables, namely the radius R2 of the convex surface of the middle arc, the radius R1 of the concave surface of the outer arc, the radius R3 of the concave surface of the inner arc, the horizontal positioning dimension L1, and the vertical positioning dimension L2, all follow the normal distribution. Based on experience, preliminarily determine the statistical distribution of these five design variables, and extract multiple groups of sample points from the five initial design variables according to the Latin hypercube sampling method;

[0027] 4) Design the corresponding number of groups of inner disc structures according to the group of sample points in step 3). Use CAE software to perform CAE thermal - mechanical coupling simulation analysis on the brake discs with the corresponding number of groups of inner disc structures by loading the surface heat flux density on the disc surface of the brake disc, and obtain the corresponding number of groups of temperature fields and Mises stress fields;

[0028] 5) According to the kriging theory, using several groups of samples θ = [R1, R2, R3, R4, R5] as the input of design variables and the average Mises stress of the disc surface in step 4) as the output, use MATLAB software for numerical fitting to establish a kriging surrogate model, that is, establish the function S (θ) ;

[0029] 6) Verify the accuracy of the kriging surrogate model obtained in step 5). Substitute two groups of sample points into the kriging surrogate model to obtain two groups of response values, and compare them with the response values obtained from the CAE simulation model to verify the accuracy of the surrogate model;

[0030] 7) Select two groups of satisfactory average Mises stresses of the brake disc surface as the observed data;

[0031] Conduct CAE simulation analysis on the brake disc with a flat inner disc surface structure in the original state. Calculate the average Mises stress of the brake disc surface by taking the Mises stresses at multiple points evenly at the friction radius of the brake disc surface;

[0032] 8) Apply Bayesian inference theory and the M-H algorithm to iteratively update the information of the design parameters. After sufficient iterations, the posterior distribution sample of the design parameters approximately follows a normal distribution.

[0033] Take the mean value of the posterior distribution of the design variables, and after rounding, obtain the design value of the inner disc surface structure of the brake disc.

[0034] 9) Round the obtained optimal solution to be used as the design value of the inner disc surface structure. After determining the inner disc surface structure parameters of the brake disc, a reinforcing rib structure is provided at the arc surface.

[0035] The beneficial effects of the present invention are as follows:

[0036] 1. In the present invention, the inner disc surface of the brake disc is designed from the planar structure of the prior art to a gradient curved surface structure. The inner disc surface of the brake disc is designed as a gradient curve structure, which means that the distance L between the inner disc surface and the disc surface near the journal continuously changes with the change of the radial distance, so as to increase the surface area of the inner disc surface and improve the heat dissipation capacity of the brake disc.

[0037] 2. By providing a ventilation rib structure at the extreme values of the inner disc surface curves R1, R2, and R3, the heat dissipation area at the internal ventilation ribs is increased, and the heat dissipation capacity of the brake disc is improved. The brake disc structure has good resistance to thermal cracking, can improve the thermal fatigue life of the brake disc. At the same time, combined with the transformation of the inner surface of the brake disc from the traditional planar design to a curved surface structure, the heat dissipation capacity at the internal ventilation ribs is further enhanced, which can improve the thermal fatigue life of the brake disc and improve the braking safety of the vehicle.

[0038] 3. The present invention uses Bayesian inference theory to update the information of the design variables, which can efficiently and accurately find the reasonable design values of the design variables, and reduce the design and manufacturing cycle and cost.

[0039] 4. The present invention uses a "black box" function to replace the traditional CAE simulation model, which greatly improves the design and R & D efficiency, reduces the time cost of simulating and verifying new products, and reduces the design and manufacturing cycle.

[0040] 5. The present invention preferably uses a stress about 15% higher than the average Mises stress of the original brake disc surface as the observed value, and calculates that the average Mises stress of the brake disc surface of the present invention is improved by 12% compared with the original brake disc surface. Verified by bench tests, the thermal fatigue life of the anti-thermal cracking structure brake disc of the present invention has achieved a breakthrough improvement of more than 50% compared with the original brake disc.

[0041] 6. In the present invention, five design variables are used as inputs through the Kriging surrogate model, and the average Mises stress on the brake disc surface obtained from CAE simulation analysis is used as the output. The numerical integration method is used to fit the functional relationship between the input and output, and this functional relationship is used as a "black box" function (i.e., implicit functional relationship) to replace the CAE simulation analysis in the optimization process. By selecting a satisfactory average Mises stress on the brake disc surface as the observed value and based on the Bayesian inference theory, the design variables are inversely optimized to update the information of the design variables, and finally the optimal solution of the design variables is determined. The present invention can significantly improve the design and R & D efficiency, reduce the time cost of simulation verification for developing new products, and shorten the design and manufacturing cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the inner disc surface structure of the brake disc of the present invention;

[0043] Figure 2 It is a schematic diagram of the position of the reinforcing ribs on the inner disc surface of the brake disc of the present invention;

[0044] Figure 3 It is a schematic diagram of the inner disc surface structure of the existing brake disc;

[0045] Figure 4 It is a flowchart of the M-H algorithm;

[0046] Figure 5 It is a Mises stress nephogram of the brake disc surface of the present invention;

[0047] Figure 6 It is a Mises stress nephogram of the brake disc surface of the original structure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0049] As Figure 1 , shown in FIGS. 2, 4 - 6, a brake disc with a thermal crack-resistant structure includes an inner disc surface 4a of the brake disc. The inner disc surface 4a of the brake disc adopts a gradually changing curved surface structure with uneven convex and concave in the radial direction. The gradually changing curved surface structure means that the distance L between the inner disc surface 4a and the disc surface 3 near the journal continuously changes with the change of the radial distance. In this embodiment, an arc transition curve is adopted.

[0050] The surface structure includes a radially outwardly convex intermediate arc convex surface, an outer arc concave surface provided on the radially outer side of the intermediate convex surface and tangent to the intermediate convex surface, and an inner arc concave surface provided on the radially inner side of the intermediate convex surface and tangent to the intermediate convex surface. Among them, the radius of the intermediate arc convex surface is R2, the radius of the outer arc concave surface is R1, and the radius of the inner arc concave surface is R3.

[0051] The center positioning dimensions of the arc R2 of the intermediate arc convex surface are the horizontal positioning dimension L1 and the vertical positioning dimension L2. Among them, the horizontal positioning dimension L1 is based on the brake disc flange surface 1, and the vertical positioning dimension L2 is based on the brake disc rotation center line 2. In the coordinate system with the center of the intermediate arc convex surface R2 as the origin, the outer arc concave surface R1 is tangent to the intermediate arc convex surface R2 in the first quadrant, and the inner arc concave surface R3 is tangent to the intermediate arc convex surface R2 in the fourth quadrant.

[0052] Reinforcing rib structures are provided on the outer arc concave surface R1, the intermediate arc convex surface R2, and the inner arc concave surface R3 to increase the ability of the brake disc to resist thermal stress.

[0053] A first reinforcing rib is provided at the minimum distance L between the intermediate arc convex surface R1 and the disc surface 3 near the journal. A second reinforcing rib is provided at the maximum distance L between the outer arc concave surface R2 and the disc surface 3 near the journal. A third reinforcing rib is provided at the minimum distance L between the inner arc concave surface R3 and the disc surface 3 near the journal.

[0054] The positions of the first reinforcing rib, the second reinforcing rib, and the third reinforcing rib are staggeredly distributed.

[0055] In the present invention, the inner disc surface 4a of the brake disc is designed as a curved arc structure, which enhances the heat dissipation capacity at the internal reinforcing ribs, and a reinforcing rib structure is provided at the extreme value of the curved surface to increase the ability of the brake disc to resist thermal deformation. Through simulation analysis and bench test verification, the brake disc with the structure of the present invention has a significantly improved thermal fatigue life compared with the brake disc with a flat inner disc surface 4a.

[0056] A brake disc anti-thermal cracking structure and its design method are as follows:

[0057] 1) Calculate the total energy sum Q generated during the test braking process through the bench test conditions 总 :

[0058]

[0059] In the formula: Q 总 —— The total energy generated during the braking process, unit J;

[0060] M —— Braking torque, unit Nm;

[0061] v —— vehicle speed, unit: m / s;

[0062] R —— wheel radius, unit: m;

[0063] In this embodiment, the test conditions are: vehicle speed = 85 km / h, braking torque M = 2800 Nm, wheel radius R = 0.51 m, and dragging time t = 40 s. The total energy generated during the braking process is Q 总 = 5182745.1 J.

[0064] 2) It is theoretically assumed that 95% of the heat will be transferred to the brake disc, and 5% of the heat will be transferred to the friction pair. Calculate the heat flux density q of the disc surface 3 near the journal and the outer disc surface 6:

[0065]

[0066] where, T —— braking time, unit: s; S —— friction surface area, unit: mm 2 ;

[0067] The calculated heat flux density q of the disc surface 3 near the journal 1 = 0.782, and the heat flux density q of the outer disc surface 6 2 = 0.786.

[0068] 3) According to experience, the distribution types of the five design variables, namely the radius R2 of the middle arc convex surface, the radius R1 of the outer arc concave surface, the radius R3 of the inner arc concave surface, the horizontal positioning dimension L1, and the vertical positioning dimension L2, all follow a normal distribution. Based on experience, the statistical distributions of these five design variables are initially determined. According to the Latin hypercube sampling method, multiple groups of sample points are extracted from the five initial design variables. In this embodiment, 7 groups of sample points are extracted, as shown in the following table:

[0069] L1 L2 R1 R2 R3 1 14.7335 164.1107 100.2335 171.1808 99.7409 2 19.7373 160.9388 101.9962 175.2373 104.7561 3 15.6808 161.8828 98.7409 168.7409 102.9962 4 16.4962 162.9976 105.2373 170.2335 103.8111 5 13.2409 165.0475 101.1808 172.8111 106.2373 6 17.3111 162.4819 102.8111 173.7561 101.2335 7 18.2561 163.5130 103.7561 171.9962 102.1808

[0070] 4) Design 7 types of inner disc surface 4a structures according to the 7 groups of sample points in step 3). By loading the surface heat flux density on the brake disc surface 3 using CAE software, perform CAE thermo-mechanical coupling simulation analysis on the brake discs with 7 types of inner disc surface 4 structures, and obtain 7 corresponding groups of temperature fields and Mises stress fields. In this example, the average Mises stress of the disc surface 3 is extracted as the response value.

[0071] 5) According to the kriging theory, using 1 - 5 groups of samples θ = [R1, R2, R3, L1, L2] as the input of the design variables and the average Mises stress of the disc surface 3 in step 4) as the output, use MATLAB software for numerical fitting to establish a kriging surrogate model, that is, establish a "black box" function S(θ).

[0072] 6) Verify the accuracy of the kriging surrogate model obtained in step 5). Substitute the sample points of groups 6 and 7 into the kriging surrogate model to obtain two groups of response values, and compare them with the response values obtained from the CAE simulation model to verify the accuracy of the surrogate model. In this embodiment, it is considered that the error within 1% meets the accuracy requirement.

[0073] 7) Select two satisfactory average Mises stresses s 1 、s 2 of the brake disc surface 3 as the observed data.

[0074] Conduct a CAE simulation analysis on the brake disc with the inner surface 4a of the original state brake disc being a flat structure. Uniformly take the Mises stresses at multiple points at the friction radius of the brake disc surface 3 to calculate the average Mises stress of the brake disc surface 3. In this embodiment, the Mises stresses at 10 points are taken to calculate the average Mises stress of the brake disc surface 3, as shown in the following table:

[0075] 1 2 3 4 5 6 7 8 9 10 321.1 300.1 312.4 344.3 314.0 320.5 370.8 323.1 350.6 332.7

[0076] It is calculated that the average Mises stress of the brake disc surface 3 in the original state is 329 Mpa.

[0077] In this embodiment, the stress about 15% better than the average Mises stress of the brake disc surface 3 in the original state is selected as the observed value s 1 = 280 Mpa, s 2 = 282 Mpa. Assume that the observation error s e = 1 Mpa, follows a normal distribution, and the prior distribution type of the parameters is a multi-dimensional normal distribution. For the observed data s 1 、s 2 The likelihood function of Bayes' theorem is:

[0078]

[0079] Obtain the posterior distribution function after two updates:

[0080]

[0081] 8) Apply Bayes' inference theory and the M-H algorithm to iteratively update the design parameters. After sufficient iterations, the posterior distribution samples of the design parameters approximately follow a normal distribution: L1 post ~ N(18.2, 1.13), L2 post ~ N(163.6, 0.62), R1 post ~ N(100.8, 1.36), R2 post ~ N(170.3, 1.43), R3 post~N(100.5, 1.36), and the number of iterations in this embodiment is set to 10,000 times.

[0082] Take the mean value of the posterior distribution of the design variables, and after rounding, obtain the design values of the structure of the inner disc surface 4a of the brake disc, as shown in the following table:

[0083]

[0084] The specific implementation process of the M-H algorithm in step 8) is as follows:

[0085] Combined with Figure 4 , Figure 4 where N is the number of loop iterations, is the initial value selected from the sample space of the design variables, is the candidate point generated from the proposal density function when the initial design variables are known, is the acceptance probability of the current candidate point, and is a random number obeying the uniform distribution.

[0086] a. Select the prior distribution function of the design variable θ, that is, the multi-dimensional normal distribution function as the proposal density function f(θ 1 |θ 0 ), and generate the candidate point θ 1 of the parameter to be estimated;

[0087] b. Calculate the acceptance probability of the candidate point, denoted as α, which can be expressed by the following formula:

[0088]

[0089] In the formula: S obv — is the observed value of the maximum Mises stress;

[0090] θ 0 — is the initially selected design variable;

[0091] θ 1 — through the proposal density function f(θ 1 |θ 0 ), generate the candidate point of the parameter to be estimated;

[0092] p(θ 1 |S obv )— is the posterior distribution function of the parameter to be estimated;

[0093] c. Generate a random number μ obeying the uniform distribution (0, 1).

[0094] d. Compare the values of μ and α. If μ ≤ α, accept the candidate point; otherwise, reject the candidate point and enter the next loop.

[0095] 9) Round the obtained optimal solution and use it as the design value of the inner disc surface structure. After determining the structural parameters of the inner disc surface 4a of the brake disc, a reinforcing rib structure is provided at the arc-shaped curved surface to increase the ability of the brake disc to resist thermal stress.

[0096] Perform CAE simulation analysis on the newly designed brake disc, and calculate the average Mises stress of the brake disc surface 3 by taking the Mises stress at 10 evenly distributed points at the friction radius of the disc surface 3 near the journal:

[0097]

[0098] Combined with Figure 5 、 Figure 6 From Figure 5 、 Figure 6 It can be seen that the Mises stress of the disc surface in the embodiments of the present invention is mainly in state a, and state b is much less than that of the existing ordinary brake discs ( Figure 6 ), while the Mises stress of the disc surface of the existing ordinary brake discs is mainly in state b. The Mises stress of the disc surface in the embodiments of the present invention is superior to that of the existing ordinary brake discs.

[0099] It is calculated that the average Mises stress of the disc surface 3 near the journal of the present invention is 289.8 Mpa. Compared with the stress of 329 Mpa of the brake disc in the original state, the disc surface stress of the brake disc of the present invention is improved by 12%. Verified by bench tests, the thermal fatigue life of the anti-thermal crack structure brake disc of the present invention has achieved a breakthrough improvement of more than 50% compared with the original brake disc.

[0100] In the description of the present invention, the orientation or positional relationship indicated by the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "vertical", "horizontal", etc. is based on the orientation or positional relationship shown in the drawings, and is only for describing the present invention rather than requiring the present invention to be constructed or operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. The "connected" and "connected" in the present invention should be understood in a broad sense. For example, it can be a connection or a detachable connection; it can be a direct connection or an indirect connection through an intermediate component. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0101] The above is the preferred embodiment of the present invention. The description of specific embodiments is only for better understanding the idea of the present invention. For those of ordinary skill in the art of this technology, several improvements or equivalent replacements can be made according to the principle of the present invention, and these improvements or equivalent replacements are also regarded as falling within the protection scope of the present invention.

Claims

1. A brake disc based on Bayesian inference, including the inner disc surface of the brake disc, Characterized in that, The inner disc surface of the brake disc adopts a gradually changing curved surface structure with uneven convex and concave in the radial direction; The gradually changing curved surface structure includes an intermediate arc convex surface protruding outward, an outer arc concave surface arranged outside the intermediate arc convex surface and tangent to the intermediate arc convex surface, and an inner arc concave surface arranged inside the intermediate arc convex surface and tangent to the intermediate arc convex surface; The intermediate arc convex surface, the outer arc concave surface, and the inner arc concave surface are circular arcs; The center positioning dimensions of the intermediate arc convex surface circular arc R2 are the horizontal positioning dimension L1 and the vertical positioning dimension L2. Among them, the horizontal positioning dimension L1 is based on the flange surface of the brake disc, and the vertical positioning dimension L2 is based on the rotation center line of the brake disc. The outer arc concave surface R1 is tangent to the intermediate arc convex surface R2 in the first quadrant, and the inner arc concave surface R3 is tangent in the fourth quadrant; The optimization design method of the brake disc based on Bayesian inference takes R1, R2, R3, L1, and L2 as design variables, selects the average Mises stress of the satisfactory brake disc surface as the observed value, and based on the Bayesian inference theory, reversely optimizes the design variables to realize the information update of the design variables, and finally determines the optimal solution of the design variables; The specific steps are as follows: 1) Calculate the total energy sum Q generated during the test braking process under the bench test conditions 总 : Where: Q 总 —— The total energy generated during the braking process, unit: J; M - braking torque, unit Nm; v - vehicle speed, unit m / s; R - wheel radius, unit m; 2) Theoretically, 95% of the heat will be transferred to the brake disc, and 5% of the heat will be transferred to the friction pair. Calculate the heat flux density q of the near-journal surface and the outer surface: Among them, T—the braking time, unit: s; S—the friction surface area, unit: mm 2 ; The heat flux density q of the disk surface near the journal is calculated 1 , and the heat flux density q of the outer disk surface 2 ; 3) According to experience, the distribution types of the five design variables, namely the radius R2 of the intermediate arc convex surface, the radius R1 of the outer arc concave surface, the radius R3 of the inner arc concave surface, the horizontal positioning dimension L1, and the vertical positioning dimension L2, all follow a normal distribution. According to experience, preliminarily determine the statistical distribution of these five design variables, and extract multiple groups of sample points from the five initial design variables according to the Latin hypercube sampling method; 4) Design the corresponding number of inner disc surface structures according to the group of sample points in step 3). Use CAE software to perform CAE thermal-structural coupling simulation analysis on the brake discs with the corresponding number of inner disc surface structures by loading the surface heat flux density on the brake disc surface, and obtain the corresponding number of groups of temperature fields and Mises stress fields; 5) According to the kriging theory, take several groups of samples θ = [R1, R2, R3, R4, R5] as the input of the design variables, and take the average Mises stress of the disc surface in step 4) as the output. Use MATLAB software for numerical fitting to establish a kriging surrogate model, that is, establish the function S(θ); 6) Verify the accuracy of the kriging surrogate model obtained in step 5). Substitute two groups of sample points into the kriging surrogate model to obtain two groups of response values, and compare them with the response values obtained from the CAE simulation model to verify the accuracy of the surrogate model; 7) Select two groups of satisfactory average Mises stresses of the brake disc surface as the observed data; Perform CAE simulation analysis on the brake disc with a flat inner disc surface in its original state, and calculate the average Mises stress of the brake disc surface by taking the Mises stress at multiple points evenly distributed at the friction radius of the brake disc surface. 8) Apply the Bayesian inference theory and the M-H algorithm to update the iterative information of the design parameters. After sufficient iterations, the posterior distribution sample of the design parameters approximately follows a normal distribution. Take the mean value of the posterior distribution of the design variables, and obtain the design value of the inner disc surface structure of the brake disc after rounding. 9) Take the optimized solution obtained after rounding as the design value of the inner disc surface structure. After determining the structural parameters of the inner disc surface of the brake disc, a reinforcing rib structure is provided at the arc surface.

2. The brake disc based on Bayesian inference according to claim 1, characterized in that, Reinforcing ribs are provided on the outer arc concave surface R1, the middle arc convex surface R2, and the inner arc concave surface R3.

3. The brake disc based on Bayesian inference according to claim 2, characterized in that, A first reinforcing rib is provided at the position where the distance L between the middle arc convex surface R2 and the disc surface near the journal is the smallest, a second reinforcing rib is provided at the position where the distance L between the outer arc concave surface R1 and the disc surface near the journal is the largest, and a third reinforcing rib is provided at the position where the distance L between the inner arc concave surface R3 and the disc surface near the journal is the smallest.

4. The brake disc based on Bayesian inference according to claim 3, characterized in that, The positions of the first reinforcing rib, the second reinforcing rib, and the third reinforcing rib are staggered.

Images