Method for controlling NVH gear rattle noise of electric power drive system of pure electric vehicle

Abstract

A method for controlling NVH gear knocking noise in the electric drive system of a pure electric vehicle includes the following steps: 1) Establishing a multiple linear regression equation for the vibration amplitude of the reducer, the slope of the motor torque change, and the backlash of the reducer gears in the electric drive system; 2) Collecting several sets of experimental data; 3) Substituting the several sets of experimental data collected in step 2) into the multiple linear regression equation in step 1), and solving the regression coefficients of the motor torque change slope, the backlash of the reducer gears, and the constant b0 according to the least squares method, to obtain the multiple linear regression equation with determined regression coefficients; 4) Setting the reducer vibration amplitude threshold according to requirements, and substituting it into the multiple linear regression equation with determined regression coefficients, and uniformly modifying and adjusting the reducer and motor of the electric drive system.

Description

Technical Field

[0001] This invention relates to the field of transmissions, and more specifically to a method for controlling NVH gear knocking noise in the electric drive system of a pure electric vehicle. Background Technology

[0002] The electric drive system of a pure electric vehicle typically uses a motor directly connected to a single-stage reducer (containing three parallel shafts), which is a two-stage cylindrical helical gear structure. During the tip-in / tip-out transition (accelerating / decelerating), the transient change in motor torque can easily cause knocking between the gear teeth. This knocking energy is transmitted through the bearings to the vehicle housing and interior, producing vibration, impact, and a "clattering" sound.

[0003] Compared to traditional cars, pure electric vehicles are quieter inside due to the elimination of the engine and multi-speed transmission. However, passengers are more sensitive to the NVH (noise, vibration, and harshness) noise from the motor and reducer. To minimize NVH noise from the electric drive system and improve the passenger experience, this issue is typically addressed through the following two approaches:

[0004] ① Reduce the vibration of the reducer in the electric drive system by controlling the slope of the motor torque change.

[0005] ② By controlling the design parameters of the reducer, the backlash of the gear teeth in the reducer is adjusted to ensure that the gears in the reducer will not knock between the backlash of the gear teeth due to the transient change of the motor torque.

[0006] However, gearbox manufacturers and motor manufacturers often operate independently. Even after the gearbox and motor of the same electric drive transmission system have been debugged and modified by the two manufacturers, the impact of the above two problems on the NVH of the electric drive system cannot be eliminated to the greatest extent. As a result, after the gearbox and motor are installed in the vehicle, the electric drive transmission system of the pure electric vehicle still has NVH noise problems that make passengers dissatisfied. This not only increases the unnecessary development costs, but also greatly reduces the passenger experience. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for controlling NVH gear knocking noise in the electric drive system of a pure electric vehicle. This method utilizes a multiple linear regression equation to find the corresponding relationship between the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash in the electric drive system. This relationship is then used to uniformly adjust and rectify the motor and reducer in the electric drive system, thereby minimizing NVH noise issues in the electric drive transmission system and improving the passenger experience.

[0008] The objective of this invention is achieved through the following solution: a method for controlling NVH (noise, vibration, and harshness) gear knocking noise in the electric drive system of a pure electric vehicle, comprising the following steps:

[0009] 1) The following multiple linear regression equations are established for the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash of the reducer in the electric drive system:

[0010]

[0011] In the formula, The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant;

[0012] 2) Vibration data experiments were conducted on the electric drive system using a triaxial acceleration sensor, and several sets of experimental data were collected. The experimental data included the vibration amplitude of the reducer, the slope of the motor torque change, and the backlash of the reducer gears at the same time.

[0013] 3) Substitute the experimental data collected in step 2) into the multiple linear regression equation in step 1), and solve the multiple linear regression equation for the motor torque change slope, the reducer gear backlash regression coefficient, and the constant using the least squares method. Thus, the multiple linear regression equation with determined regression coefficients is obtained;

[0014] 4) Set the vibration amplitude threshold of the reducer according to requirements, and substitute it into the multiple linear regression equation determined by the regression coefficients. Then, uniformly modify and debug the reducer and motor of the electric drive system in the following manner:

[0015] 4-1) First, determine the value of the gear backlash of the reducer according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding motor torque change slope. Control the motor of the electric drive system according to the motor torque change slope.

[0016] 4-2) First, determine the value of the slope of the motor torque change according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding gear backlash of the reducer. Adjust the design parameters of the reducer according to the gear backlash of the reducer.

[0017] Preferably, based on the principle of least squares, the regression coefficients of the motor torque change slope, the gear backlash of the reducer, and the constants are obtained by solving the multiple linear regression equation. The specific steps to obtain the multiple linear regression equation with determined regression coefficients are as follows:

[0018] ① According to the principle of least squares, the regression coefficients , , It should satisfy:

[0019]

[0020] In the formula, This represents the sum of squares of the deviations between the measured and theoretical values ​​of the reducer vibration amplitude in the vibration data experiment. The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant. This represents the theoretical value of the reducer vibration amplitude at time n in the vibration data experiment.

[0021] ② In the formula of step ①, the regression coefficients , , Taking the partial derivative and then setting it to zero, we obtain the following system of normal equations:

[0022]

[0023] ③ Solve the normal equation system from step ② using the matrix method to obtain the regression coefficients. , , The values ​​are as follows:

[0024] .

[0025] The advantage of this invention is that it uses a multiple linear regression equation to find the correspondence between the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash of the reducer in the electric drive system. It then uses this correspondence to uniformly debug and rectify the motor and reducer in the electric drive system, so as to minimize the NVH noise problem of the electric drive transmission system and improve the passenger experience. Attached Figure Description

[0026] Figure 1 This is a flowchart of the present invention;

[0027] Figure 2 This is a schematic diagram of the residual analysis of the multiple linear regression equation of the present invention. Detailed Implementation

[0028] like Figure 1 As shown, a method for controlling NVH gear knocking noise in the electric drive system of a pure electric vehicle includes the following steps:

[0029] 1) The following multiple linear regression equations are established for the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash of the reducer in the electric drive system:

[0030]

[0031] In the formula, The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant;

[0032] 2) Vibration data experiments were conducted on the electric drive system using a triaxial accelerometer, and several sets of experimental data were collected. The experimental data included the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash of the reducer at the same moment, as shown in Table 1:

[0033] Table 1

[0034]

[0035] 3) Substitute the experimental data collected in step 2) into the multiple linear regression equation in step 1), and solve the multiple linear regression equation for the motor torque change slope, the reducer gear backlash regression coefficient, and the constant using the least squares method. The specific steps to obtain the multiple linear regression equation with determined regression coefficients are as follows:

[0036] ① According to the principle of least squares, the regression coefficients , , It should satisfy:

[0037]

[0038] In the formula, This represents the sum of squares of the deviations between the measured and theoretical values ​​of the reducer vibration amplitude in the vibration data experiment. The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant. This represents the theoretical value of the reducer vibration amplitude at time n in the vibration data experiment.

[0039] The regression coefficient should be obtained when the sum of the squares of the deviations between the measured vibration amplitude values ​​and the theoretical values ​​of the reducer in the vibration data experiment is minimized. , , .

[0040] ② In the formula of step ①, the regression coefficients , , Taking the partial derivative and then setting it to zero, we obtain the following system of normal equations:

[0041]

[0042] ③ Solve the normal equation system from step ② using the matrix method to obtain the regression coefficients. , , The values ​​are as follows:

[0043] .

[0044] In this embodiment, the regression coefficient =-24.29, =26.03, =30.03.

[0045] 4) Set a vibration amplitude threshold for the reducer to ensure that the vibration amplitude of the reducer always meets the requirements (i.e., the vibration amplitude of the reducer when no gear knocking occurs is generally used as the threshold), and substitute it into the multiple linear regression equation determined by the regression coefficients. Then, uniformly rectify and debug the reducer and motor of the electric drive system in the following manner:

[0046] 4-1) First, determine the value of the gear backlash of the reducer according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding motor torque change slope. After the whole machine is installed in the vehicle, the motor of the electric drive system is controlled according to the motor torque change slope.

[0047] 4-2) First, determine the value of the slope of the motor torque change according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding gear backlash of the reducer. Adjust the design parameters of the reducer according to the gear backlash of the reducer.

[0048] This invention not only underwent bench / vehicle NVH testing verification, but also used Minitab software to establish a regression model based on a multiple linear regression equation. Significance tests and residual analysis were performed on the regression equation. In other words, the results were validated for consistency using significance metrics and residual analysis. All of these demonstrate that this invention can achieve the optimal match between the motor torque change slope and the reducer gear backlash, improving the NVH noise performance of new energy electric vehicles and saving on vehicle development and verification costs.

[0049] ① Significance measure

[0050] Let P be the value for rejecting the null hypothesis, and a be the significance level. If P < a, that is, below the significance level a, the regression coefficient is significant.

[0051] The test showed that the p-value was zero and the α-value was 0.05, meaning the p-value was < 0.05. This proves that the total effect of the regression equation is significant below the significance level α = 0.05.

[0052] The P-values ​​of the independent variables, the slope of the change in motor torque and the gear backlash of the reducer, are both less than a=0.05, therefore both factors are significant.

[0053] The coefficient of determination and the adjusted coefficient of the regression model are both greater than 85% and are quite close.

[0054] The regression model has a prediction coefficient of 90.7%, which means that the prediction coefficient of the regression model is greater than 80%, indicating that the regression model has a good predictive effect.

[0055] ② Residual Analysis

[0056] like Figure 2 As shown, the residual plot is discrete, without the "large residual", "abnormal residual", "residual point cluster" and "unequal variation of residual" situations on the right side. Therefore, the residuals of the regression model are valid.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications made to the present invention by those skilled in the art without departing from the spirit of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method of controlling NVH gear rattle noise of an electric power drive system of a pure electric vehicle, characterized in that, Includes the following steps: 1) The following multiple linear regression equations are established for the vibration amplitude of the reducer, the slope of the motor torque change, and the gear backlash of the reducer in the electric drive system: ； In the formula, The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant; 2) Vibration data experiments were conducted on the electric drive system using a triaxial acceleration sensor, and several sets of experimental data were collected. The experimental data included the vibration amplitude of the reducer, the slope of the motor torque change, and the backlash of the reducer gears at the same time. 3) Substitute the experimental data collected in step 2) into the multiple linear regression equation in step 1), and solve the multiple linear regression equation for the motor torque change slope, the reducer gear backlash regression coefficient, and the constant using the least squares method. Thus, the multiple linear regression equation with determined regression coefficients is obtained; 4) Set the vibration amplitude threshold of the reducer according to requirements, and substitute it into the multiple linear regression equation determined by the regression coefficients. Then, uniformly modify and debug the reducer and motor of the electric drive system in the following manner: 4-1) First, determine the value of the gear backlash of the reducer according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding motor torque change slope. Control the motor of the electric drive system according to the motor torque change slope. 4-2) First, determine the value of the slope of the motor torque change according to the requirements, and then substitute this value into the multiple linear regression equation determined by the regression coefficient to obtain the corresponding gear backlash of the reducer. Adjust the design parameters of the reducer according to the gear backlash of the reducer.

2. The method of claim 1, wherein, Based on the principle of least squares, the regression coefficients of the slope of the motor torque change in this multiple linear regression equation are obtained. Regression coefficient of gear backlash in reducer and constants The specific steps to obtain the multiple linear regression equation with determined regression coefficients are as follows: ① According to the principle of least squares, the regression coefficients , , should satisfy: ； In the formula, This represents the sum of squares of the deviations between the measured and theoretical values ​​of the reducer vibration amplitude in the vibration data experiment. The measured value of the reducer vibration amplitude at time n. Let be the slope of the motor torque change at time n. The regression coefficient represents the slope of the motor torque change. Let n be the gear backlash of the reducer at time n. The regression coefficient for the gear backlash of the reducer is given. It is a constant. This represents the theoretical value of the reducer vibration amplitude at time n in the vibration data experiment. ② Partial derivatives of the formula in step ① are taken with respect to the regression coefficients , , The following normal equations are obtained by setting the partial derivatives to zero. ； (3) The regression coefficients are obtained by solving the normal equations in step (2) by matrix method , , The numerical values of the regression coefficients are as follows: 。

Images