CN-116911045-B - Speed reducer gear parameter optimization method, device, equipment and storage medium

CN116911045BCN 116911045 BCN116911045 BCN 116911045BCN-116911045-B

Abstract

The invention provides a method, a device, equipment and a storage medium for optimizing gear parameters of a speed reducer, wherein the method comprises the steps of obtaining a sample set of a gear pair in the speed reducer, and gear parameters and modification ranges of the gear parameters of the gear pair; the method comprises the steps of taking a part of a sample set as a training sample, inputting the training sample set into a preset speed reducer simulation model, obtaining first simulation values of the training samples under various working conditions, wherein the first simulation values comprise transmission errors and transmission error peak values, calculating the respective mean value and variance of the transmission errors and the transmission error peak values of the training samples according to the first simulation values of the same training sample under different working conditions, determining the mean value and the variance as second simulation values, carrying out fitting calculation according to the training samples and the second simulation values, determining a proxy model, constructing a gear parameter optimization model based on the proxy model and a modification range, and enabling the gear parameter optimization model to output optimized values of various gear parameters according to the second simulation values. The invention not only improves the NVH performance of the speed reducer, but also improves the universality of gear parameters.

Inventors

YANG JING
CHANG XING
ZHENG WEI
YU FUYONG
ZHAO CHEN

Assignees

深蓝汽车科技有限公司

Dates

Publication Date: 20260505
Application Date: 20230726

Claims (9)

1. A method for optimizing gear parameters of a speed reducer, comprising: Acquiring a sample set of a gear pair in a speed reducer, gear parameters of the gear pair and a modification range of the gear parameters, wherein the sample is a combination of any value of each gear parameter in the modification range; taking a part of samples in the sample set as training samples, inputting the training samples into a preset speed reducer simulation model, and obtaining first simulation values of the training samples under various working conditions, wherein the working conditions are vehicle torque when the speed reducer works, and the first simulation values comprise transmission errors and transmission error peak-to-peak values; According to the first simulation values of the same training sample under different working conditions, respectively calculating the mean value and variance of the transmission error corresponding to the training sample and the transmission error peak-to-peak value, and determining the mean value and variance as a second simulation value; performing fitting calculation according to the training sample and the second simulation value to determine a proxy model, wherein the proxy model is divided into a first proxy model, a second proxy model, a third proxy model and a fourth proxy model; The gear parameter optimization model is constructed based on the agent model and the modification range, the gear parameter optimization model is made to output optimized values of the gear parameters according to the second simulation value, the transmission errors and the transmission error peak values output by the first agent model and the second agent model are weighted and calculated respectively to determine target functions of the transmission errors, the transmission errors and the transmission error peak values output by the third agent model and the fourth agent model are weighted and calculated respectively to determine target functions of the transmission error peak values, the gear parameter optimization model is constructed according to the target functions of the transmission errors, the target functions of the transmission error peak values, the modification range of the gear parameters and constraint conditions that the transmission error peak values are within a preset range threshold, initial values of the gear parameters are obtained and input to the gear parameter optimization model to obtain an optimization set of the gear parameters, the optimization set is formed by any combination of the optimization values of the gear parameters within the modification range, and the optimization set of the gear parameters is determined.
2. The method for optimizing parameters of a gear of a speed reducer according to claim 1, wherein the step of inputting a part of samples in the sample set as training samples into a preset simulation model of the speed reducer to obtain first simulation values of the training samples under each working condition comprises the steps of: Constructing an inner surface and an outer surface of a field mouth according to the training sample and the working condition, wherein the training sample forms the inner surface and the outer surface of the field mouth, and the working condition forms the outer surface and the inner surface of the field mouth; and inputting each training sample and each working condition into the speed reducer simulation model based on the field mouth inner and outer surfaces so that the speed reducer simulation model outputs the first simulation value of each training sample in each working condition.
3. The method of optimizing the parameters of the speed reducer gear according to claim 1, wherein the second simulation value includes a mean value of the transmission errors output by the first proxy model, a variance of the transmission errors output by the second proxy model, a mean value of the transmission error peaks output by the third proxy model, and a variance of the transmission error peaks output by the fourth proxy model; and performing fitting calculation according to the training sample and the second simulation value to determine a proxy model, wherein the method comprises the following steps: fitting the training sample with the mean value and the variance of the transfer error respectively to determine the first proxy model and the second proxy model; fitting the training sample with the mean value and the variance of the peak-to-peak value of the transmission error respectively, and determining the third generation model and the fourth agent model.
4. A method of optimizing a gear parameter of a retarder according to claim 3, comprising, after said fitting calculation based on said training samples and said second simulation values, determining a proxy model: Inputting the rest samples of the sample set as verification samples to the speed reducer simulation model and each agent model, and respectively outputting the first simulation value and the second simulation value of each verification sample, wherein the first simulation value of each verification sample comprises a first transmission error and a first transmission error peak-to-peak value, and the second simulation value of each verification sample comprises a mean value and a variance of a second transmission error and a mean value and a variance of a second transmission error peak-to-peak value; Calculating average relative errors of the agent models according to the first simulation values and the second simulation values of the verification samples; Wherein the mean and variance of the first transfer error and the mean and variance of the first transfer error peak-to-peak value of each of the verification samples are calculated; Average calculation is carried out on the relative error between the average value of the first transmission error and the average value of the second transmission error corresponding to each verification sample, and the average relative error of the first proxy model is determined; average calculation is carried out on the relative error between the variance of the first transmission error and the variance of the second transmission error corresponding to each verification sample, and the average relative error of the second agent model is determined; Average calculation is carried out on the relative error between the average value of the first transmission error and the average value of the second transmission error corresponding to each verification sample, and the average relative error of the third generation model is determined; And carrying out average calculation on the relative error between the variance of the first transmission error and the variance of the second transmission error corresponding to each verification sample, and determining the average relative error of the fourth agent model.
5. The method of optimizing parameters of a reduction gear as set forth in claim 4, including, after calculating an average relative error of each of said proxy models from said first simulation value and said second simulation value of each of said validation samples: comparing the average relative error of each agent model with a preset error threshold value respectively; If the average relative error of the proxy model is larger than the preset error threshold, adding the training sample, and reconstructing the proxy model; and if the average relative error of the proxy model is smaller than or equal to the preset error threshold, constructing the gear parameter optimization model.
6. The method of optimizing gear parameters of a reduction gear according to claim 1, comprising, after said constructing a gear parameter optimizing model based on said proxy model and said modification range, causing said gear parameter optimizing model to output optimized values of each of said gear parameters according to said second simulation values: respectively inputting the optimized value and the initial value of each gear parameter into the speed reducer simulation model to output the transmission error corresponding to the optimized value and the transmission error corresponding to the initial value under each working condition through the speed reducer simulation model; Comparing the transmission error corresponding to the optimized value and the transmission error corresponding to the initial value under each working condition, and determining an optimized result of the gear parameter based on a comparison result; if the optimization result does not meet the preset expectations, the training sample is added, and the gear parameter optimization model is reconstructed.
7. A speed reducer gear parameter optimization device, comprising: The sample module is used for acquiring a sample set of a gear pair in the speed reducer, gear parameters of the gear pair and a modification range of the gear parameters, wherein the sample is a combination of any value of each gear parameter in the modification range; The first simulation module is used for taking a part of samples in the sample set as training samples, inputting the training samples into a preset speed reducer simulation model, and obtaining first simulation values of the training samples under various working conditions, wherein the working conditions are vehicle torque when the speed reducer works, and the first simulation values comprise transmission errors and transmission error peak-to-peak values; The second simulation module is used for respectively calculating the mean value and the variance of the transmission error corresponding to the training sample and the peak-to-peak value of the transmission error according to the first simulation values of the same training sample under different working conditions, and determining the transmission error and the peak-to-peak value as second simulation values; the fitting module is used for carrying out fitting calculation according to the training sample and the second simulation value to determine a proxy model, and the proxy model is divided into a first proxy model, a second proxy model, a third proxy model and a fourth proxy model; The optimization module is used for constructing a gear parameter optimization model based on the proxy model and the modification range, enabling the gear parameter optimization model to output optimized values of gear parameters according to the second simulation value, carrying out weighted sum calculation on the transmission errors and the transmission error peaks and peaks respectively output by the first proxy model and the second proxy model, determining an objective function of the transmission errors, carrying out weighted sum calculation on the transmission errors and the transmission error peaks respectively output by the third proxy model and the fourth proxy model, determining an objective function of the transmission error peaks and peaks, constructing the gear parameter optimization model according to the objective function of the transmission errors, the objective function of the transmission error peaks and peaks, the modification range of each gear parameter and constraint conditions, wherein the constraint conditions are that the transmission error peaks and the transmission error peaks are within a preset range threshold, obtaining initial values of each gear parameter, inputting the initial values into the gear parameter optimization model, obtaining an optimization set of the gear parameter, carrying out weighted sum calculation on the transmission errors and the transfer error peaks and the objective function of the transmission error peaks and the transmission error peaks, constructing the gear parameter optimization model according to the objective function of the transmission errors, carrying out optimization set selection of any optimization value combination of the gear parameter optimization value in the modification range, and determining the optimization set.
8. An electronic device, comprising: One or more processors; Storage means for storing one or more programs that, when executed by the one or more processors, cause the electronic device to implement the method of optimizing the retarder gear parameter as claimed in any of claims 1 to 6.
9. A computer-readable storage medium, characterized in that a computer program for causing a computer to execute the speed reducer gear parameter optimization method according to any one of claims 1 to 6 is stored thereon.

Description

Speed reducer gear parameter optimization method, device, equipment and storage medium Technical Field The invention relates to the technical field of gear optimization, in particular to a method, a device, equipment and a storage medium for optimizing parameters of a gear of a speed reducer. Background With the rapid development of the automobile industry, the power transmission system of the vehicle is increasingly promoted to be high-speed, light-weight and electric, and the speed reducer is taken as a main component in the power transmission system of the vehicle, and mainly plays roles of reducing the rotating speed and increasing the torque in the power transmission process. The NVH performance problem of the speed reducer is mainly caused by transmission errors of gear pairs of the speed reducer, the transmission errors are important parameters for evaluating the meshing quality of gears, when fluctuation of the transmission errors is reduced, vibration and noise during gear meshing can be reduced, and the squeak noise of the speed reducer can be effectively reduced, so that the transmission errors of the speed reducer are reduced by adopting a mode of optimizing gear parameters and carrying out microscopic shaping on gears to improve the NVH performance of the speed reducer, however, when the transmission errors of the speed reducer are optimized, on one hand, the same speed reducer cannot take into account multiple types of different working conditions, so that the NVH performance of the speed reducer on different types of vehicles is inconsistent, and the gear parameters have no universality; on the other hand, the gear parameters involved in the microscopic modification of the gear have more variables, so that the dimension of the seeking process is higher, and the optimization efficiency is low. Chinese patent CN115481499a discloses a gear transmission error optimization method and system, by introducing samples of gear parameters into a gear transmission error calculation model, determining and reducing transmission errors according to the feature parameters to be optimized. However, although the scheme completes optimization of the transmission error, how the characteristic parameters to be optimized are suitable for various working conditions is not described, and the universality of the parameters is poor. Chinese patent CN106763642B provides a noise reduction method for an electric vehicle reducer and an electric vehicle reducer, in which microscopic shaping parameters are determined by transmission errors occurring during gear processing, so as to perform gear shaping on the reducer according to the microscopic shaping parameters. However, the efficiency of the scheme is low, and the gear shaping does not consider the factor of working conditions, so that the NVH performance of the speed reducer under various working conditions cannot be considered. Therefore, how to optimize the gear parameters of the speed reducer to improve the universality of the gear parameters is a problem to be solved urgently at present. Disclosure of Invention In view of the above-mentioned drawbacks of the prior art, the present invention provides a method, apparatus, device and storage medium for optimizing parameters of a gear of a speed reducer, so as to solve at least one of the above-mentioned technical problems. In a first aspect, the invention provides a gear parameter optimization method of a speed reducer, which comprises the steps of obtaining a sample set of a gear pair in the speed reducer, and gear parameters of the gear pair and a modification range of the gear parameters, wherein the sample is a combination of any value of each gear parameter in the modification range, inputting a part of samples in the sample set as training samples into a preset speed reducer simulation model, obtaining first simulation values of each training sample under each working condition, wherein the working condition is vehicle torque when the speed reducer works, the first simulation values comprise transmission errors and transmission error peak values, calculating respective mean values and variances of the transmission errors and the transmission error peak values corresponding to the training samples according to the first simulation values of the same training sample under different working conditions respectively, determining a second simulation value, calculating according to the training samples and the second simulation values, determining a proxy model, and constructing a gear parameter optimization model based on the proxy model and the modification range, and enabling the gear parameters to be optimized according to the second simulation model, and outputting optimized gear parameter values. In an embodiment of the invention, the step of inputting a part of samples in the sample set as training samples into a preset speed reducer simulation model to obtain first simulation values of the training samples unde