Reduction gear shifting working condition bearing force prediction method, device, equipment and medium

Abstract

This application relates to a method, apparatus, equipment, and medium for predicting bearing forces in variable speed operation of a speed reducer. The method includes: determining the corresponding gear shaft transmission geometric model and material parameters of each component of the electric vehicle speed reducer; constructing a dynamic simulation model of the speed reducer based on the gear shaft transmission geometric model and material parameters; calculating and determining the bearing force data and corresponding steady-state speeds of the speed reducer under multiple steady-state operating conditions based on the dynamic simulation model; determining the bearing force of the electric vehicle speed reducer under variable speed operation based on the bearing force data and corresponding steady-state speeds under the steady-state operating conditions using the bearing force prediction model, wherein the bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module; and determining the noise of the electric vehicle speed reducer under variable speed operation based on the bearing force calculation under the variable speed operation. This application can quickly and accurately obtain the speed reducer bearing force that varies with the input speed.

Description

Technical Field

[0001] This application relates to the field of electric vehicles, and more particularly to a method for predicting bearing force under gearbox shifting conditions, a corresponding device, electronic equipment, and a computer-readable storage medium. Background Technology

[0002] The current mainstream trend in the automotive industry is the vigorous development of new energy vehicles, with pure electric vehicles being the mainstream product in this field. As pure electric vehicle technology advances, noise issues are increasingly attracting attention. Unlike traditional vehicles, the noise of pure electric vehicles primarily originates from the electric drive assembly, especially the reducer connected to the motor. Due to the high speed of the motor, it often produces whistling noise, reducing the comfort of the driver and passengers. How to quickly predict and optimize reducer noise during the design phase has become a common concern and urgent issue for automakers. The noise in the electric vehicle reducer is mainly generated by the power output from the motor being transmitted to the gear transmission system. Gear vibration is transmitted through the shaft system to the bearings, and then the noise is radiated by the housing connected to the bearing housing. Therefore, accurately predicting the bearing force of the reducer under transmission conditions is crucial for calculating its noise.

[0003] Currently, the prediction and calculation of bearing forces in electric vehicle reducers mostly employ analytical methods or the finite element method. The analytical method involves establishing a mathematical model of the reducer's transmission system—gear, shaft, and bearing—and solving the dynamic equations of this model to obtain the bearing force. The finite element method uses commercial finite element software to calculate the bearing force by importing the model, meshing, setting material and boundary conditions, and simulating actual operating conditions to ultimately obtain the bearing force. For obtaining the bearing force of an electric vehicle reducer under variable speed conditions, an interpolation function is used to obtain the bearing force that varies with speed, i.e., the bearing force under variable speed conditions, based on the obtained bearing forces under multiple steady-state conditions.

[0004] Analytical methods, which use mathematical models to analyze bearing forces, consider too few influencing factors, resulting in significant calculation errors and limiting their application to single-condition calculations. The finite element method (FEM) uses commercial FEM software for simulation calculations of bearing forces, but its effectiveness is also affected by modeling accuracy, mesh generation, and computer performance. Furthermore, most commercial FEM software currently only supports single-speed conditions. To predict bearing forces under variable-speed conditions, interpolation methods using interpolation functions are necessary in addition to the above methods. This repetitive calculation is labor-intensive, time-consuming, and inefficient. For example, to calculate reducer noise during acceleration from 2000-5000 r / min, an interpolation node can be selected at intervals of 200 r / min based on the housing's natural frequency. The obtained bearing forces under steady-state conditions corresponding to the reducer can then be used to calculate the bearing force as a function of speed using higher-order interpolation.

[0005] In summary, existing technologies using interpolation to calculate bearing forces under variable speed conditions suffer from problems such as large computational workload, long computation time, and low efficiency. Furthermore, analytical methods using mathematical models to analyze bearing forces consider too few influencing factors, resulting in significant calculation errors. Therefore, the applicant has made corresponding explorations to address these issues. Summary of the Invention

[0006] The purpose of this application is to solve the above-mentioned problems by providing a method for predicting bearing force under variable speed conditions of a speed reducer, a corresponding device, electronic equipment, and a computer-readable storage medium.

[0007] To achieve the various objectives of this application, the following technical solution is adopted:

[0008] A method for predicting bearing force under variable speed operation of a speed reducer, proposed to meet one of the purposes of this application, includes:

[0009] In response to the bearing force prediction command of the electric vehicle reducer under the variable speed condition, determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer.

[0010] Based on the gear shaft transmission geometric model and the material parameters, a corresponding dynamic simulation model of the reducer is constructed. Based on the dynamic simulation model, the bearing force data of the reducer under multiple steady-state conditions and its corresponding steady-state speed are calculated and determined.

[0011] The pre-trained bearing force prediction model for variable speed operating conditions determines the bearing force of the electric vehicle reducer under variable speed operating conditions based on the bearing force data under the steady-state operating conditions and its corresponding steady-state operating speed. The variable speed operating condition bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module.

[0012] The noise of the electric vehicle reducer under variable speed conditions is determined based on the bearing force calculation under variable speed conditions, so as to complete the bearing force prediction of the electric vehicle reducer under variable speed conditions.

[0013] Optional steps for constructing a dynamic simulation model include:

[0014] Determine the initial data corresponding to the reducer housing, gears, shafts, and bearings of the electric vehicle reducer, and construct the corresponding gear and shaft transmission geometric model based on the initial data corresponding to the reducer housing, gears, shafts, and bearings. The various parts of the reducer include one or any combination of the reducer housing, gears, shafts, and bearings.

[0015] The material parameters corresponding to the reducer housing, gears, shafts, and bearings, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model are determined. The material parameters corresponding to the reducer housing, gears, shafts, and bearings, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model are then input into the gear shaft transmission geometric model to construct a dynamic simulation model.

[0016] Optionally, the step of calculating and determining the bearing force data of the reducer and its corresponding steady-state speed under multiple steady-state operating conditions based on the dynamic simulation model includes:

[0017] The excitation of bearing force caused by the reducer is mainly driven by the helical gear transmission error and the time-varying meshing stiffness of the helical gear pair. Simulation analysis is performed based on the aforementioned dynamic simulation model, where the helical gear transmission error is expressed as:

[0018]

[0019] Where TE is the helical gear transmission error, θ1 and θ2 are the theoretical rotation angles of the driving and driven gears, θ2'2 is the actual rotation angle of the driven gear, and R... b1 R b2 Let be the base circle radius of the driving gear and the driven gear, and i be the theoretical transmission ratio;

[0020] The time-varying meshing stiffness of the helical gear pair:

[0021]

[0022] Where K(t) is the time-varying meshing stiffness of the helical gear pair, ω m The meshing frequency, Let ε be the initial phase angle. α For the degree of overlap, K c This refers to the meshing stiffness of a single gear pair;

[0023] The calculated helical gear transmission error and the time-varying meshing stiffness of the helical gear pair are applied to the dynamic simulation model to obtain bearing force data under multiple steady-state operating conditions.

[0024] The bearing force data obtained from the simulation under steady-state operating conditions, along with their corresponding steady-state operating speed and operating time, form the sample data set for the bearing force prediction model under variable speed conditions.

[0025] Optionally, after the step of combining the bearing force data obtained from the simulation under steady-state operating conditions with its corresponding steady-state operating speed and operating time to form a sample data set for the bearing force prediction model under variable speed conditions, the following steps are included:

[0026] In response to the instruction to construct a tag data set, the actual bearing force of the bearing force sensor in the electric vehicle reducer and the corresponding actual steady-state speed are obtained under steady-state conditions.

[0027] The actual bearing force, the actual steady-state operating speed, and the operating time are used to construct a labeled data set for the bearing force prediction model under variable speed conditions.

[0028] Optionally, the steps for training the bearing force prediction model under variable speed conditions include:

[0029] Based on the characteristics of bearing force change over time under steady-state conditions in the sample dataset, a loop unit is selected, and the loss function is determined according to the task type of bearing force prediction to construct the bearing force prediction model for variable speed conditions.

[0030] The data in the sample data set and the label data set are cut into data segments of the same length. These data segments are subjected to discrete Fourier transform, and the transformed data are normalized to have a variance of 1 and a mean of 0.

[0031] A portion of the normalized sample data set is used as the training set, and the training set is input into the variable speed bearing force prediction model for training.

[0032] When the variable speed bearing force prediction model reaches the preset number of iterations, the training of the variable speed bearing force prediction model is completed.

[0033] Optionally, the step of determining the bearing force of the electric vehicle reducer under variable speed conditions based on the pre-trained bearing force prediction model under variable speed conditions and its corresponding steady-state speed under the bearing force data under the steady-state conditions includes:

[0034] The bearing force data under steady-state conditions and their corresponding steady-state speeds are input into the pre-trained bearing force prediction model for variable speed conditions.

[0035] The frequency domain transformation module in the variable speed bearing force prediction model overlaps and truncates the bearing force data in the time domain output by the adaptive neural network in the variable speed bearing force prediction model.

[0036] The truncated bearing force data is subjected to Discrete Fourier Transform, and the data after Discrete Fourier Transform is then subjected to frequency domain windowed averaging.

[0037] Output the bearing force data of the electric vehicle reducer under variable speed conditions in the time and frequency domains.

[0038] Optionally, the adaptive neural network in the variable speed bearing force prediction model includes an input layer, a long short-time memory layer, several fully connected layers, and a data output layer.

[0039] A bearing force prediction device for speed reducer shifting conditions, provided for another purpose of this application, includes:

[0040] The material parameter acquisition module is configured to respond to the bearing force prediction command of the electric vehicle reducer under the variable speed condition, and determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer.

[0041] The steady-state bearing force determination module is configured to construct a dynamic simulation model of the reducer based on the gear shaft transmission geometric model and the material parameters, and calculate and determine the bearing force data of the reducer under multiple steady-state conditions and its corresponding steady-state speed based on the dynamic simulation model.

[0042] The variable speed bearing force determination module is configured to determine the bearing force of the electric vehicle reducer under variable speed conditions based on the bearing force data under the steady-state conditions and its corresponding steady-state speed, according to a pre-trained variable speed bearing force prediction model. The variable speed bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module.

[0043] The reducer noise calculation module is configured to determine the noise of the electric vehicle reducer under the variable speed condition based on the bearing force calculation under the variable speed condition, so as to complete the bearing force prediction of the electric vehicle reducer under the variable speed condition.

[0044] An electronic device provided for another purpose of this application includes a central processing unit and a memory, the central processing unit being configured to invoke and run a computer program stored in the memory to perform the steps of the method for predicting bearing force under speed change conditions of a reducer as described in this application.

[0045] A computer-readable storage medium is provided for another purpose of this application, which stores, in the form of computer-readable instructions, a computer program implemented according to the method for predicting bearing force under the speed reduction condition of the reducer, which, when called by a computer, executes the steps included in the corresponding method.

[0046] Compared to existing technologies, this application addresses the problems of existing technologies using interpolation methods to calculate bearing forces under variable speed conditions, which suffer from large computational workload, long calculation time, and low efficiency, as well as the problems of analytical methods using mathematical models to analyze bearing forces but considering too few influencing factors, resulting in large calculation errors. This application provides, but is not limited to, the following beneficial effects:

[0047] Firstly, this application addresses the problems of large errors in obtaining the bearing force under steady-state conditions, long calculation time, and large amount of repetitive work in the current electric vehicle reducer. It proposes a method for predicting the bearing force under the variable speed conditions of electric vehicle reducers. This method can quickly and accurately obtain the reducer bearing force that varies with the input speed. Furthermore, the obtained bearing force data can be used to predict the noise of the electric vehicle reducer under the variable speed conditions, which greatly saves time and costs.

[0048] Secondly, the bearing force prediction model for variable speed operation of this application has the advantages of fast calculation speed and high computational efficiency. It can quickly and accurately obtain the reducer bearing force that changes with the input speed. Furthermore, the obtained bearing force data can be used to predict the noise of electric vehicle reducer variable speed operation. Attached Figure Description

[0049] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0050] Figure 1 This is a flowchart illustrating the bearing force prediction method for the speed reducer under variable speed conditions in the embodiments of this application.

[0051] Figure 2 This is a schematic diagram of the dynamic simulation model in the embodiments of this application;

[0052] Figure 3 This is a schematic diagram illustrating the meshing transmission error in an embodiment of this application;

[0053] Figure 4 This is a schematic diagram of the time-varying meshing stiffness of the helical gear pair in the embodiments of this application;

[0054] Figure 5 This is an exemplary architecture diagram of the adaptive neural network in the embodiments of this application;

[0055] Figure 6 This is a schematic diagram of the bearing force prediction device for the speed reducer in the variable speed operation condition in the embodiments of this application;

[0056] Figure 7 This is a schematic diagram of the structure of the computer device in the embodiments of this application. Detailed Implementation

[0057] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0058] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0059] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0060] Those skilled in the art will understand that the terms "client," "terminal," and "terminal device" as used herein include both devices that receive wireless signals, devices that only possess wireless signal receiver capabilities without transmission capabilities, and devices with receiving and transmitting hardware, devices that have receiving and transmitting hardware capable of bidirectional communication over a bidirectional communication link. Such devices may include: cellular or other communication devices such as personal computers or tablets, having single-line displays, multi-line displays, or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service) that can combine voice, data processing, fax, and / or data communication capabilities; PDA (Personal Digital Assistant) that may include a radio frequency receiver, pager, internet / intranet access, web browser, notepad, calendar, and / or GPS (Global Positioning System) receiver; and conventional laptops and / or handheld computers or other devices that have and / or include radio frequency receivers. As used herein, "client," "terminal," and "terminal device" can be portable, transportable, installed in a means of transportation (air, sea, and / or land), or suitable and / or configured to operate locally and / or in a distributed manner, operating in any other location on Earth and / or in space. "Client," "terminal," and "terminal device" as used herein can also be a communication terminal, an internet access terminal, or a music / video playback terminal, such as a PDA, a MID (Mobile Internet Device), and / or a mobile phone with music / video playback capabilities, or a smart TV, set-top box, etc.

[0061] The hardware referred to by the names "server," "client," and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann architecture, such as a central processing unit (including an arithmetic logic unit and a control unit), memory, input devices, and output devices. The computer program is stored in its memory, and the central processing unit loads the program stored in the secondary storage into the main memory to run it, execute the instructions in the program, and interact with the input and output devices to complete specific functions.

[0062] It should be noted that the concept of "server" used in this application can also be extended to the case of server clusters. Based on the network deployment principles understood by those skilled in the art, the servers should be logically divided. Physically, these servers can be independent of each other but accessible through interfaces, or they can be integrated into a single physical computer or a computer cluster. Those skilled in the art should understand this flexibility and should not use it to constrain the implementation of the network deployment method in this application.

[0063] One or more of the technical features of this application, unless explicitly specified herein, can be deployed on a server and accessed by a client remotely calling the online service interface provided by the server, or can be directly deployed and run on a client for access.

[0064] Unless otherwise specified, the neural network models referenced or potentially referenced in this application may be deployed on a remote server and invoked remotely on the client, or deployed on a client with the capability to invoke directly. In some embodiments, when running on the client, the corresponding intelligence may be acquired through transfer learning in order to reduce the requirements on the client's hardware resources and avoid excessive consumption of the client's hardware resources.

[0065] Unless otherwise specified, all data involved in this application may be stored remotely on a server or on a local terminal device, as long as it is suitable for use by the technical solution of this application.

[0066] Those skilled in the art will understand that although the various methods in this application are described based on the same concept and thus present commonality among them, they can be performed independently unless otherwise specified. Similarly, the various embodiments disclosed in this application are all based on the same inventive concept; therefore, concepts expressed in the same way, as well as concepts that are appropriately changed for convenience but are expressed differently, should be understood equivalently.

[0067] Unless otherwise expressly stated, the various embodiments disclosed in this application can be combined in a cross-cutting manner to flexibly construct new embodiments, as long as such combination does not depart from the inventive spirit of this application and can meet the needs of the prior art or solve a certain deficiency in the prior art. Those skilled in the art should be aware of such modifications.

[0068] The current mainstream trend in the automotive industry is the vigorous development of new energy vehicles, with pure electric vehicles being the mainstream product in this field. As pure electric vehicle technology advances, noise issues are increasingly attracting attention. Unlike traditional vehicles, the noise of pure electric vehicles primarily originates from the electric drive assembly, especially the reducer connected to the motor. Due to the high speed of the motor, it often produces whistling noise, reducing the comfort of the driver and passengers. How to quickly predict and optimize reducer noise during the design phase has become a common concern and urgent issue for automakers. The noise in the electric vehicle reducer is mainly generated by the power output from the motor being transmitted to the gear transmission system. Gear vibration is transmitted through the shaft system to the bearings, and then the noise is radiated by the housing connected to the bearing housing. Therefore, accurately predicting the bearing force of the reducer under transmission conditions is crucial for calculating its noise.

[0069] Based on the above exemplary scenarios, please refer to Figure 1 In one embodiment of the method for predicting bearing force under gearbox transmission conditions, this application includes:

[0070] Step S10: Respond to the bearing force prediction command of the electric vehicle reducer under the speed change condition, and determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer.

[0071] The computer terminal device can respond to the bearing force prediction command of the electric vehicle reducer under the variable speed condition, determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer. Each part of the reducer includes one or more of the reducer housing, gears, shafts and bearings. First, the initial data corresponding to the reducer housing, gears, shafts and bearings of each part of the electric vehicle reducer are obtained. Based on the initial data corresponding to the reducer housing, gears, shafts and bearings, the corresponding gear shaft transmission geometric model is constructed.

[0072] In some embodiments, a simplified gear and shaft transmission geometric model is established based on the initial data of the electric vehicle reducer housing, gears, shafts, and bearings. For example, when establishing a differential model, it is simplified to model an irregular shaft.

[0073] Step S20: Based on the gear shaft transmission geometric model and the material parameters, construct the corresponding dynamic simulation model of the reducer, and calculate and determine the bearing force data and the corresponding steady-state speed of the reducer under multiple steady-state conditions based on the dynamic simulation model.

[0074] After determining the gear shaft transmission geometric model corresponding to the electric vehicle reducer and the material parameters corresponding to each part of the reducer, a dynamic simulation model corresponding to the reducer is constructed based on the gear shaft transmission geometric model and the material parameters. Based on the dynamic simulation model, the bearing force data of the reducer under multiple steady-state conditions and its corresponding steady-state speed are calculated and determined.

[0075] The steps for constructing a dynamic simulation model include:

[0076] Step S201: Determine the initial data corresponding to the reducer housing, gears, shafts and bearings of the electric vehicle reducer, and construct the corresponding gear and shaft transmission geometric model based on the initial data corresponding to the reducer housing, gears, shafts and bearings.

[0077] Step S203: Determine the material parameters corresponding to the reducer housing, gear, shaft, and bearing, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model, and input the material parameters corresponding to the reducer housing, gear, shaft, and bearing, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model into the gear shaft transmission geometric model to construct a dynamic simulation model.

[0078] Specifically, the initial data corresponding to the reducer housing, gears, shafts, and bearings of the electric vehicle reducer are determined. Based on the initial data of the reducer housing, gears, shafts, and bearings, a corresponding gear-shaft transmission geometric model is constructed. The material parameters of the reducer housing, gears, shafts, and bearings, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear-shaft transmission geometric model are determined. The material parameters of the reducer housing, gears, shafts, and bearings, the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear-shaft transmission geometric model are input into the gear-shaft transmission geometric model to construct a dynamic simulation model. The dynamic simulation model is as follows: Figure 2 As shown.

[0079] Furthermore, the step of calculating and determining the bearing force data of the reducer under multiple steady-state operating conditions and its corresponding steady-state operating speed based on the dynamic simulation model includes:

[0080] Step S100: The excitation of bearing force caused by the reducer is mainly based on the helical gear transmission error and the time-varying meshing stiffness of the helical gear pair. Simulation analysis is performed based on the aforementioned dynamic simulation model, wherein the helical gear transmission error is expressed as:

[0081]

[0082] Where TE is the helical gear transmission error, θ1 and θ2 are the theoretical rotation angles of the driving and driven gears, θ′2 is the actual rotation angle of the driven gear, and R... b1 R b2 Let be the base circle radius of the driving gear and the driven gear, and i be the theoretical transmission ratio;

[0083] The time-varying meshing stiffness of the helical gear pair:

[0084]

[0085] Where K(t) is the time-varying meshing stiffness of the helical gear pair, ω m The meshing frequency, Let ε be the initial phase angle. α For the degree of overlap, K c This refers to the meshing stiffness of a single gear pair;

[0086] Step S200: The calculated helical gear transmission error and the time-varying meshing stiffness of the helical gear pair are applied to the dynamic simulation model to obtain bearing force data under multiple steady-state operating conditions.

[0087] Step S300: The bearing force data obtained from the simulation under steady-state operating conditions and its corresponding steady-state operating speed and running time are combined to form a sample data set for the bearing force prediction model under the variable speed condition. The running time and steady-state operating speed must be correctly matched with their corresponding bearing force data.

[0088] In some embodiments, such as the meshing transmission error of the first-stage gear pair of the reducer and the time-varying meshing stiffness of the helical gear pair, Figure 3 , Figure 4 As shown.

[0089] Further, after the step of combining the bearing force data obtained from the simulation under steady-state operating conditions with its corresponding steady-state operating speed and operating time to form a sample data set for the bearing force prediction model under variable speed conditions, the following steps are included:

[0090] Step S2001: Respond to the tag data set construction instruction to obtain the actual bearing force of the bearing force sensor in the electric vehicle reducer under steady-state conditions and the corresponding actual steady-state speed.

[0091] Step S2003: Construct a tag data set for the bearing force prediction model under variable speed conditions by combining the actual bearing force, the actual steady-state operating speed, and the running time.

[0092] In some embodiments, such as the left bearing 6207 radial ball bearing of the reducer input shaft, the BROSA bearing force sensor can be arranged at the bearing housing, the electric drive assembly can be installed on the test bench, and the data can be collected and conditioned by the Sony or AVL data acquisition system. Finally, the bearing force time domain data can be processed and displayed on the computer.

[0093] Step S30: Based on the pre-trained bearing force prediction model for variable speed operating conditions, the bearing force of the electric vehicle reducer under variable speed operating conditions is determined according to the bearing force data under the steady-state operating conditions and its corresponding steady-state operating speed. The bearing force prediction model for variable speed operating conditions is constructed by an adaptive neural network followed by a frequency domain transformation module.

[0094] After calculating and determining the bearing force data and corresponding steady-state speed of the reducer under multiple steady-state operating conditions based on the dynamic simulation model, the bearing force of the electric vehicle reducer under variable speed conditions is determined based on the bearing force data and corresponding steady-state speed under the steady-state operating conditions using a pre-trained variable speed bearing force prediction model. The variable speed bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module.

[0095] Further steps in training the bearing force prediction model for variable speed operating conditions include:

[0096] Step S301: Select a cyclic unit based on the characteristics of bearing force change over time under steady-state conditions in the sample data set, and determine the loss function according to the task type of bearing force prediction to construct the bearing force prediction model for variable speed conditions.

[0097] Step S303: Cut each data item in the sample data set and the label data set into data segments of the same length, perform discrete Fourier transform on these data segments, and normalize the transformed data to have a variance of 1 and a mean of 0.

[0098] Step S305: Take a portion of the normalized sample data set as the training set and input the training set into the variable speed bearing force prediction model for training.

[0099] Step S307: When the variable speed bearing force prediction model reaches the preset number of iterations, the training of the variable speed bearing force prediction model is completed.

[0100] Furthermore, the step of determining the bearing force of the electric vehicle reducer under variable speed conditions based on the pre-trained bearing force prediction model under variable speed conditions and the bearing force data under the steady-state conditions and their corresponding steady-state speeds includes:

[0101] Step S3001: Input the bearing force data under steady-state conditions and its corresponding steady-state speed into the pre-trained bearing force prediction model under variable speed conditions.

[0102] Step S3003: The frequency domain transformation module in the variable speed bearing force prediction model overlaps and truncates the bearing force data in the time domain output by the adaptive neural network in the variable speed bearing force prediction model.

[0103] Step S3005: Perform Discrete Fourier Transform on the truncated bearing force data, and then perform frequency domain windowed averaging on the data after Discrete Fourier Transform.

[0104] Step S3007: Output the bearing force data of the electric vehicle reducer under the variable speed condition in the time domain and frequency domain.

[0105] Specifically, based on the characteristics of the bearing dynamic force changing over time under steady-state conditions in the sample dataset, a recurrent unit is selected, and a loss function is chosen according to the task type of bearing force prediction. This forms the adaptive neural network in the variable-speed bearing force prediction model. Long Short-Time Memory (LSTM) units can be selected as the recurrent units, and loss functions such as Mean Absolute Error (MAE) or Mean Square Error (MSE) can be chosen. (See [link to relevant documentation]). Figure 5 The constructed adaptive neural network includes an input layer, a long short-term memory layer, several fully connected layers, and a data output layer.

[0106] Furthermore, the label dataset and the sample dataset are checked for missing or outlier values ​​to ensure the accuracy of the datasets.

[0107] Furthermore, each data item in the sample dataset and the label dataset is cut into data segments of the same length; these data segments are subjected to discrete Fourier transform, and the transformed data are normalized to have a variance of 1 and a mean of 0.

[0108] Furthermore, a portion of the normalized sample data set is used as the training set and input into the adaptive calibration recurrent neural network to train the network. 70% of the sample data set is used as the training set, while 70% of the data from the label data set is selected as the validation set to test the prediction accuracy of the neural network model. The timestamps and rotational speeds of the 70% of data taken from the label data set are consistent with those of the 70% of data in the sample data set. The remaining 30% of the data set, along with the sample data set and the label data set, is used as the test set to test the model's ability to validate under unknown operating conditions.

[0109] In some embodiments, the model prediction accuracy of the neural network is evaluated by tracking the model's loss (e.g., mean squared error) on both the training and validation sets, ensuring that both are decreasing. When the validation set loss stops decreasing or begins to increase, it may indicate model overfitting, and training should be stopped.

[0110] The adaptive neural network in the variable speed bearing force prediction model is followed by the frequency domain transformation module in the same model, enabling prediction of bearing force under variable speed conditions in the reducer. The frequency domain transformation module truncates the time-domain bearing force data output from the neural network, performs a discrete Fourier transform on the truncated data, and finally performs a frequency-domain windowed average on the transformed data. This allows the model to simultaneously output predicted bearing force data in both the time and frequency domains, enabling prediction of bearing force under variable speed conditions in electric vehicle reducers. The adaptive neural network in the variable speed bearing force prediction model includes an input layer, a long short-time memory layer, several fully connected layers, and a data output layer.

[0111] Step S40: Determine the noise of the electric vehicle reducer under the variable speed condition based on the bearing force calculation under the variable speed condition, so as to complete the bearing force prediction of the electric vehicle reducer under the variable speed condition.

[0112] After determining the bearing force of the electric vehicle reducer under variable speed conditions, the noise of the electric vehicle reducer under variable speed conditions is calculated based on the bearing force under variable speed conditions, so as to complete the prediction of the bearing force of the electric vehicle reducer under variable speed conditions.

[0113] As can be seen from the above embodiments, compared with the prior art, this application addresses the problems of large computational workload, long computation time, and low efficiency in the prior art of using interpolation method to calculate bearing force under variable speed conditions, as well as the problems of insufficient consideration of influencing factors and large calculation errors in analytical method using mathematical model to analyze bearing force. This application has, but is not limited to, the following beneficial effects:

[0114] Firstly, this application addresses the problems of large errors in obtaining the bearing force under steady-state conditions, long calculation time, and large amount of repetitive work in the current electric vehicle reducer. It proposes a method for predicting the bearing force under the variable speed conditions of electric vehicle reducers. This method can quickly and accurately obtain the reducer bearing force that varies with the input speed. Furthermore, the obtained bearing force data can be used to predict the noise of the electric vehicle reducer under the variable speed conditions, which greatly saves time and costs.

[0115] Secondly, the bearing force prediction model for variable speed operation of this application has the advantages of fast calculation speed and high computational efficiency. It can quickly and accurately obtain the reducer bearing force that changes with the input speed. Furthermore, the obtained bearing force data can be used to predict the noise of electric vehicle reducer variable speed operation.

[0116] Please see Figure 6 A bearing force prediction device for speed reducer under variable speed conditions is provided to meet one of the purposes of this application, including a material parameter acquisition module 1100, a steady-state bearing force determination module 1200, a variable speed bearing force determination module 1300, and a speed reducer noise calculation module 1400. The module includes the following components: a material parameter acquisition module 1100, configured to respond to a bearing force prediction command for an electric vehicle reducer under variable speed conditions, and determine the corresponding gear shaft transmission geometric model and material parameters of each component of the reducer; a steady-state bearing force determination module 1200, configured to construct a dynamic simulation model of the reducer based on the gear shaft transmission geometric model and the material parameters, and calculate the bearing force data and corresponding steady-state speed of the reducer under multiple steady-state conditions based on the dynamic simulation model; a variable speed bearing force determination module 1300, configured to determine the bearing force of the electric vehicle reducer under variable speed conditions based on a pre-trained variable speed bearing force prediction model, according to the bearing force data and corresponding steady-state speed under steady-state conditions, wherein the variable speed bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module; and a reducer noise calculation module 1400, configured to calculate and determine the noise of the electric vehicle reducer under variable speed conditions based on the bearing force under variable speed conditions, thereby completing the bearing force prediction of the electric vehicle reducer under variable speed conditions.

[0117] Based on any embodiment of this application, please refer to Figure 7 Another embodiment of this application also provides an electronic device, which can be implemented by a computer device, such as... Figure 7 The diagram shows the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. The computer-readable storage medium stores an operating system, a database, and computer-readable instructions. The database may store control information sequences. When executed by the processor, the computer-readable instructions enable the processor to implement a method for predicting bearing force under variable speed conditions in a speed reducer. The processor provides computational and control capabilities, supporting the operation of the entire computer device. The memory stores computer-readable instructions, which, when executed by the processor, enable the processor to execute the speed reducer variable speed bearing force prediction method of this application. The network interface of the computer device is used for communication with a terminal. Those skilled in the art will understand that… Figure 7The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0118] In this embodiment, the processor executes the specific functions of each module and its sub-modules in section 6, and the memory stores the program code and various types of data required to execute the aforementioned modules or sub-modules. The network interface is used for data transmission between the user terminal and the server. In this embodiment, the memory stores the program code and data required to execute all modules / sub-modules in the reducer shifting bearing force prediction device of this application, and the server can call the server's program code and data to execute the functions of all sub-modules.

[0119] This application also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the bearing force prediction method for speed reducer shifting conditions described in any embodiment of this application.

[0120] This application also provides a computer program product, including a computer program / instructions that, when executed by one or more processors, implement the steps of the method for predicting bearing force under speed change conditions of a reducer as described in any embodiment of this application.

[0121] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0122] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

[0123] In summary, this application addresses the problems of large errors in obtaining steady-state bearing force in electric vehicle reducers, long calculation time for bearing force under variable speed conditions, and large amount of repetitive work. It proposes a method for predicting bearing force under variable speed conditions in electric vehicle reducers. This method can quickly and accurately obtain the reducer bearing force that varies with the input speed. Furthermore, the obtained bearing force data can be used to predict noise under variable speed conditions in electric vehicle reducers, greatly saving time and costs.

Claims

1. A method for predicting bearing force under variable speed operation of a speed reducer, characterized in that, include: In response to the bearing force prediction command of the electric vehicle reducer under the variable speed condition, determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer. Based on the gear shaft transmission geometric model and the material parameters, a corresponding dynamic simulation model of the reducer is constructed. Based on the dynamic simulation model, the bearing force data of the reducer under multiple steady-state conditions and its corresponding steady-state speed are calculated and determined. The pre-trained variable-speed bearing force prediction model determines the bearing force of the electric vehicle reducer under variable-speed conditions based on the bearing force data under steady-state conditions and its corresponding steady-state speed. This includes: inputting the bearing force data under steady-state conditions and its corresponding steady-state speed into the pre-trained variable-speed bearing force prediction model; truncating the bearing force data in the time domain output by the adaptive neural network in the variable-speed bearing force prediction model using the frequency domain transformation module; performing a discrete Fourier transform on the truncated bearing force data; and performing a frequency domain windowed averaging operation on the data after the discrete Fourier transform; and outputting the bearing force data of the electric vehicle reducer under variable-speed conditions in both the time and frequency domains. The variable-speed bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module. The steps for training the variable-speed bearing force prediction model include: Based on the characteristics of bearing force variation over time under steady-state conditions in the sample dataset, a loop unit is selected, and the loss function is determined according to the task type of bearing force prediction to construct the variable speed bearing force prediction model. Each data item in the sample dataset and the label dataset is segmented into data segments of equal length. These data segments undergo Discrete Fourier Transform, and the transformed data is normalized to have a variance of 1 and a mean of 0. A portion of the normalized sample dataset is used as the training set, and this training set is input into the variable speed bearing force prediction model for training. When the variable speed bearing force prediction model reaches a preset number of iterations, the training of the variable speed bearing force prediction model is completed. The noise of the electric vehicle reducer under variable speed conditions is determined based on the bearing force calculation under variable speed conditions, so as to complete the bearing force prediction of the electric vehicle reducer under variable speed conditions.

2. The method for predicting bearing force under variable speed conditions of a reducer according to claim 1, characterized in that, The steps for constructing a dynamic simulation model include: Determine the initial data corresponding to the reducer housing, gears, shafts, and bearings of the electric vehicle reducer, and construct the corresponding gear and shaft transmission geometric model based on the initial data corresponding to the reducer housing, gears, shafts, and bearings. The various parts of the reducer include one or any combination of the reducer housing, gears, shafts, and bearings. Determine the material parameters corresponding to the reducer housing, gears, shafts, and bearings, as well as the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model. Input the material parameters corresponding to the reducer housing, gears, shafts, and bearings, as well as the motion boundary conditions, periodic boundary conditions, number of solution steps, and solution step size of the gear shaft transmission geometric model into the gear shaft transmission geometric model to construct a dynamic simulation model.

3. The method for predicting bearing force under variable speed conditions of a reducer according to claim 1, characterized in that, The steps for calculating and determining the bearing force data and corresponding steady-state speed of the reducer under multiple steady-state operating conditions based on the dynamic simulation model include: The excitation of bearing force caused by the reducer is mainly driven by the helical gear transmission error and the time-varying meshing stiffness of the helical gear pair. Simulation analysis is performed based on the aforementioned dynamic simulation model, where the helical gear transmission error is expressed as: ， in, For the transmission error of helical gears, , Let be the theoretical rotation angle of the driving gear and the driven gear. The actual rotation angle of the driven gear. , Let be the base circle radius of the driving gear and the driven gear. This is the theoretical transmission ratio; The time-varying meshing stiffness of the helical gear pair: ， in, For time-varying meshing stiffness of helical gear pairs, The meshing frequency, The initial phase angle, For the degree of overlap, This refers to the meshing stiffness of a single gear pair; The calculated helical gear transmission error and the time-varying meshing stiffness of the helical gear pair are applied to the dynamic simulation model to obtain the bearing force data of the electric vehicle reducer under multiple steady-state operating conditions. The bearing force data obtained from the simulation under steady-state operating conditions, along with their corresponding steady-state operating speed and operating time, form the sample data set for the bearing force prediction model under variable speed conditions.

4. The method for predicting bearing force under variable speed conditions of a reducer according to claim 3, characterized in that, After the step of combining the bearing force data obtained from simulation under steady-state operating conditions with its corresponding steady-state operating speed and operating time to form a sample data set for the bearing force prediction model under variable speed conditions, the following steps are included: In response to the instruction to construct a tag data set, the actual bearing force of the bearing force sensor in the electric vehicle reducer and the corresponding actual steady-state speed are obtained under steady-state conditions. The actual bearing force, the actual steady-state operating speed, and the operating time are used to construct a labeled data set for the bearing force prediction model under variable speed conditions.

5. The method for predicting bearing force under variable speed operation of a reducer according to any one of claims 1 to 4, characterized in that, The adaptive neural network in the bearing force prediction model for variable speed operating conditions includes an input layer, a long short-time memory layer, several fully connected layers, and a data output layer.

6. A bearing force prediction device for speed reducer operation, characterized in that, include: The material parameter acquisition module is configured to respond to the bearing force prediction command of the electric vehicle reducer under the variable speed condition, and determine the corresponding gear shaft transmission geometric model of the electric vehicle reducer and the corresponding material parameters of each part of the reducer. The steady-state bearing force determination module is configured to construct a dynamic simulation model of the reducer based on the gear shaft transmission geometric model and the material parameters, and calculate and determine the bearing force data of the reducer under multiple steady-state conditions and its corresponding steady-state speed based on the dynamic simulation model. The variable speed bearing force determination module is configured to determine the bearing force of the electric vehicle reducer under variable speed conditions based on the bearing force data under steady-state conditions and its corresponding steady-state speed, according to a pre-trained variable speed bearing force prediction model. The module includes: inputting the steady-state bearing force data and its corresponding steady-state speed into the pre-trained variable speed bearing force prediction model; truncating the time-domain bearing force data output from the adaptive neural network in the variable speed bearing force prediction model using a frequency domain transformation module; performing a discrete Fourier transform on the truncated bearing force data; and performing a frequency domain windowed averaging operation on the data after the discrete Fourier transform; and outputting the bearing force data of the electric vehicle reducer under variable speed conditions in both the time and frequency domains. The variable speed bearing force prediction model is constructed by an adaptive neural network followed by a frequency domain transformation module. The steps for training the variable speed bearing force prediction model include: Based on the characteristics of bearing force variation over time under steady-state conditions in the sample dataset, a loop unit is selected, and the loss function is determined according to the task type of bearing force prediction to construct the variable speed bearing force prediction model. Each data item in the sample dataset and the label dataset is segmented into data segments of equal length. These data segments undergo Discrete Fourier Transform, and the transformed data is normalized to have a variance of 1 and a mean of 0. A portion of the normalized sample dataset is used as the training set, and this training set is input into the variable speed bearing force prediction model for training. When the variable speed bearing force prediction model reaches a preset number of iterations, the training of the variable speed bearing force prediction model is completed. The reducer noise calculation module is configured to determine the noise of the electric vehicle reducer under the variable speed condition based on the bearing force calculation under the variable speed condition, so as to complete the bearing force prediction of the electric vehicle reducer under the variable speed condition.

7. An electronic device comprising a central processing unit and a memory, characterized in that, The central processing unit is used to invoke and run a computer program stored in the memory to perform the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, It stores, in the form of computer-readable instructions, a computer program implemented according to any one of claims 1 to 5, which, when invoked by a computer, executes the steps included in the corresponding method.

Images