CN-119749569-B - Electric automobile wheel slip pre-judging method and system

CN119749569BCN 119749569 BCN119749569 BCN 119749569BCN-119749569-B

Abstract

The invention relates to the technical field of automobiles, in particular to a method and a system for predicting wheel slip of an electric automobile, which comprise the steps of obtaining various types of data of different attached roads and driving the electric automobile, obtaining all data unit sets through the various types of data, performing off-line training on a neural network model according to all the data unit sets to obtain a trained model, transplanting the trained model to a motor MCU (micro control unit), inputting the data unit sets of the electric automobile to the motor MCU to predict the wheel slip, and implementing torque reduction and torque transfer measures on a driving motor of the wheel with a slip trend through a VCU when the wheel slips. According to the invention, through the pre-judging analysis of the wheel slip of the electric automobile, the driving safety of a user and the controllability of the vehicle are improved, and the user experience is also improved.

Inventors

DONG YITAO
WANG YILIN
ZHAO GUANGCHANG

Assignees

奇瑞汽车股份有限公司

Dates

Publication Date: 20260505
Application Date: 20250206

Claims (6)

1. The method for predicting the wheel slip of the electric automobile is characterized by comprising the following steps of: acquiring various types of data of different attached pavements and running electric vehicles, and acquiring all data unit sets by the various types of data, wherein the data unit sets comprise Millisecond is the length of time a set of data units to The millisecond is the sampling time interval, and all data unit sets of the electric automobile on different attached roads are obtained, wherein one data unit set contains various types of data, including motor rotation angular velocity and angular acceleration, motor output torque, vehicle speed and vehicle acceleration, tire speed and tire acceleration, steering wheel rotation angle and accelerator pedal signals, For the parameter of the length of time to be preset, The method comprises the steps of presetting time interval parameters; The method comprises the steps of performing offline training on a neural network model according to all data unit sets to obtain a trained model, wherein the method comprises the steps of marking a slip state corresponding to each data unit set, and obtaining a slip state characteristic corresponding to each data unit set, wherein the slip state characteristics are only two types, namely slip and non-slip; The data unit set of the electric automobile is input into the motor MCU controller to predict the wheel slip condition, including, before the current moment is obtained The data unit set in millisecond is recorded as the data unit set at the current moment, the data unit set at the current moment is input into the motor MCU controller, and then the motor MCU controller is used for pre-judging the state information of the wheel slip of the electric automobile after the current moment, wherein, When the wheels of the electric automobile slip, the torque reducing and torque transferring measures are implemented on the driving motors of the wheels with the slip trend through the VCU, and the overall slip condition of the electric automobile is reduced by implementing the torque reducing and torque transferring measures on the driving motors of the wheels with the slip trend through the VCU according to the slip state information of all the wheels of the electric automobile.
2. The method for predicting wheel slip of an electric vehicle of claim 1, wherein the motor MCU controller comprises a processor core, a memory and an input/output interface.
3. The method for predicting wheel slip of an electric vehicle according to claim 2, wherein the memory comprises a program memory and a data memory.
4. An electric vehicle wheel slip prognosis system, comprising: the data acquisition module is used for acquiring various types of data of different attached roads and running electric vehicles, and acquiring all data unit sets through the various types of data, wherein the data acquisition module comprises Millisecond is the length of time a set of data units to The millisecond is the sampling time interval, and all data unit sets of the electric automobile on different attached roads are obtained, wherein one data unit set contains various types of data, including motor rotation angular velocity and angular acceleration, motor output torque, vehicle speed and vehicle acceleration, tire speed and tire acceleration, steering wheel rotation angle and accelerator pedal signals, For the parameter of the length of time to be preset, The method comprises the steps of presetting time interval parameters; the model training module is used for carrying out offline training on the neural network model according to all the data unit sets to obtain a trained model, and comprises the step of marking the slip state corresponding to each data unit set, so that one slip state characteristic corresponding to each data unit set can be obtained; The wheel slip pre-judging and adjusting module is used for transplanting the trained model to the motor MCU controller, inputting the data unit set of the electric automobile to the motor MCU controller to pre-judge the wheel slip condition, including, before the current moment is obtained The data unit set in millisecond is recorded as the data unit set at the current moment, the data unit set at the current moment is input into the motor MCU controller, and then the motor MCU controller is used for pre-judging the state information of the wheel slip of the electric automobile after the current moment, wherein, When the wheels of the electric automobile slip, the torque reducing and torque transferring measures are implemented on the driving motors of the wheels with the slip trend through the VCU, and the overall slip condition of the electric automobile is reduced by implementing the torque reducing and torque transferring measures on the driving motors of the wheels with the slip trend through the VCU according to the slip state information of all the wheels of the electric automobile.
5. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing a method of predicting wheel slip of an electric vehicle as claimed in any one of claims 1 to 3 when the computer program is executed by the processor.
6. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program, which when executed by a processor, implements a wheel slip pre-determination method for an electric vehicle according to any one of claims 1-3.

Description

Electric automobile wheel slip pre-judging method and system Technical Field The invention relates to the technical field of automobiles, in particular to a method and a system for predicting wheel slip of an electric automobile. Background The electric automobile has the advantages of energy conservation, environmental protection, strong power, good NVH (Motor Vehicle Noise ) performance, smooth driving and the like, and the research of key technology is paid attention to each automobile enterprise. The electric automobile can have different slip risks under different road conditions (such as ice and snow, rainwater, mud and the like), the pre-judging method needs to be flexible and adaptive, has high instantaneity, can judge whether the slip risks exist in extremely short time and timely react, so that the electric automobile wheel slip pre-judging method needs to be designed, and can improve driving safety, particularly under severe weather conditions, and can provide support for future intelligent driving and automatic driving technologies. Along with the continuous development of artificial intelligence and sensing technology, the pre-judging precision and response speed can be gradually improved, and safer driving experience is ensured, so that the method has important significance for the research of electric automobile wheel slip pre-judging. The driving power of the electric automobile comes from the motor, and the motor has the characteristics of large output torque and quick torque response, so that the electric automobile is extremely easy to generate tire slipping on the road surfaces with low adhesion and the like. In a traditional automobile, the speed and the wheel speed of the automobile are calculated through an ESP (Electronic Stability Program and an electronic automobile body stabilizing system), so that the slip rate is calculated to judge the slip, and the anti-slip control is realized by adjusting the output torque of a motor and the braking force on a driving wheel. The anti-slip control of the ESP requires the cooperation of each sensor and an actuating mechanism, is expensive to realize, has low timeliness compared with the quick response of a motor, and has poor experience when the control is implemented again under the condition of slipping. Disclosure of Invention The invention provides a method and a system for predicting wheel slip of an electric automobile, which are used for solving the existing problems. The aim of the invention can be achieved by the following technical scheme: The first aspect of the invention provides a method for predicting wheel slip of an electric automobile, which comprises the following steps: acquiring various types of data of different attached pavements and driving an electric automobile, and acquiring all data unit sets through the various types of data; The method comprises the steps of transplanting a trained model to a motor MCU controller, inputting a data unit set of an electric automobile to the motor MCU controller to predict the wheel slip condition, and implementing measures of torque reduction and torque transfer on a driving motor of a wheel with a slip trend through a VCU when the wheel slips. Further, the obtaining various types of data of different attached roads and running electric vehicles, and obtaining all data unit sets through the various types of data includes: To be used for Millisecond is the length of time a set of data units toThe millisecond is a sampling time interval, and all data unit sets of the electric automobile on different attached roads are obtained, wherein one data unit set contains various types of data, including motor rotation angular velocity and angular acceleration, motor output torque, vehicle speed and vehicle acceleration, tire speed and tire acceleration, steering wheel rotation angle and accelerator pedal signals; Wherein, the For the parameter of the length of time to be preset,Is a preset time interval parameter. Further, the offline training of the neural network model according to all the data unit sets includes: marking the slip state corresponding to each data unit set, and obtaining a slip state characteristic corresponding to each data unit set, wherein the slip state characteristics are slip and non-slip respectively; and inputting all the data unit sets and the slip state characteristics corresponding to each data unit set into a neural network model for offline training to obtain a trained model. Further, the motor MCU controller comprises a processor core, a memory and an input-output interface. Further, the memory comprises a program memory and a data memory. Further, the inputting the data unit set of the electric automobile to the motor MCU controller to predict the wheel slip condition includes: Before the current moment is acquired The method comprises the steps of marking a data unit set in milliseconds as the data unit set at the current moment, inputting the data unit set at