Intelligent electric vehicle steering stability prediction method

Abstract

The invention discloses an electromobile steering stability prediction method based on intelligence, which belongs to the technical field of intellectualization and comprises the following steps: S1, acquiring operating parameter data information of an electromobile in real time through various sensors on the electromobile; the method comprises the following steps: S1, acquiring the detailed construction and design data of a power system, a steering system, a suspension system and a vehicle body system of an electric vehicle, S2, constructing a simulation model according to the acquired data of the electric vehicle, and accurately configuring and simulating the model to generate virtual data under various driving conditions, by integrating various sensor data on the electric vehicle, including detailed construction and design data of a power system, a steering system, a suspension system and a vehicle body system, a comprehensive electric vehicle state data set is constructed, the steering stability of the electric vehicle can be evaluated more accurately, and when a simulation model is constructed, the steering stability of the electric vehicle can be evaluated more accurately. Various driving conditions are considered to generate abundant virtual data, and the accuracy and applicability of the prediction model are improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent technology, and specifically refers to an intelligent-based method for predicting the steering stability of an electric vehicle. Background Art

[0002] With the rapid development of electric vehicle technology and the continuous advancement of autonomous driving technology, the steering stability of electric vehicles has received increasing attention. Existing electric vehicle steering stability control methods mainly focus on anti-lock braking systems and drive anti-skid systems, which adjust the vehicle's running posture by controlling the vehicle's power source. However, these methods often fail to achieve ideal control effects when faced with complex and changeable road conditions and driving environments.

[0003] However, the existing electric vehicle steering stability prediction still has certain defects. The existing electric vehicle steering stability prediction relies on limited sensor data. A single data source cannot fully reflect the status of the vehicle, resulting in inaccurate evaluation results. Existing methods usually adopt simplified simulation models and fail to fully consider the vehicle behavior under various driving conditions, resulting in insufficient richness of the generated virtual data. The accuracy and applicability of the simulation results are limited. The virtual data generated by the existing methods are relatively single and cannot cover all possible driving conditions, which limits the generalization ability of the prediction model and only provides prediction results. There is a lack of automatic adjustment mechanism, which requires manual intervention by the driver, increasing the driving burden. When potential instability is detected, there is a lack of effective early warning signals, and the driver cannot be reminded to take measures in time. Therefore, an intelligent electric vehicle steering stability prediction method is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent electric vehicle steering stability prediction method to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent electric vehicle steering stability prediction method, comprising the following steps:

[0006] S1. Obtaining real-time operating parameter data information of electric vehicles through various sensors on electric vehicles, and obtaining detailed structure and design data of the power system, steering system, suspension system and body system of electric vehicles;

[0007] S2. Build a simulation model based on the acquired electric vehicle data, and accurately configure and simulate the model to generate virtual data under various driving conditions;

[0008] S3, obtaining electric vehicle simulation data according to the electric vehicle simulation model to perform data preprocessing;

[0009] S4, extracting key features for analyzing steering stability through preprocessed data;

[0010] S5. Training a steering stability prediction model based on the extracted key features and combined with historical data;

[0011] S6, inputting the real-time data into the trained steering stability prediction model, and combining it with the electric vehicle simulation model for real-time analysis to evaluate the steering stability index of the vehicle in the current state;

[0012] S7: Automatically adjust and provide feedback based on steering stability indicators.

[0013] Wherein, the S1, obtains the electric vehicle operating parameter data information in real time through various sensors on the electric vehicle, and obtains the detailed structure and design data of the electric vehicle's power system, steering system, suspension system and body system;

[0014] Power system: obtain motor speed, torque, battery voltage, and current information from the motor controller and battery management system, and obtain relevant data of the engine control unit through the on-board diagnostic interface;

[0015] Steering system: Obtain steering angle and steering torque information from the electric power steering system, and obtain relevant data from the lane keeping assist system;

[0016] Suspension system: obtain the status of the shock absorber, spring compression information and dynamic adjustment parameters from the suspension control system;

[0017] Body system: obtains status information of doors, windows, seats, etc. from the body control module, and obtains airbag status and trigger condition information from the airbag control system;

[0018] Integrate all data from different sensors and systems into a unified data format to form a complete electric vehicle status data set, add a timestamp to each data record, and store the real-time collected data through the on-board storage device.

[0019] Among them, the S2 constructs a simulation model based on the acquired electric vehicle data, and accurately configures and simulates the model to generate virtual data under various driving conditions; in the simulation software, based on the detailed data of the electric vehicle, gradually constructs simulation models of various systems, including the motor model and battery model of the power system, the steering mechanism model of the steering system, the spring-damper model of the suspension system, and the mass-inertia model of the body system, integrates the simulation models of various systems to form a complete electric vehicle simulation model, and preliminarily verifies the integrated model to check the correctness and stability of the model.

[0020] Among them, the S2 accurately configures the simulation model according to the constructed electric vehicle simulation model, including setting vehicle parameters, road conditions and environmental parameters. After the configuration is completed, the simulation model is run to simulate the behavior of the electric vehicle under various driving conditions, including different vehicle speeds, different steering angles and different road conditions, so as to generate rich virtual data, and compare the simulation results with the actual vehicle data to verify the accuracy of the model. If a large deviation is found, the model parameters are adjusted and the simulation is re-run, the virtual data generated during the simulation process is exported, and the virtual data generated by the simulation is merged with the actual collected data to form a more comprehensive and diverse data set.

[0021] Among them, S3, according to the electric vehicle simulation model, obtains the electric vehicle simulation data, performs format conversion, removes invalid data, outliers and denoises, and aligns the data to the same time axis through an interpolation method.

[0022] Wherein, said S4 extracts key features for analyzing steering stability through the preprocessed data, including comprehensive sideslip index, dynamic stability coefficient and steering response sensitivity;

[0023] The comprehensive sideslip index takes into account vehicle speed, lateral acceleration and steering angle. The formula for the comprehensive sideslip index is:

[0024]

[0025] In the formula, a y represents lateral acceleration, v represents vehicle speed, θ represents steering angle, and ∈ represents a constant;

[0026] Dynamic stability coefficient comprehensive lateral acceleration a y , longitudinal acceleration a x and vehicle speed v, the dynamic stability coefficient is realized by:

[0027]

[0028] Steering response sensitivity The steering response sensitivity is calculated through the relationship between the steering angle change rate and the vehicle speed. The implementation formula is:

[0029]

[0030] In the formula, dθ / dt represents the rate of change of the steering angle, v represents the vehicle speed, and ∈ represents a constant.

[0031] In S4, the features related to steering stability are analyzed according to the extracted key features. The extracted feature data is assumed to be X and the target variable is y. The target variable is represented as a steering stability index. The correlation between each feature and the target variable is calculated by the Pearson correlation coefficient. The implementation formula is:

[0032]

[0033] In the formula, x ij represents the jth eigenvalue of the i-th sample, represents the mean of the jth feature, y i represents the target variable value of the i-th sample, represents the mean of the target variable, n represents the number of samples, and m represents the number of features;

[0034] Set the correlation coefficient threshold T and select the top k features with the highest absolute correlation coefficient.

[0035] In S5, a steering stability prediction model is trained based on the extracted key features and combined with historical data. Assuming the relationship between the target variable y and the input feature X, the steering stability prediction model is:

[0036] y=Xβ+∈,

[0037] In the formula, y represents the target vector, X represents the input eigenvalue, β represents the regression coefficient vector, and ∈ represents the error vector;

[0038] According to the results of the steering stability prediction model, the optimal regression coefficient β is found, and the implementation formula is:

[0039] J(β)=(y-Xβ) T (y-Xβ).

[0040] Among them, the S6 inputs real-time data into a trained steering stability prediction model, and performs real-time analysis in combination with an electric vehicle simulation model, inputs real-time collected data into the simulation model, simulates the behavior of the vehicle in the current state, analyzes the output results of the simulation model, identifies possible unstable situations, combines the results of the prediction model and the simulation model, and performs a comprehensive evaluation to evaluate the steering stability indicators of the vehicle in the current state.

[0041] Among them, the S7 automatically adjusts and provides feedback based on the steering stability index, continuously monitors the steering stability index calculated by the prediction model and the simulation model, and sets a threshold value to determine whether the current state is unstable or potentially unstable. If it is detected that the steering stability index exceeds the threshold, a corresponding warning signal is generated, and corresponding adjustments are automatically made in combination with the electronic stability program, and real-time feedback information is provided to the driver to inform him of the current vehicle status.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. The present invention constructs a comprehensive electric vehicle state data set by integrating various sensor data on electric vehicles, including detailed construction and design data of the power system, steering system, suspension system and body system, which helps to more accurately evaluate the steering stability of electric vehicles. When constructing the simulation model, various driving conditions are considered to generate rich virtual data, which further improves the accuracy and applicability of the prediction model;

[0044] 2. The present invention can analyze the steering stability index of electric vehicles in real time through intelligent prediction models and simulation models, and automatically adjust and feedback according to the indexes, which greatly reduces the burden on drivers and improves driving safety and comfort. By setting thresholds and early warning signals, the present invention can promptly remind drivers when potential unstable situations are detected, and automatically make corresponding adjustments to avoid accidents.

[0045] 3. The present invention can more comprehensively reflect the steering stability of the vehicle by extracting key features such as the comprehensive sideslip index, dynamic stability coefficient and steering response sensitivity. The first k features with the highest absolute correlation coefficient are selected through the Pearson correlation coefficient, and the features that have the greatest impact on the prediction of steering stability are effectively screened out. This method not only improves the efficiency of feature selection, but also enhances the explanatory power and prediction accuracy of the model.

[0046] 4. The present invention trains an accurate steering stability prediction model by extracting key features and historical data. The linear regression model finds the optimal regression coefficient through the least squares method and establishes a concise and effective prediction model. The model can quickly and accurately predict the vehicle's steering stability index, providing a solid foundation for real-time analysis and feedback. The model is trained in combination with historical data, and the generalization ability of the model is evaluated through methods such as cross-validation, ensuring the stability and reliability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The operation flow of the intelligent electric vehicle steering stability prediction method of the present invention is as follows Figure 1 ;

[0048] Figure 2 The operation flow of the intelligent electric vehicle steering stability prediction method of the present invention is as follows Figure 2 ;

[0049] Figure 3 The operation flow of the intelligent electric vehicle steering stability prediction method of the present invention is as follows Figure 3 . DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Example

[0052] See also Figure 1-Figure 3 As shown, the present invention provides a technical solution: comprising the following steps:

[0053] S1. Obtaining real-time operating parameter data information of electric vehicles through various sensors on electric vehicles, and obtaining detailed structure and design data of the power system, steering system, suspension system and body system of electric vehicles;

[0054] S2. Build a simulation model based on the acquired electric vehicle data, and accurately configure and simulate the model to generate virtual data under various driving conditions;

[0055] S3, obtaining electric vehicle simulation data according to the electric vehicle simulation model to perform data preprocessing;

[0056] S4, extracting key features for analyzing steering stability through preprocessed data;

[0057] S5. Training a steering stability prediction model based on the extracted key features and combined with historical data;

[0058] S6, inputting the real-time data into the trained steering stability prediction model, and combining it with the electric vehicle simulation model for real-time analysis to evaluate the steering stability index of the vehicle in the current state;

[0059] S7: Automatically adjust and provide feedback based on steering stability indicators.

[0060] Wherein, the S1, obtains the electric vehicle operating parameter data information in real time through various sensors on the electric vehicle, and obtains the detailed structure and design data of the electric vehicle's power system, steering system, suspension system and body system;

[0061] Power system: obtain motor speed, torque, battery voltage, and current information from the motor controller and battery management system, and obtain relevant data of the engine control unit through the on-board diagnostic interface;

[0062] Steering system: Obtain steering angle and steering torque information from the electric power steering system, and obtain relevant data from the lane keeping assist system;

[0063] Suspension system: obtain the status of the shock absorber, spring compression information and dynamic adjustment parameters from the suspension control system;

[0064] Body system: obtains status information of doors, windows, seats, etc. from the body control module, and obtains airbag status and trigger condition information from the airbag control system;

[0065] Integrate all data from different sensors and systems into a unified data format to form a complete electric vehicle status data set, add a timestamp to each data record, and store the real-time collected data through the on-board storage device.

[0066] Among them, the S2 constructs a simulation model based on the acquired electric vehicle data, and accurately configures and simulates the model to generate virtual data under various driving conditions; in the simulation software, based on the detailed data of the electric vehicle, gradually constructs simulation models of various systems, including the motor model and battery model of the power system, the steering mechanism model of the steering system, the spring-damper model of the suspension system, and the mass-inertia model of the body system, integrates the simulation models of various systems to form a complete electric vehicle simulation model, and preliminarily verifies the integrated model to check the correctness and stability of the model.

[0067] Among them, the S2 accurately configures the simulation model according to the constructed electric vehicle simulation model, including setting vehicle parameters, road conditions and environmental parameters. After the configuration is completed, the simulation model is run to simulate the behavior of the electric vehicle under various driving conditions, including different vehicle speeds, different steering angles and different road conditions, so as to generate rich virtual data, and compare the simulation results with the actual vehicle data to verify the accuracy of the model. If a large deviation is found, the model parameters are adjusted and the simulation is re-run, the virtual data generated during the simulation process is exported, and the virtual data generated by the simulation is merged with the actual collected data to form a more comprehensive and diverse data set.

[0068] Among them, S3, according to the electric vehicle simulation model, obtains the electric vehicle simulation data, performs format conversion, removes invalid data, outliers and denoises, and aligns the data to the same time axis through an interpolation method.

[0069] Wherein, said S4 extracts key features for analyzing steering stability through the preprocessed data, including comprehensive sideslip index, dynamic stability coefficient and steering response sensitivity;

[0070] The comprehensive sideslip index takes into account vehicle speed, lateral acceleration and steering angle. The formula for the comprehensive sideslip index is:

[0071]

[0072] In the formula, a y represents lateral acceleration, v represents vehicle speed, θ represents steering angle, and ∈ represents a constant;

[0073] Dynamic stability coefficient comprehensive lateral acceleration a y , longitudinal acceleration a x and vehicle speed v, the dynamic stability coefficient is realized by:

[0074]

[0075] Steering response sensitivity The steering response sensitivity is calculated through the relationship between the steering angle change rate and the vehicle speed. The implementation formula is:

[0076]

[0077] In the formula, dθ / dt represents the rate of change of the steering angle, v represents the vehicle speed, and ∈ represents a constant.

[0078] In S4, the features related to steering stability are analyzed according to the extracted key features. The extracted feature data is assumed to be X and the target variable is y. The target variable is represented as a steering stability index. The correlation between each feature and the target variable is calculated by the Pearson correlation coefficient. The implementation formula is:

[0079]

[0080] In the formula, x ij represents the jth eigenvalue of the i-th sample, represents the mean of the jth feature, y i represents the target variable value of the i-th sample, represents the mean of the target variable, n represents the number of samples, and m represents the number of features;

[0081] Set the correlation coefficient threshold T and select the top k features with the highest absolute correlation coefficient.

[0082] In S5, a steering stability prediction model is trained based on the extracted key features and combined with historical data. Assuming the relationship between the target variable y and the input feature X, the steering stability prediction model is:

[0083] y=Xβ+∈,

[0084] In the formula, y represents the target vector, X represents the input eigenvalue, β represents the regression coefficient vector, and ∈ represents the error vector;

[0085] According to the results of the steering stability prediction model, the optimal regression coefficient β is found, and the implementation formula is:

[0086] J(β)=(y-Xβ)T (y-Xβ).

[0087] Among them, the S6 inputs real-time data into a trained steering stability prediction model, and performs real-time analysis in combination with an electric vehicle simulation model, inputs real-time collected data into the simulation model, simulates the behavior of the vehicle in the current state, analyzes the output results of the simulation model, identifies possible unstable situations, combines the results of the prediction model and the simulation model, and performs a comprehensive evaluation to evaluate the steering stability indicators of the vehicle in the current state.

[0088] Among them, the S7 automatically adjusts and provides feedback based on the steering stability index, continuously monitors the steering stability index calculated by the prediction model and the simulation model, and sets a threshold value to determine whether the current state is unstable or potentially unstable. If it is detected that the steering stability index exceeds the threshold, a corresponding warning signal is generated, and corresponding adjustments are automatically made in combination with the electronic stability program, and real-time feedback information is provided to the driver to inform him of the current vehicle status.

[0089] Working principle: The key operating parameters of electric vehicles, such as vehicle speed, steering angle, acceleration, etc., are acquired in real time through a variety of high-precision sensors on electric vehicles. At the same time, detailed data of power, steering, suspension and body systems are acquired from motor controllers, battery management systems, electric power steering systems, suspension control systems and body control modules. All data from different sensors and systems are integrated into a unified data format to form a complete electric vehicle status data set. Timestamps are added to each data record, and the collected data is stored in real time through on-board storage devices.

[0090] In the simulation software, simulation models of various systems are gradually constructed according to the detailed data of electric vehicles, including the motor model and battery model of the power system, the steering mechanism model of the steering system, the spring-damper model of the suspension system, and the mass-inertia model of the body system, etc. The simulation models of various systems are integrated to form a complete electric vehicle simulation model, and the integrated simulation model is preliminarily verified to check the correctness and stability of the model. The model is accurately configured according to the actual vehicle data, including setting vehicle parameters, road conditions and environmental parameters, running the simulation model, simulating the behavior of electric vehicles under various driving conditions, generating rich virtual data, and comparing the simulation results with the actual vehicle data to verify the accuracy of the model. If a large deviation is found, the model parameters are adjusted and the simulation is re-run until the model accuracy meets the requirements. The virtual data generated by the simulation model and the actual collected data are preprocessed by format conversion, removal of invalid values ​​and outliers, denoising and other preprocessing operations, and the data are aligned to the same time axis through interpolation methods. The data related to the analysis of steering stability are extracted from the preprocessed data. Key features, further analyze the extracted key features, calculate the correlation between each feature and the target variable through the Pearson correlation coefficient, set the threshold of the correlation coefficient, select the first k features with the highest absolute correlation coefficient as input features, train the steering stability prediction model based on the extracted key features and historical data, establish the correlation relationship between the target variable and the input features through regression analysis method, find the optimal regression coefficient through optimization algorithm, so that the prediction accuracy of the prediction model is the highest, input the real-time collected data into the trained steering stability prediction model, and combine it with the electric vehicle simulation model for real-time analysis, simulate the vehicle's behavior in the current state, analyze the output results of the simulation model, identify possible unstable situations, combine the results of the prediction model and the simulation model for comprehensive evaluation, and obtain the steering stability index of the vehicle in the current state, continuously monitor the steering stability index, and set the threshold to determine whether the current state is unstable or potentially unstable. If the steering stability index is detected to exceed the threshold, a corresponding warning signal is generated, and the electronic stability program assist system is combined to automatically make corresponding adjustments.

[0091] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0092] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An intelligent electric vehicle steering stability prediction method is characterized by: The following steps are involved: S1. Obtaining real-time operating parameter data information of electric vehicles through various sensors on electric vehicles, and obtaining detailed structure and design data of the power system, steering system, suspension system and body system of electric vehicles; S2. Build a simulation model based on the acquired electric vehicle data, and accurately configure and simulate the model to generate virtual data under various driving conditions; S3, obtaining electric vehicle simulation data according to the electric vehicle simulation model to perform data preprocessing; S4, extracting key features for analyzing steering stability through preprocessed data; S5. Training a steering stability prediction model based on the extracted key features and combined with historical data; S6, inputting the real-time data into the trained steering stability prediction model, and combining it with the electric vehicle simulation model for real-time analysis to evaluate the steering stability index of the vehicle in the current state; S7: Automatically adjust and provide feedback based on steering stability indicators.

2. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: S1, obtaining real-time operating parameter data information of the electric vehicle through various sensors on the electric vehicle, and obtaining detailed structure and design data of the power system, steering system, suspension system and body system of the electric vehicle; Power system: obtain motor speed, torque, battery voltage, and current information from the motor controller and battery management system, and obtain relevant data of the engine control unit through the on-board diagnostic interface; Steering system: Obtain steering angle and steering torque information from the electric power steering system, and obtain relevant data from the lane keeping assist system; Suspension system: obtain the status of the shock absorber, spring compression information and dynamic adjustment parameters from the suspension control system; Body system: obtains status information of doors, windows, seats, etc. from the body control module, and obtains airbag status and trigger condition information from the airbag control system; Integrate all data from different sensors and systems into a unified data format to form a complete electric vehicle status data set, add a timestamp to each data record, and store the real-time collected data through the on-board storage device.

3. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: The S2 constructs a simulation model based on the acquired electric vehicle data, and accurately configures and simulates the model to generate virtual data under various driving conditions; in the simulation software, based on the detailed data of the electric vehicle, gradually constructs simulation models of various systems, including a motor model and a battery model of the power system, a steering mechanism model of the steering system, a spring-damper model of the suspension system, and a mass-inertia model of the body system, integrates the simulation models of various systems to form a complete electric vehicle simulation model, and preliminarily verifies the integrated model to check the correctness and stability of the model.

4. The intelligent electric vehicle steering stability prediction method according to claim 3 is characterized in that: The S2 is to accurately configure the simulation model according to the constructed electric vehicle simulation model, including setting vehicle parameters, road conditions and environmental parameters. After the configuration is completed, the simulation model is run to simulate the behavior of the electric vehicle under various driving conditions, including different vehicle speeds, different steering angles and different road conditions, so as to generate rich virtual data, and compare the simulation results with the actual vehicle data to verify the accuracy of the model. If a large deviation is found, the model parameters are adjusted and the simulation is re-run. The virtual data generated during the simulation process is exported, and the virtual data generated by the simulation is merged with the actual collected data to form a more comprehensive and diverse data set.

5. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: The S3 obtains the electric vehicle simulation data according to the electric vehicle simulation model, performs format conversion, removes invalid data, outliers and denoises the data, and aligns the data to the same time axis through an interpolation method.

6. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: S4, extracting key features for analyzing steering stability through the pre-processed data, including comprehensive sideslip index, dynamic stability coefficient and steering response sensitivity; The comprehensive sideslip index takes into account vehicle speed, lateral acceleration and steering angle. The formula for the comprehensive sideslip index is: In the formula, a y represents lateral acceleration, v represents vehicle speed, θ represents steering angle, and ∈ represents a constant; Dynamic stability coefficient comprehensive lateral acceleration a y , longitudinal acceleration a x and vehicle speed v, the dynamic stability coefficient is realized by: Steering response sensitivity The steering response sensitivity is calculated through the relationship between the steering angle change rate and the vehicle speed. The implementation formula is: In the formula, dθ / dt represents the rate of change of the steering angle, v represents the vehicle speed, and ∈ represents a constant.

7. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: In S4, the features related to steering stability are analyzed according to the extracted key features. The extracted feature data is assumed to be X and the target variable is y. The target variable is represented as a steering stability index. The correlation between each feature and the target variable is calculated by the Pearson correlation coefficient. The implementation formula is: In the formula, x ij represents the jth eigenvalue of the i-th sample, represents the mean of the jth feature, y i represents the target variable value of the i-th sample, represents the mean of the target variable, n represents the number of samples, and m represents the number of features; Set the correlation coefficient threshold T and select the top k features with the highest absolute correlation coefficient.

8. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: In S5, the steering stability prediction model is trained based on the extracted key features and combined with historical data. Assuming the relationship between the target variable y and the input feature X, the steering stability prediction model is: y=Xβ+∈, In the formula, y represents the target vector, X represents the input eigenvalue, β represents the regression coefficient vector, and ∈ represents the error vector; According to the results of the steering stability prediction model, the optimal regression coefficient β is found, and the implementation formula is: J(β)=(y-Xβ) T (y-Xβ).

9. The intelligent electric vehicle steering stability prediction method according to claim 1 is characterized in that: The S6 inputs the real-time data into the trained steering stability prediction model, and performs real-time analysis in combination with the electric vehicle simulation model, inputs the real-time collected data into the simulation model, simulates the behavior of the vehicle in the current state, analyzes the output results of the simulation model, identifies possible unstable situations, combines the results of the prediction model and the simulation model, and performs a comprehensive evaluation to evaluate the steering stability index of the vehicle in the current state.

10. The intelligent electric vehicle steering stability prediction method according to claim 1, characterized in that: The S7 automatically adjusts and provides feedback based on the steering stability index, continuously monitors the steering stability index calculated by the prediction model and the simulation model, and sets a threshold value to determine whether the current state is unstable or potentially unstable. If it is detected that the steering stability index exceeds the threshold value, a corresponding warning signal is generated, and corresponding adjustments are automatically made in combination with the electronic stability program, and real-time feedback information is provided to the driver to inform him of the current vehicle state.

Images