An optimization control method for an intelligent steering pump of a pure electric vehicle

Abstract

The present invention discloses a pure electric vehicle intelligent steering pump optimization control method, which aims to improve the vehicle's handling performance and reduce energy consumption through intelligent means. The method first collects multiple data such as vehicle speed, steering angle and oil pressure in real time, and uses advanced statistical analysis and machine learning techniques to identify the driver's habits and vehicle needs. A steering pump dynamic model suitable for various driving conditions is established, and its accuracy is verified through actual tests. By designing PID (proportional-integral-differential) control and fuzzy control strategies, the method realizes real-time dynamic adjustment of steering assistance to ensure that the best assistance effect can be provided under various driving conditions. In addition, combined with the data of the navigation system and the front camera, a machine learning algorithm is used to predict the future steering needs of the vehicle, and the assistance adjustment is performed in advance, which enhances the stability and response speed of the vehicle. It improves the driving experience and effectively reduces energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile steering systems, and in particular to an optimization control method for an intelligent steering pump of a pure electric vehicle. Background Art

[0002] With the rapid development of the automobile industry, the requirements for vehicle handling and NVH comfort are getting higher and higher. At present, the steering system of pure electric buses adopts electric hydraulic steering pump power steering.

[0003] The principle of electric hydraulic power steering is that the steering controller inverts the DC voltage of the electric vehicle battery into AC power and provides it to the electric steering pump to make the electric steering pump work. The electric steering pump drives the steering fluid to provide assistance for the hydraulic power steering gear. At present, the control strategy of most steering pumps is that the steering controller only outputs one frequency, so that the electric steering pump runs at a specified maximum speed. This can only ensure the ease of operation at low speeds. However, when the vehicle is driving at high speeds, the steering force of the vehicle is too light, and there are phenomena such as floating, the handling feel is not good, unnecessary energy is consumed, and the steering pump noise is more obvious when the vehicle is idling. Therefore, it is imperative to optimize the control strategy of the steering pump. It is urgent to improve the vehicle's idle NVH comfort and the user's sense of control at high speeds and reduce energy consumption while ensuring low-speed portability. Summary of the invention

[0004] A pure electric vehicle intelligent steering pump optimization control method comprises the following steps:

[0005] S1. Data collection and analysis: Real-time data collection, real-time data collection from sensors: vehicle speed, steering angle, oil pressure, pay attention to the steering angle and oil pressure changes at low speeds when driving at low speeds, slippery roads, and increase data monitoring of wheel slippage. Use statistical analysis and machine learning technology to analyze data and identify driving patterns and needs;

[0006] S2. Dynamic model establishment: Establish a dynamic model of the steering pump and its control system, covering the response characteristics under different driving conditions, and verify the accuracy of the model through actual test data;

[0007] S3. Control strategy design: Model-based control: Design control strategies: PID control, fuzzy control, etc. Use real-time data to adjust control strategies to adapt to different driving conditions;

[0008] S4. Machine learning optimization: Use the data from the vehicle’s navigation system and front camera sensor to build a prediction model and integrate it to get the prediction formula: , P is the working state of the steering pump (required oil pressure or torque), where v Represents the current speed of the vehicle, which affects the steering torque and steering response.k Indicates the curvature of the road, which affects the steering angle and steering force required. k is the inverse of the road curvature, i.e. ,in R is the radius of curvature, d Indicates the distance between the lane line detected by the camera and the lane center, which affects the degree of vehicle deviation and thus affects the steering requirements. i Indicates the rotation angle of the wheel, which affects the vehicle's driving trajectory. M , a , b ,and c It is a model parameter, which is determined by experimental data. The optimal working state of the steering pump is accurately predicted through the formula, so as to adjust the power of the steering pump in advance. The steering power can be properly adjusted before entering the curve to improve the stability of the vehicle.

[0009] S5. Environmental perception: Integrate data from multiple sensors (radar, lidar, camera) to automatically adjust the steering assist according to road conditions (such as wet, dry, rough, etc.). On wet roads, increase the steering assist and adjust the steering pump working mode according to environmental information to adapt to different environmental conditions.

[0010] S6. Augmented reality integrated energy consumption optimization: Display the steering system status and suggestions on the dashboard and HUD, and further optimize the working mode of the steering pump to reduce overall energy consumption while ensuring the driving experience;

[0011] S7. Driving mode selection: Provides different driving modes: Sport, Comfort, Economy, allowing the driver to choose according to their needs, automatically switching modes according to driving environment and operation, improving the overall driving experience;

[0012] S8. System Integration and Testing: Integrate the optimized control strategy into the vehicle's electronic control unit (ECU) and conduct comprehensive tests under various driving conditions to ensure the effectiveness and stability of the system. Continuously optimize the control algorithm based on actual usage feedback and test results, and regularly update the control strategy to adapt to new needs and technological advances.

[0013] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0014] In step S2, a dynamic model of the steering pump and its control system is established, covering the response characteristics under different driving conditions. The specific steps are as follows:

[0015] S21. Requirements analysis and goal definition: clarify the main goals of the model: improve steering accuracy, reduce energy consumption, optimize response time, and define different driving conditions that need to be modeled (such as urban driving, high-speed driving, emergency avoidance);

[0016] S22. System modeling: Decompose the steering pump and its control system into basic components, such as pump, actuator, sensor and controller. According to the physical principles of the system, establish the dynamic equations of each component, which may include fluid dynamics equations, mechanical equations and electrical control equations. Establish the relationship model of the flow, pressure and power of the steering pump. Establish the response characteristic model of the actuator (such as the motor), including its speed, acceleration and load response. Establish the dynamic response model of the controller, including the behavior of the control algorithm (such as PID control).

[0017] S23. Parameter identification: Obtain key parameters of each component through experiments, such as the flow constant and pressure constant of the steering pump, the inertia and friction coefficient of the actuator, etc. Use the experimental data to fit the model parameters to ensure that the model can accurately reflect the characteristics of the actual system;

[0018] S24. Nonlinear and time-varying characteristic modeling: Nonlinear characteristics, taking into account nonlinear factors in the system, such as the nonlinear relationship between pump flow and pressure; time-varying characteristics, modeling the system in a time-varying manner, taking into account the impact of environmental changes and operating conditions on system performance;

[0019] S25. Model verification: Use simulation tools (such as Matlab / Simulink) to simulate the established model to verify the consistency between the dynamic response of the model and the actual system. Real vehicle testing: Test on the actual vehicle to verify the prediction accuracy of the model under different working conditions, compare with the actual measurement data, and adjust the model parameters to improve the accuracy;

[0020] S26. Model optimization: Analyze the error between the model and the actual system, find possible improvement points, adjust model parameters according to the error analysis results, and optimize the accuracy and robustness of the model.

[0021] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0022] In step S3, the real-time data is used to adjust the control strategy to adapt to different driving conditions, as follows:

[0023] After the vehicle is powered on, the vehicle controller reads the position information of the ignition switch to determine whether the high voltage is connected. After it is connected, the vehicle controller reads the speed sensor information to determine the vehicle's driving speed at this moment;

[0024] When the vehicle speed = 0Km / h, the vehicle controller first determines the handbrake on / off information. If the handbrake is not released and the gear is not in the R / D gear, the steering pump will not start; if the handbrake is released and the gear is not in the R / D gear, the vehicle controller determines the load state of the steering pump at this moment by reading the load current information of the steering pump. If the current of the steering pump is <3A, the steering pump runs at 70% of the rated speed. If the current of the steering pump is ≥3A, the steering pump runs at the rated speed:

[0025] When the vehicle speed is 0Km / h<40Km / h, the steering pump runs at the rated speed;

[0026] When the vehicle speed is greater than 40Km / h, the vehicle controller continues to judge the current load at this moment. When the current of the steering pump is less than 3A, the steering pump runs at 70% of the rated speed. When the current of the steering pump is ≥3A, the steering pump runs at the rated speed.

[0027] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0028] In step S4, a prediction model is established using data from the vehicle's navigation system or the front camera sensor to accurately predict the optimal working state of the steering pump, thereby adjusting the power assist of the steering pump in advance. The specific steps are as follows:

[0029] S41. Data collection and preparation: Collect GPS data, vehicle speed, acceleration, steering wheel angle, front road image, lane lines, obstacles, and record the current working status of the steering pump, power assist, oil pressure, and perform data preprocessing: remove noise and outliers to ensure data quality, align data from different sensors to the same timestamp for comprehensive analysis, and extract useful features from the raw data: road curvature, vehicle speed change rate, obstacle distance;

[0030] S42. Feature Engineering: Select the most relevant features according to the prediction target, environmental features: road curvature, lane offset, relative position of the obstacle ahead, driving features: vehicle speed, acceleration, steering wheel angle, steering angular velocity, system status features: current steering pump power, oil pressure, and perform necessary transformations on the selected features: normalization and standardization for model training;

[0031] S43. Prediction model training: Use the collected historical data to train the linear regression prediction model and integrate it to obtain the prediction formula: , P is the working state of the steering pump, where v Indicates the current speed of the vehicle. k Indicates the curvature of the road. k is the inverse of the road curvature, i.e. ,in Ris the radius of curvature, d Indicates the distance between the lane line detected by the camera and the center of the lane. i represents the rotation angle of the wheel, M , a , b ,and c are the model parameters, which are determined through experimental data and validated using an independent dataset to evaluate the generalization ability and accuracy of the model;

[0032] S44. Model evaluation and validation: Use mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) indicators to evaluate the prediction performance of the model, analyze the source of the model prediction error, identify the shortcomings of the model, test the prediction effect of the model in the actual driving environment, including different driving conditions and road types, adjust the model according to the actual vehicle test results, and further optimize the prediction accuracy;

[0033] S45. Prediction and control: Input real-time sensor data into the prediction model to make real-time predictions. According to the prediction results, adjust the power of the steering pump to provide the best steering response. Test the integrated system to ensure the effectiveness and stability of the model in actual application.

[0034] S46. System maintenance and update: Continuously monitor sensor data and model prediction results, identify potential problems, regularly update the model to maintain its accuracy based on new data and changes in the driving environment, collect user feedback, and understand the model's performance in actual driving.

[0035] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0036] In step S6, the working mode of the steering pump is further optimized to reduce the overall energy consumption while ensuring the driving experience. The specific steps are as follows:

[0037] S61. Analyze and evaluate existing systems;

[0038] S611. Data collection: Collect the working data of the existing steering pump, including the power size, oil pressure, steering angle, vehicle speed, and record the driver's operating habits, including steering frequency and steering amplitude;

[0039] S612. Data analysis: Analyze data to identify steering demand patterns and energy consumption patterns under different driving conditions, and evaluate the energy efficiency of the current steering pump and its impact on the driving experience;

[0040] S62. Optimize steering pump control strategy;

[0041] S621. Dynamic adjustment: Dynamically adjust the power assist of the steering pump according to the prediction results, reduce the power assist in low demand situations, and increase the power assist in high demand situations;

[0042] S622. Energy-saving mode: Low-energy mode: Switch to energy-saving mode when low steering demand is predicted to reduce the power consumption of the steering pump. High-efficiency mode: When high steering demand is expected, optimize the working state of the steering pump to provide necessary assistance while reducing unnecessary energy consumption.

[0043] S63. Adjust the steering assist curve: Use sensors to monitor parameters such as steering wheel rotation angle, speed and vehicle status, and use algorithms to calculate the required assist in real time, so that the motor can work on demand and adjust the degree of steering assist according to the vehicle speed, increasing the assist at low speeds and reducing the assist at high speeds. This can maintain a good driving experience and reduce unnecessary energy consumption.

[0044] Beneficial effects of the present invention: Through intelligent control and real-time feedback mechanism, steering can be made more accurate and reliable, and driving safety is improved. By improving the energy efficiency of the steering system, unnecessary energy consumption is reduced. The optimized steering system can provide appropriate assistance under different driving conditions, making driving easier and more comfortable, especially in terms of operating feel and high-precision control when driving at low speeds. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of an optimization control method for an intelligent steering pump for a pure electric vehicle; DETAILED DESCRIPTION

[0046] A pure electric vehicle intelligent steering pump optimization control method comprises the following steps:

[0047] S1. Data collection and analysis: Real-time data collection, real-time data collection from sensors: vehicle speed, steering angle, oil pressure, pay attention to the steering angle and oil pressure changes at low speeds when driving at low speeds, slippery roads, and increase data monitoring of wheel slippage. Use statistical analysis and machine learning technology to analyze data and identify driving patterns and needs;

[0048] S2. Dynamic model establishment: Establish a dynamic model of the steering pump and its control system, covering the response characteristics under different driving conditions, and verify the accuracy of the model through actual test data;

[0049] S3. Control strategy design: Model-based control: Design control strategies: PID control, fuzzy control, etc. Use real-time data to adjust control strategies to adapt to different driving conditions;

[0050] S4. Machine learning optimization: Use the data from the vehicle’s navigation system and front camera sensor to build a prediction model and integrate it to get the prediction formula: , P is the working state of the steering pump (required oil pressure or torque), where v Represents the current speed of the vehicle, which affects the steering torque and steering response. k Indicates the curvature of the road, which affects the steering angle and steering force required. k is the inverse of the road curvature, i.e. ,in R is the radius of curvature, d Indicates the distance between the lane line detected by the camera and the lane center, which affects the degree of vehicle deviation and thus affects the steering requirements. i Indicates the rotation angle of the wheel, which affects the vehicle's driving trajectory. M , a , b ,and c It is a model parameter, which is determined by experimental data. The optimal working state of the steering pump is accurately predicted through the formula, so as to adjust the power of the steering pump in advance. The steering power can be properly adjusted before entering the curve to improve the stability of the vehicle.

[0051] S5. Environmental perception: Integrate data from multiple sensors (radar, lidar, camera) to automatically adjust the steering assist according to road conditions (such as wet, dry, rough, etc.). On wet roads, increase the steering assist and adjust the steering pump working mode according to environmental information to adapt to different environmental conditions.

[0052] S6. Augmented reality integrated energy consumption optimization: Display the steering system status and suggestions on the dashboard and HUD, and further optimize the working mode of the steering pump to reduce overall energy consumption while ensuring the driving experience;

[0053] S7. Driving mode selection: Provides different driving modes: Sport, Comfort, Economy, allowing the driver to choose according to their needs, automatically switching modes according to driving environment and operation, improving the overall driving experience;

[0054] S8. System Integration and Testing: Integrate the optimized control strategy into the vehicle's electronic control unit (ECU) and conduct comprehensive tests under various driving conditions to ensure the effectiveness and stability of the system. Continuously optimize the control algorithm based on actual usage feedback and test results, and regularly update the control strategy to adapt to new needs and technological advances.

[0055] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0056] In step S2, a dynamic model of the steering pump and its control system is established, covering the response characteristics under different driving conditions. The specific steps are as follows:

[0057] S21. Requirements analysis and goal definition: clarify the main goals of the model: improve steering accuracy, reduce energy consumption, optimize response time, and define different driving conditions that need to be modeled (such as urban driving, high-speed driving, emergency avoidance);

[0058] S22. System modeling: Decompose the steering pump and its control system into basic components, such as pump, actuator, sensor and controller. According to the physical principles of the system, establish the dynamic equations of each component, which may include fluid dynamics equations, mechanical equations and electrical control equations. Establish the relationship model of the flow, pressure and power of the steering pump. Establish the response characteristic model of the actuator (such as the motor), including its speed, acceleration and load response. Establish the dynamic response model of the controller, including the behavior of the control algorithm (such as PID control).

[0059] S23. Parameter identification: Obtain key parameters of each component through experiments, such as the flow constant and pressure constant of the steering pump, the inertia and friction coefficient of the actuator, etc. Use the experimental data to fit the model parameters to ensure that the model can accurately reflect the characteristics of the actual system;

[0060] S24. Nonlinear and time-varying characteristic modeling: Nonlinear characteristics, taking into account nonlinear factors in the system, such as the nonlinear relationship between pump flow and pressure; time-varying characteristics, modeling the system in a time-varying manner, taking into account the impact of environmental changes and operating conditions on system performance;

[0061] S25. Model verification: Use simulation tools (such as Matlab / Simulink) to simulate the established model to verify the consistency between the dynamic response of the model and the actual system. Real vehicle testing: Test on the actual vehicle to verify the prediction accuracy of the model under different working conditions, compare with the actual measurement data, and adjust the model parameters to improve the accuracy;

[0062] S26. Model optimization: Analyze the error between the model and the actual system, find possible improvement points, adjust model parameters according to the error analysis results, and optimize the accuracy and robustness of the model.

[0063] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0064] In step S3, the real-time data is used to adjust the control strategy to adapt to different driving conditions, as follows:

[0065] After the vehicle is powered on, the vehicle controller reads the position information of the ignition switch to determine whether the high voltage is connected. After it is connected, the vehicle controller reads the speed sensor information to determine the vehicle's driving speed at this moment;

[0066] When the vehicle speed = 0Km / h, the vehicle controller first determines the handbrake on / off information. If the handbrake is not released and the gear is not in the R / D gear, the steering pump will not start; if the handbrake is released and the gear is not in the R / D gear, the vehicle controller determines the load state of the steering pump at this moment by reading the load current information of the steering pump. If the current of the steering pump is <3A, the steering pump runs at 70% of the rated speed. If the current of the steering pump is ≥3A, the steering pump runs at the rated speed:

[0067] When the vehicle speed is 0Km / h<40Km / h, the steering pump runs at the rated speed;

[0068] When the vehicle speed is greater than 40Km / h, the vehicle controller continues to judge the current load at this moment. When the current of the steering pump is less than 3A, the steering pump runs at 70% of the rated speed. When the current of the steering pump is ≥3A, the steering pump runs at the rated speed.

[0069] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0070] In step S4, a prediction model is established using data from the vehicle's navigation system or the front camera sensor to accurately predict the optimal working state of the steering pump, thereby adjusting the power assist of the steering pump in advance. The specific steps are as follows:

[0071] S41. Data collection and preparation: Collect GPS data, vehicle speed, acceleration, steering wheel angle, front road image, lane lines, obstacles, and record the current working status of the steering pump, power assist, oil pressure, and perform data preprocessing: remove noise and outliers to ensure data quality, align data from different sensors to the same timestamp for comprehensive analysis, and extract useful features from the raw data: road curvature, vehicle speed change rate, obstacle distance;

[0072] S42. Feature Engineering: Select the most relevant features according to the prediction target, environmental features: road curvature, lane offset, relative position of the obstacle ahead, driving features: vehicle speed, acceleration, steering wheel angle, steering angular velocity, system status features: current steering pump power, oil pressure, and perform necessary transformations on the selected features: normalization and standardization for model training;

[0073] S43. Prediction model training: Use the collected historical data to train the linear regression prediction model and integrate it to obtain the prediction formula: , P is the working state of the steering pump, where v Indicates the current speed of the vehicle. k Indicates the curvature of the road. k is the inverse of the road curvature, i.e. ,in R is the radius of curvature, d Indicates the distance between the lane line detected by the camera and the center of the lane. i represents the rotation angle of the wheel, M , a , b ,and c are the model parameters, which are determined through experimental data and validated using an independent dataset to evaluate the generalization ability and accuracy of the model;

[0074] S44. Model evaluation and validation: Use mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) indicators to evaluate the prediction performance of the model, analyze the source of the model prediction error, identify the shortcomings of the model, test the prediction effect of the model in the actual driving environment, including different driving conditions and road types, adjust the model according to the actual vehicle test results, and further optimize the prediction accuracy;

[0075] S45. Prediction and control: Input real-time sensor data into the prediction model to make real-time predictions. According to the prediction results, adjust the power of the steering pump to provide the best steering response. Test the integrated system to ensure the effectiveness and stability of the model in actual application.

[0076] S46. System maintenance and update: Continuously monitor sensor data and model prediction results, identify potential problems, regularly update the model to maintain its accuracy based on new data and changes in the driving environment, collect user feedback, and understand the model's performance in actual driving.

[0077] Furthermore, a pure electric vehicle intelligent steering pump optimization control method is provided.

[0078] In step S6, the working mode of the steering pump is further optimized to reduce the overall energy consumption while ensuring the driving experience. The specific steps are as follows:

[0079] S61. Analyze and evaluate existing systems;

[0080] S611. Data collection: Collect the working data of the existing steering pump, including the power size, oil pressure, steering angle, vehicle speed, and record the driver's operating habits, including steering frequency and steering amplitude;

[0081] S612. Data analysis: Analyze data to identify steering demand patterns and energy consumption patterns under different driving conditions, and evaluate the energy efficiency of the current steering pump and its impact on the driving experience;

[0082] S62. Optimize steering pump control strategy;

[0083] S621. Dynamic adjustment: Dynamically adjust the power assist of the steering pump according to the prediction results, reduce the power assist in low demand situations, and increase the power assist in high demand situations;

[0084] S622. Energy-saving mode: Low-energy mode: Switch to energy-saving mode when low steering demand is predicted to reduce the power consumption of the steering pump. High-efficiency mode: When high steering demand is expected, optimize the working state of the steering pump to provide necessary assistance while reducing unnecessary energy consumption.

[0085] S63. Adjust the steering assist curve: Use sensors to monitor parameters such as steering wheel rotation angle, speed and vehicle status, and use algorithms to calculate the required assist in real time, so that the motor can work on demand and adjust the degree of steering assist according to the vehicle speed, increasing the assist at low speeds and reducing the assist at high speeds. This can maintain a good driving experience and reduce unnecessary energy consumption.

Claims

1. A pure electric vehicle intelligent steering pump optimization control method, characterized in that: The steps include: S1. Data collection and analysis: Collect data from sensors in real time: vehicle speed, steering angle, oil pressure, steering angle and oil pressure changes at low speed, and add data monitoring of wheel slippage. Use statistical analysis methods to analyze data and identify driving patterns and needs. S2. Dynamic model establishment: Establish a dynamic model of the steering pump and its control system, covering the response characteristics under different driving conditions, and verify the accuracy of the model through actual test data; S3. Control strategy design: Model-based control, design control strategies: PID control, fuzzy control, use real-time data to adjust control strategies to adapt to different driving conditions; S4. Machine learning optimization: Use the data from the vehicle’s navigation system and front camera sensor to build a prediction model and integrate it to get the prediction formula: , P is the working state of the steering pump, where v Indicates the current speed of the vehicle. k Indicates the curvature of the road. k is the inverse of the road curvature, i.e. ,in R is the radius of curvature, d Indicates the distance between the lane line detected by the camera and the center of the lane. θ represents the rotation angle of the wheel, M , a , b ,and c It is a model parameter, which is determined by experimental data. The optimal working state of the steering pump is predicted by the formula, so as to adjust the power of the steering pump in advance. The steering power can be adjusted appropriately before entering the curve to improve the stability of the vehicle. S5. Environmental perception: Integrate data from multiple sensors, including radar, lidar, and cameras, and automatically adjust the steering assist according to road conditions: wet, dry, and rough. On wet roads, increase the steering assist and adjust the steering pump's operating mode according to environmental information to adapt to different environmental conditions. S6. Augmented reality integrated energy consumption optimization: Display the steering system status and suggestions on the dashboard and HUD, and further optimize the working mode of the steering pump to reduce overall energy consumption while ensuring the driving experience; S7. Driving mode selection: Provides different driving modes: Sport, Comfort, Economy, allowing the driver to choose according to their needs, automatically switching modes according to driving environment and operation, improving the overall driving experience; S8. System Integration and Testing: Integrate the optimized control strategy into the vehicle's electronic control unit ECU and test it under various driving conditions to ensure the effectiveness and stability of the system. Continuously optimize the control algorithm based on actual usage feedback and test results, and regularly update the control strategy to adapt to new needs and technological advances.

2. The method for optimizing and controlling a pure electric vehicle intelligent steering pump according to claim 1, characterized in that: In step S2, a dynamic model of the steering pump and its control system is established, covering the response characteristics under different driving conditions. The specific steps are as follows: S21. Requirements analysis and goal definition: clarify the main goals of the model: improve steering accuracy, reduce energy consumption, optimize response time, and define different driving conditions that need to be modeled: urban driving, high-speed driving, and emergency avoidance; S22. System modeling: Decompose the steering pump and its control system into basic components: pump, actuator, sensor and controller. According to the physical principles of the system, establish the dynamic equations of each component, including fluid dynamics equations, mechanical equations and electrical control equations; S23. Parameter identification: obtain the key parameters of each component through experiments: flow constant, pressure constant of steering pump, inertia and friction coefficient of actuator, and fit the model parameters using experimental data; S24. Nonlinear and time-varying characteristics modeling: Nonlinear characteristics, considering the nonlinear factors in the system, the nonlinear relationship between pump flow and pressure, time-varying characteristics, time-varying modeling of the system, considering the impact of environmental changes and working conditions on system performance; S25. Model verification: Use the simulation tool Matlab to simulate the established model to verify the consistency between the dynamic response of the model and the actual system, and test it on the actual vehicle to verify the prediction accuracy of the model under different working conditions, compare it with the actual measurement data, and adjust the model parameters to improve the accuracy; S26. Model optimization: Analyze the error between the model and the actual system, find out the improvement points, adjust the model parameters according to the error analysis results, and optimize the accuracy and robustness of the model.

3. The method for optimizing and controlling a pure electric vehicle intelligent steering pump according to claim 1, characterized in that: In step S3, the real-time data is used to adjust the control strategy to adapt to different driving conditions. The control strategy is as follows: After the vehicle is powered on, the vehicle controller reads the position information of the ignition switch to determine whether the high voltage is connected. After it is connected, the vehicle controller reads the speed sensor information to determine the vehicle's driving speed at this moment; When the vehicle speed = 0Km / h, the vehicle controller first determines the handbrake on / off information. When the handbrake is not released and the gear is not in the R / D gear, the steering pump will not start; when the handbrake is released and the gear is not in the R / D gear, the vehicle controller determines the load state of the steering pump at this moment by reading the load current information of the steering pump. If the current of the steering pump is <3A, the steering pump runs at 70% of the rated speed. If the current of the steering pump is ≥3A, the steering pump runs at a fixed speed: When the vehicle speed is 0Km / h<40Km / h, the steering pump runs at the rated speed; When the vehicle speed is greater than 40Km / h, the vehicle controller continues to judge the current load at this moment. When the current of the steering pump is less than 3A, the steering pump runs at 70% of the rated speed. When the current of the steering pump is ≥3A, the steering pump runs at the rated speed.

4. The method for optimizing and controlling a pure electric vehicle intelligent steering pump according to claim 1, characterized in that: In step S4, a prediction model is established using data from the vehicle's navigation system or the front camera sensor to accurately predict the optimal working state of the steering pump, thereby adjusting the power assist of the steering pump in advance. The specific steps are as follows: S41. Data collection and preparation: Collect GPS data, vehicle speed, acceleration, steering wheel angle, front road image, lane lines, obstacles, and record the current working status of the steering pump, power assist, oil pressure, and perform data preprocessing: remove noise and outliers to ensure data quality, align data from different sensors to the same timestamp for comprehensive analysis, and extract useful features from the raw data: road curvature, vehicle speed change rate, obstacle distance; S42. Feature Engineering: Select the most relevant features according to the prediction target, environmental features: road curvature, lane offset, relative position of the obstacle ahead, driving features: vehicle speed, acceleration, steering wheel angle, steering angular velocity, system status features: current steering pump power, oil pressure, and perform necessary transformations on the selected features: normalization and standardization; S43. Prediction model training: Use the collected historical data to train the linear regression prediction model and integrate it to obtain the prediction formula: , P is the working state of the steering pump, where v Indicates the current speed of the vehicle. k Indicates the curvature of the road. k is the inverse of the road curvature, i.e. ,in R is the radius of curvature, d Indicates the distance between the lane line detected by the camera and the center of the lane. θ represents the rotation angle of the wheel, M , a , b ,and c are the model parameters, which are determined through experimental data and validated using an independent dataset to evaluate the generalization ability and accuracy of the model; S44. Model evaluation and validation: Use mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) to evaluate the prediction performance of the model, analyze the source of the model prediction error, identify the shortcomings of the model, test the prediction effect of the model in the actual driving environment, including different driving conditions and road types, adjust the model according to the actual vehicle test results, and further optimize the prediction accuracy; S45. Prediction and control: Input real-time sensor data into the prediction model to make real-time predictions. According to the prediction results, adjust the power of the steering pump to provide the best steering response. Test the integrated system to ensure the effectiveness and stability of the model in actual application. S46. System maintenance and update: Continuously monitor sensor data and model prediction results, identify potential problems, regularly update the model to maintain its accuracy based on new data and changes in the driving environment, collect user feedback, and understand the model's performance in actual driving.

5. The method for optimizing and controlling a pure electric vehicle intelligent steering pump as claimed in claim 1, characterized in that: In step S6, the working mode of the steering pump is further optimized to reduce the overall energy consumption while ensuring the driving experience. The specific steps are as follows: S61. Analyze and evaluate existing systems; S611. Data collection: Collect the working data of the existing steering pump, including the power size, oil pressure, steering angle, vehicle speed, and record the driver's operating habits, including steering frequency and steering amplitude; S612. Data analysis: Analyze data to identify steering demand patterns and energy consumption patterns under different driving conditions, and evaluate the energy efficiency of the current steering pump and its impact on the driving experience; S62. Optimize steering pump control strategy; S621. Dynamic adjustment: Dynamically adjust the power assist of the steering pump according to the prediction results, reduce the power assist in low demand situations, and increase the power assist in high demand situations; S622. Energy-saving mode: Low-energy mode: Switch to energy-saving mode when low steering demand is predicted to reduce the power consumption of the steering pump. High-efficiency mode: When high steering demand is expected, optimize the working state of the steering pump to provide the necessary assistance. S63. Adjust the steering assist curve: Use sensors to monitor the steering wheel rotation angle, speed and vehicle status parameters, and use algorithms to calculate the required assist in real time, so that the motor can work as needed and adjust the degree of steering assist according to the vehicle speed, increasing the assist at low speeds and reducing the assist at high speeds.

Images