A passenger car braking energy recovery control method and system based on series control

Abstract

The present invention provides a bus braking energy recovery control method and system based on series control. The method monitors the brake pedal position and four-wheel speed through a linear Hall sensor and a magnetoelectric wheel speed sensor, and collects battery status data through a distributed battery status monitoring system. The Kalman filter is used to process the data to detect abnormal values, and a braking demand prediction model combining LSTM and GNN is constructed. The braking demand is predicted according to the real-time driving state. A non-dominated sorting genetic algorithm II is used for multi-objective dynamic optimization to determine the optimal braking torque distribution strategy. An adaptive slope calculation model is established according to the real-time load and road friction coefficient. The coasting brake intervention and exit rates are dynamically adjusted, a hierarchical energy recovery control strategy is established, and a priority mechanism is set to ensure that the optimal energy recovery strategy is selected when multiple conditions are met. In addition, a forced exit mechanism is set for situations such as activation of the active safety system and excessive battery temperature to ensure safety.

Description

Technical Field

[0001] The invention relates to the technical field of passenger car braking, and in particular to a passenger car braking energy recovery control method and system based on series control. Background Art

[0002] With the development of new energy vehicle technology, brake energy recovery technology has become one of the key technologies to improve vehicle energy efficiency and extend cruising range. As an important carrier of urban transportation, the optimization of the brake energy recovery system of buses is of great significance to energy conservation and emission reduction. Traditional brake energy recovery systems mostly adopt rule-based control strategies or single-objective optimization methods, but there are still significant deficiencies in adaptability under complex working conditions, and the balance between energy recovery efficiency and driving comfort.

[0003] Traditional methods usually use a single sensor signal for braking control and lack a multi-source data fusion mechanism. For example, the wheel speed signal is easily affected by road interference, resulting in error accumulation, and battery status monitoring is not dynamically associated with braking demand. Existing methods mostly use a fixed threshold method to handle data anomalies, which cannot effectively deal with noise interference under dynamic conditions, resulting in a decrease in the reliability of the control strategy.

[0004] Existing prediction models are mostly based on linear regression of historical vehicle speed or brake pedal travel, without considering the dynamic impact of the vehicle's surrounding environment on driving behavior. The mainstream torque distribution method uses static weights or heuristic rules, which makes it difficult to take into account the real-time balance of energy recovery efficiency, motor efficiency and acceleration change rate.

[0005] Existing coasting brakes mostly use fixed slope control, which cannot adapt to load changes and fluctuations in road adhesion coefficient. The energy recovery classification standards are mostly based on a single parameter and are not deeply coupled with the active safety system. Summary of the invention

[0006] The purpose of the present invention is to provide a passenger car braking energy recovery control method and system based on series control.

[0007] The problem to be solved by the present invention is to solve the problems of low energy recovery efficiency, unreasonable braking force distribution and unsafe system response during the braking process of a bus. By integrating multiple sensor data and applying Kalman filtering for preprocessing, the braking demand is judged in advance in combination with a braking prediction model, and the optimal braking force distribution is achieved using a multi-objective dynamic optimization algorithm. At the same time, the coordinated work of the motor and the hydraulic system is considered, the sliding braking strategy is adjusted based on the real-time load and road friction coefficient, and a hierarchical control mechanism is established to adapt to different driving conditions, ensuring driving safety while improving the energy recovery efficiency.

[0008] A passenger car braking energy recovery control method based on series control, the technical scheme adopted is as follows:

[0009] S1: A linear Hall sensor is used to monitor the brake pedal position, a magnetoelectric wheel speed sensor is used to monitor the four-wheel speed and the wheel speed difference is compensated in combination with the anti-lock braking system signal, and a distributed battery status monitoring system is used to collect battery status data, including state of charge, health status and temperature status. Kalman filtering is applied to process sensor data and battery status data for abnormal value detection;

[0010] S2: Establish a braking prediction model to predict the braking demand of the bus, and combine the processed sensor data and battery status data under real-time driving status to predict the braking demand of the bus;

[0011] S3: Establish a multi-objective dynamic optimization function combined with non-dominated sorting genetic algorithm II to distribute braking torque. The multi-objectives are to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration change rate. The constraints are total braking force conservation, motor temperature limit and hydraulic response delay. The adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed and road adhesion coefficient, and the solution in the optimization function is selected according to the adaptive weight.

[0012] S4: An adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery.

[0013] S5: Establish a hierarchical energy recovery control strategy, set the levels according to the bus speed, battery charge state, and braking demand, set the energy recovery control priority mechanism, and when multiple trigger conditions are met, trigger the energy recovery control strategy with the highest priority according to the priority mechanism, set forced exit conditions according to the active safety system, battery temperature status and vehicle speed status, and adopt emergency strategies.

[0014] Furthermore, in S1, a magnetoelectric wheel speed sensor is used to monitor the speed of four wheels and the wheel speed difference is compensated in combination with the anti-lock braking system signal, and the sensor data and battery status data are processed by Kalman filtering to perform abnormal value detection, including:

[0015] The wheel speed difference between each wheel is calculated, and a threshold range of the wheel speed difference is defined. When the speed difference between a certain wheel and the other wheels is within the threshold range, the anti-lock braking system is activated, and the braking force is reduced in proportion to the wheel speed difference. The proportion of the reduced braking force is the proportion of the wheel speed difference to the normal wheel speed.

[0016] The Kalman filter recursively processes the sensor data and battery status data, and uses the standard deviation criterion to detect outliers. When the received data deviates from the normal range and enters the standard deviation range, it is judged as an outlier, and the data of the previous cycle is used for interpolation and triggering the fault code.

[0017] Furthermore, the braking prediction model is established in S2 to predict the braking demand of the bus, including:

[0018] Extracting feature vectors from sensor data processed by Kalman filtering and battery status data, including vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency;

[0019] Building a Braking Prediction Model ,in is the feature vector at time point t, A is the adjacency matrix, which is constructed by monitoring the spatial relationship information of the vehicle's surrounding environment using radar data, LSTM is a neural network that processes sequence data, and GNN is a neural network that processes graph structure data. is the prediction result at time point t;

[0020] Extract feature vectors from processed sensor data and battery status data under real-time driving conditions and input the feature vectors to , use LSTM network to process the input feature vector, extract the time series features, use GNN combined with adjacency matrix A to analyze the environment around the vehicle, and output the braking demand at time point t .

[0021] Furthermore, in S3, a multi-objective dynamic optimization function is established to distribute the braking torque in combination with a non-dominated sorting genetic algorithm II, and an adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed, and road adhesion coefficient, and a solution in the optimization function is selected according to the adaptive weight, including:

[0022] S31: Establish multiple objectives to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration rate of change. Maximize energy recovery efficiency by minimizing hydraulic braking torque The motor efficiency is The acceleration rate is ;

[0023] S32: The constraints are total braking force conservation, motor temperature limit and hydraulic response delay. Equal to the braking force provided by the motor Plus the braking force provided by the hydraulic system , the motor temperature limit is , is the maximum permissible torque calculated based on the current motor temperature, and the hydraulic response delay is the rate of change of the hydraulic braking torque;

[0024] S33: Construct a multi-objective optimization function as , the constraints are , , ,in For time Changes in hydraulic braking force within the

[0025] S34: A set of optimal solutions is calculated using a non-dominated sorting genetic algorithm II combined with a multi-objective optimization function, which represents different trade-offs between maximizing energy recovery efficiency, motor efficiency, and minimizing acceleration change rate;

[0026] S35: Calculate adaptive weights in real time based on battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient and , , ,in They are the influencing factors of battery state of charge, battery temperature, road adhesion coefficient, and vehicle speed. is the maximum state of charge of the battery, is the current battery state of charge, is the current battery temperature, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the maximum speed allowed under current driving conditions, is the current vehicle speed;

[0027] S36: Use the calculated weights and Calculate the score of each solution in a set of optimal solutions , ,in They are the objective function values ​​corresponding to energy recovery efficiency, motor efficiency and acceleration change rate respectively. The solution with the highest score is selected as the final torque distribution strategy, and the selected solution is converted into a specific torque distribution instruction, which is sent to the motor controller and hydraulic braking system to execute the corresponding braking force distribution.

[0028] Furthermore, in S4, an adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the sliding brake, including:

[0029] S41: Establishing an adaptive slope calculation model , is the calculated adaptive slope value, is the basic slope, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the mass of the current vehicle, is the mass of the vehicle when unloaded;

[0030] S42: After the driver releases the accelerator pedal, the calculated Increase the coasting braking torque. Coasting braking is when the vehicle speed is greater than 5 km / h and the anti-lock braking system is not triggered. When energy recovery is activated, the calculated Reduce the coasting braking torque. The activation conditions are that the vehicle speed is greater than 20km / h and the battery charge state is less than 90%.

[0031] Furthermore, in S5, an energy recovery hierarchical control strategy is established, an energy recovery control priority mechanism is set, and a forced exit condition is set according to the active safety system, battery temperature status and vehicle speed status and an emergency strategy is adopted, including:

[0032] Three levels are set according to the bus speed, battery state of charge and braking demand. Level 1 is when the speed is greater than 20km / h and the battery state of charge is less than 90%, and 20kw power is used for energy recovery. Level 2 is when any of the following conditions is met: braking demand is greater than 50% and battery state of charge is less than 70%, and 50kw power is used for energy recovery. Level 3 is when the emergency braking deceleration is greater than 0.4g, and 100kw power is used for energy recovery. Level 1 corresponds to low priority, level 2 corresponds to medium priority, and level 3 corresponds to high priority.

[0033] When multiple trigger conditions are met at the same time, the energy recovery control strategy with the highest trigger priority is selected according to a pre-set priority mechanism;

[0034] When the active safety system is detected to be activated, the energy recovery mode is immediately forced to exit. When the battery temperature exceeds the set upper limit, energy recovery is exited and measures are taken to cool the vehicle. When the vehicle speed drops below 5km / h, the energy recovery mode is exited and traditional braking is used instead.

[0035] Furthermore, a bus braking energy recovery control system based on tandem control is used to implement any of the above-mentioned bus braking energy recovery control methods based on tandem control, and the bus braking energy recovery control system based on tandem control includes: a data acquisition and preprocessing module, a braking demand prediction module, a multi-objective dynamic optimization module, an adaptive slope adjustment module, and an energy recovery hierarchical control module:

[0036] Data acquisition and preprocessing module: linear Hall sensors are used to monitor the brake pedal position, magnetoelectric wheel speed sensors are used to monitor the four-wheel speed, and the wheel speed difference is compensated in combination with the anti-lock braking system signal. The distributed battery status monitoring system is used to collect the battery's state of charge, health status and temperature status. Kalman filtering is used to process sensor data and battery status data, and abnormal value detection and data smoothing are performed;

[0037] Braking demand prediction module: Extracts the vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency feature vectors from the processed sensor data, and uses LSTM and GNN combined with the adjacency matrix A to build a braking prediction model. LSTM processes time series features, and GNN analyzes the environment around the vehicle to output the prediction results;

[0038] Multi-objective dynamic optimization module: The objectives are defined as minimizing hydraulic braking torque, maximizing motor efficiency, and minimizing acceleration change rate. Constraints are set including conservation of total braking force, motor temperature limit, and hydraulic response delay. The non-dominated sorting genetic algorithm II is used to solve the multi-objective optimization problem and generate the optimal solution set. The weights are adjusted in real time based on the current battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient, and the optimal solution is selected accordingly.

[0039] Adaptive slope adjustment module: an adaptive slope calculation model is established based on the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery.

[0040] Energy recovery hierarchical control module: Sets energy recovery strategy based on bus speed, battery charge state, and braking demand factors, determines priority mechanism, and sets three energy recovery levels. Each level corresponds to different trigger conditions and power ranges. When multiple trigger conditions are met at the same time, the highest priority energy recovery control strategy is selected according to the pre-set priority. Emergency measures are taken to force exit from energy recovery mode based on activation of active safety system, battery temperature exceeding the upper limit, and vehicle speed dropping below the threshold.

[0041] The beneficial effects of the present invention are as follows: through the braking demand prediction model and the multi-objective dynamic optimization function, the future braking demand is predicted, and the optimal braking force distribution is performed accordingly to maximize the energy recovery efficiency; the adaptive slope calculation model dynamically adjusts the intervention and exit rates of the sliding brake according to the real-time load and the road friction coefficient, so that energy recovery can be effectively performed under different driving conditions;

[0042] The non-dominated sorting genetic algorithm II is used to solve the multi-objective optimization problem and find the balance point between multiple conflicting objectives, providing an efficient solution for complex braking torque distribution.

[0043] When the anti-lock braking system, electronic stability control system and other active safety systems are detected to be activated, the energy recovery mode will be forced to exit immediately to ensure the stability of the vehicle in emergency situations;

[0044] By minimizing the acceleration change rate, the system ensures smooth braking, reduces discomfort caused by sudden deceleration, and improves passenger comfort. It also provides a more personalized response strategy based on the driver's operating habits and current road conditions, thus enhancing the driving experience.

[0045] Establish a clear energy recovery hierarchical control strategy and set a corresponding priority mechanism so that the best decision can be made quickly when multiple trigger conditions are met at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a bus braking energy recovery control method based on series control;

[0047] Figure 2 This is a module diagram of a passenger car braking energy recovery control system based on series control. DETAILED DESCRIPTION

[0048] The present invention is further clearly and completely described below, but the protection scope of the present invention is not limited thereto.

[0049] A passenger car braking energy recovery control method based on series control, the technical scheme adopted is as follows:

[0050] S1: A linear Hall sensor is used to monitor the brake pedal position, a magnetoelectric wheel speed sensor is used to monitor the four-wheel speed and the wheel speed difference is compensated in combination with the anti-lock braking system signal, and a distributed battery status monitoring system is used to collect battery status data, including state of charge, health status and temperature status. Kalman filtering is applied to process sensor data and battery status data for abnormal value detection;

[0051] S2: Establish a braking prediction model to predict the braking demand of the bus, and combine the processed sensor data and battery status data under real-time driving status to predict the braking demand of the bus;

[0052] S3: Establish a multi-objective dynamic optimization function combined with non-dominated sorting genetic algorithm II to distribute braking torque. The multi-objectives are to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration change rate. The constraints are total braking force conservation, motor temperature limit and hydraulic response delay. The adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed and road adhesion coefficient, and the solution in the optimization function is selected according to the adaptive weight.

[0053] S4: An adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery.

[0054] S5: Establish a hierarchical energy recovery control strategy, set the levels according to the bus speed, battery charge state, and braking demand, set the energy recovery control priority mechanism, and when multiple trigger conditions are met, trigger the energy recovery control strategy with the highest priority according to the priority mechanism, set forced exit conditions according to the active safety system, battery temperature status and vehicle speed status, and adopt emergency strategies.

[0055] refer to Figure 1 As shown, it is a flow chart of a passenger car braking energy recovery control method based on series control.

[0056] Furthermore, in S1, a magnetoelectric wheel speed sensor is used to monitor the speed of four wheels and the wheel speed difference is compensated in combination with the anti-lock braking system signal, and the sensor data and battery status data are processed by Kalman filtering to perform abnormal value detection, including:

[0057] The wheel speed difference between each wheel is calculated, and a threshold range of the wheel speed difference is defined. When the speed difference between a certain wheel and the other wheels is within the threshold range, the anti-lock braking system is activated, and the braking force is reduced in proportion to the wheel speed difference. The proportion of the reduced braking force is the proportion of the wheel speed difference to the normal wheel speed.

[0058] The Kalman filter recursively processes the sensor data and battery status data, and uses the standard deviation criterion to detect outliers. When the received data deviates from the normal range and enters the standard deviation range, it is judged as an outlier, and the data of the previous cycle is used for interpolation and triggering the fault code.

[0059] Furthermore, the braking prediction model is established in S2 to predict the braking demand of the bus, including:

[0060] Extracting feature vectors from sensor data processed by Kalman filtering and battery status data, including vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency;

[0061] Building a Braking Prediction Model ,in is the feature vector at time point t, A is the adjacency matrix, which is constructed by monitoring the spatial relationship information of the vehicle's surrounding environment using radar data, LSTM is a neural network that processes sequence data, and GNN is a neural network that processes graph structure data. is the prediction result at time point t;

[0062] Extract feature vectors from processed sensor data and battery status data under real-time driving conditions and input the feature vectors to , use LSTM network to process the input feature vector, extract the time series features, use GNN combined with adjacency matrix A to analyze the environment around the vehicle, and output the braking demand at time point t .

[0063] Furthermore, in S3, a multi-objective dynamic optimization function is established to distribute the braking torque in combination with a non-dominated sorting genetic algorithm II, and an adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed, and road adhesion coefficient, and a solution in the optimization function is selected according to the adaptive weight, including:

[0064] S31: Establish multiple objectives to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration rate of change. Maximize energy recovery efficiency by minimizing hydraulic braking torque The motor efficiency is The acceleration rate is ;

[0065] S32: Constraints are total braking force conservation, motor temperature limit and hydraulic response delay. Equal to the braking force provided by the motor Plus the braking force provided by the hydraulic system , the motor temperature limit is , is the maximum permissible torque calculated based on the current motor temperature, and the hydraulic response delay is the rate of change of the hydraulic braking torque;

[0066] S33: Construct a multi-objective optimization function as , the constraints are , , ,in For time Changes in hydraulic braking force within the

[0067] S34: A set of optimal solutions is calculated using a non-dominated sorting genetic algorithm II combined with a multi-objective optimization function, which represents different trade-offs between maximizing energy recovery efficiency, motor efficiency, and minimizing acceleration change rate;

[0068] S35: Calculate adaptive weights in real time based on battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient and , , ,in They are the influencing factors of battery state of charge, battery temperature, road adhesion coefficient, and vehicle speed. is the maximum state of charge of the battery, is the current battery state of charge, is the current battery temperature, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the maximum speed allowed under current driving conditions, is the current vehicle speed;

[0069] S36: Use the calculated weights and Calculate the score of each solution in a set of optimal solutions , ,in They are the objective function values ​​corresponding to energy recovery efficiency, motor efficiency and acceleration change rate respectively. The solution with the highest score is selected as the final torque distribution strategy, and the selected solution is converted into a specific torque distribution instruction, which is sent to the motor controller and hydraulic braking system to execute the corresponding braking force distribution.

[0070] Furthermore, in S4, an adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the sliding brake, including:

[0071] S41: Establishing an adaptive slope calculation model , is the calculated adaptive slope value, is the basic slope, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the mass of the current vehicle, is the mass of the vehicle when unloaded;

[0072] S42: After the driver releases the accelerator pedal, the calculated Increase the coasting braking torque. Coasting braking is when the vehicle speed is greater than 5 km / h and the anti-lock braking system is not triggered. When energy recovery is activated, the calculated Reduce the coasting braking torque. The activation conditions are that the vehicle speed is greater than 20km / h and the battery charge state is less than 90%.

[0073] Furthermore, in S5, an energy recovery hierarchical control strategy is established, an energy recovery control priority mechanism is set, and a forced exit condition is set according to the active safety system, battery temperature status and vehicle speed status and an emergency strategy is adopted, including:

[0074] Three levels are set according to the bus speed, battery state of charge and braking demand. Level 1 is when the speed is greater than 20km / h and the battery state of charge is less than 90%, and 20kw power is used for energy recovery. Level 2 is when any of the following conditions is met: braking demand is greater than 50% and battery state of charge is less than 70%, and 50kw power is used for energy recovery. Level 3 is when the emergency braking deceleration is greater than 0.4g, and 100kw power is used for energy recovery. Level 1 corresponds to low priority, level 2 corresponds to medium priority, and level 3 corresponds to high priority.

[0075] When multiple trigger conditions are met at the same time, the energy recovery control strategy with the highest trigger priority is selected according to a pre-set priority mechanism;

[0076] When the active safety system is detected to be activated, the energy recovery mode is immediately forced to exit. When the battery temperature exceeds the set upper limit, energy recovery is exited and measures are taken to cool the vehicle. When the vehicle speed drops below 5km / h, the energy recovery mode is exited and traditional braking is used instead.

[0077] Furthermore, a bus braking energy recovery control system based on tandem control is used to implement any of the above-mentioned bus braking energy recovery control methods based on tandem control, and the bus braking energy recovery control system based on tandem control includes: a data acquisition and preprocessing module, a braking demand prediction module, a multi-objective dynamic optimization module, an adaptive slope adjustment module, and an energy recovery hierarchical control module:

[0078] Data acquisition and preprocessing module: linear Hall sensors are used to monitor the brake pedal position, magnetoelectric wheel speed sensors are used to monitor the four-wheel speed, and the wheel speed difference is compensated in combination with the anti-lock braking system signal. The distributed battery status monitoring system is used to collect the battery's state of charge, health status and temperature status. Kalman filtering is used to process sensor data and battery status data, and abnormal value detection and data smoothing are performed;

[0079] Braking demand prediction module: Extracts the vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency feature vectors from the processed sensor data, and uses LSTM and GNN combined with the adjacency matrix A to build a braking prediction model. LSTM processes time series features, and GNN analyzes the environment around the vehicle to output the prediction results;

[0080] Multi-objective dynamic optimization module: The objectives are defined as minimizing hydraulic braking torque, maximizing motor efficiency, and minimizing acceleration change rate. Constraints are set including conservation of total braking force, motor temperature limit, and hydraulic response delay. The non-dominated sorting genetic algorithm II is used to solve the multi-objective optimization problem and generate the optimal solution set. The weights are adjusted in real time based on the current battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient, and the optimal solution is selected accordingly.

[0081] Adaptive slope adjustment module: an adaptive slope calculation model is established based on the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery.

[0082] Energy recovery hierarchical control module: Sets energy recovery strategy based on bus speed, battery charge state, and braking demand factors, determines priority mechanism, and sets three energy recovery levels. Each level corresponds to different trigger conditions and power ranges. When multiple trigger conditions are met at the same time, the highest priority energy recovery control strategy is selected according to the pre-set priority. Emergency measures are taken to force exit from energy recovery mode based on activation of active safety system, battery temperature exceeding the upper limit, and vehicle speed dropping below the threshold.

[0083] refer to Figure 2 The figure shows a module diagram of a passenger car braking energy recovery control system based on series control.

[0084] The present invention provides a bus braking energy recovery control method and system based on series control. The method monitors the brake pedal position and four-wheel speed through a linear Hall sensor and a magnetoelectric wheel speed sensor, and collects battery status data through a distributed battery status monitoring system. The Kalman filter is used to process the data to detect abnormal values, and a braking demand prediction model combining LSTM and GNN is constructed. The braking demand is predicted according to the real-time driving state. A non-dominated sorting genetic algorithm II is used for multi-objective dynamic optimization to determine the optimal braking torque distribution strategy. An adaptive slope calculation model is established according to the real-time load and road friction coefficient. The coasting brake intervention and exit rates are dynamically adjusted, a hierarchical energy recovery control strategy is established, and a priority mechanism is set to ensure that the optimal energy recovery strategy is selected when multiple conditions are met. In addition, a forced exit mechanism is set for situations such as activation of the active safety system and excessive battery temperature to ensure safety.

Claims

1. A passenger car braking energy recovery control method based on series control, characterized in that: include: S1: A linear Hall sensor is used to monitor the brake pedal position, a magnetoelectric wheel speed sensor is used to monitor the four-wheel speed and the wheel speed difference is compensated in combination with the anti-lock braking system signal, and a distributed battery status monitoring system is used to collect battery status data, including state of charge, health status and temperature status. Kalman filtering is applied to process sensor data and battery status data for abnormal value detection; S2: Establish a braking prediction model to predict the braking demand of the bus, and combine the processed sensor data and battery status data under real-time driving status to predict the braking demand of the bus; S3: Establish a multi-objective dynamic optimization function combined with non-dominated sorting genetic algorithm II to distribute braking torque. The multi-objectives are to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration change rate. The constraints are total braking force conservation, motor temperature limit and hydraulic response delay. The adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed and road adhesion coefficient, and the solution in the optimization function is selected according to the adaptive weight. S4: An adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery. S5: Establish a hierarchical energy recovery control strategy, set the levels according to the bus speed, battery charge state, and braking demand, set the energy recovery control priority mechanism, and when multiple trigger conditions are met, trigger the energy recovery control strategy with the highest priority according to the priority mechanism, set forced exit conditions according to the active safety system, battery temperature status and vehicle speed status, and adopt emergency strategies.

2. A passenger car braking energy recovery control method based on series control as claimed in claim 1, characterized in that: In the S1, a magnetoelectric wheel speed sensor is used to monitor the speed of the four wheels and the wheel speed difference is compensated in combination with the anti-lock braking system signal. The sensor data and battery status data are processed by Kalman filtering to perform abnormal value detection, including: The wheel speed difference between each wheel is calculated, and a threshold range of the wheel speed difference is defined. When the speed difference between a certain wheel and the other wheels is within the threshold range, the anti-lock braking system is activated, and the braking force is reduced in proportion to the wheel speed difference. The proportion of the reduced braking force is the proportion of the wheel speed difference to the normal wheel speed. The Kalman filter recursively processes the sensor data and battery status data, and uses the standard deviation criterion to detect outliers. When the received data deviates from the normal range and enters the standard deviation range, it is judged as an outlier, and the data of the previous cycle is used for interpolation and triggering the fault code.

3. A passenger car braking energy recovery control method based on tandem control as claimed in claim 1, characterized in that: The braking prediction model is established in S2 to predict the braking demand of the passenger car, including: Extracting feature vectors from sensor data processed by Kalman filtering and battery status data, including vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency; Building a Braking Prediction Model ,in is the feature vector at time point t, A is the adjacency matrix, which is constructed by monitoring the spatial relationship information of the vehicle's surrounding environment using radar data, LSTM is a neural network that processes sequence data, and GNN is a neural network that processes graph structure data. is the prediction result at time point t; Extract feature vectors from processed sensor data and battery status data under real-time driving conditions and input the feature vectors to , use LSTM network to process the input feature vector, extract the time series features, use GNN combined with adjacency matrix A to analyze the environment around the vehicle, and output the braking demand at time point t .

4. A passenger car braking energy recovery control method based on series control as claimed in claim 1, characterized in that: In S3, a multi-objective dynamic optimization function is established in combination with a non-dominated sorting genetic algorithm II to distribute the braking torque, and an adaptive weight is calculated in real time according to the battery state of charge, battery temperature state, vehicle speed, and road adhesion coefficient, and a solution in the optimization function is selected according to the adaptive weight, including: S31: Establish multiple objectives to maximize energy recovery efficiency, maximize motor efficiency and minimize acceleration rate of change. Maximize energy recovery efficiency by minimizing hydraulic braking torque The motor efficiency is The acceleration rate is ; S32: Constraints are total braking force conservation, motor temperature limit and hydraulic response delay. Equal to the braking force provided by the motor Plus the braking force provided by the hydraulic system , the motor temperature limit is , is the maximum permissible torque calculated based on the current motor temperature, and the hydraulic response delay is the rate of change of the hydraulic braking torque; S33: Construct a multi-objective optimization function as , the constraints are , , ,in For time Changes in hydraulic braking force within the S34: A set of optimal solutions is calculated using a non-dominated sorting genetic algorithm II combined with a multi-objective optimization function, which represents different trade-offs between maximizing energy recovery efficiency, motor efficiency, and minimizing acceleration change rate; S35: Calculate adaptive weights in real time based on battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient and , , ,in They are the influencing factors of battery state of charge, battery temperature, road adhesion coefficient, and vehicle speed. is the maximum state of charge of the battery, is the current battery state of charge, is the current battery temperature, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the maximum speed allowed under current driving conditions, is the current vehicle speed; S36: Use the calculated weights and Calculate the score of each solution in a set of optimal solutions , ,in They are the objective function values ​​corresponding to energy recovery efficiency, motor efficiency and acceleration change rate respectively. The solution with the highest score is selected as the final torque distribution strategy, and the selected solution is converted into a specific torque distribution instruction, which is sent to the motor controller and hydraulic braking system to execute the corresponding braking force distribution.

5. A passenger car braking energy recovery control method based on series control as claimed in claim 1, characterized in that: In S4, an adaptive slope calculation model is established according to the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the sliding brake, including: S41: Establishing an adaptive slope calculation model , is the calculated adaptive slope value, is the basic slope, is the real-time road adhesion coefficient, is the standard adhesion coefficient on dry road surface, is the mass of the current vehicle, is the mass of the vehicle when unloaded; S42: After the driver releases the accelerator pedal, the calculated Increase the coasting braking torque. Coasting braking is when the vehicle speed is greater than 5 km / h and the anti-lock braking system is not triggered. When energy recovery is activated, the calculated Reduce the coasting braking torque. The activation conditions are that the vehicle speed is greater than 20km / h and the battery charge state is less than 90%.

6. A passenger car braking energy recovery control method based on tandem control as claimed in claim 1, characterized in that: In S5, an energy recovery hierarchical control strategy is established, an energy recovery control priority mechanism is set, and a forced exit condition is set according to the active safety system, battery temperature status and vehicle speed status and an emergency strategy is adopted, including: Three levels are set according to the bus speed, battery state of charge and braking demand. Level 1 is when the speed is greater than 20km / h and the battery state of charge is less than 90%, and 20kw power is used for energy recovery. Level 2 is when any of the following conditions is met: braking demand is greater than 50% and battery state of charge is less than 70%, and 50kw power is used for energy recovery. Level 3 is when the emergency braking deceleration is greater than 0.4g, and 100kw power is used for energy recovery. Level 1 corresponds to low priority, level 2 corresponds to medium priority, and level 3 corresponds to high priority. When multiple trigger conditions are met at the same time, the energy recovery control strategy with the highest trigger priority is selected according to a pre-set priority mechanism; When the active safety system is detected to be activated, the energy recovery mode is immediately forced to exit. When the battery temperature exceeds the set upper limit, energy recovery is exited and measures are taken to cool the vehicle. When the vehicle speed drops below 5km / h, the energy recovery mode is exited and traditional braking is used instead.

7. A passenger car braking energy recovery control system based on series control, characterized in that: Used to implement a bus braking energy recovery control method based on tandem control as described in any one of claims 1 to 6, the bus braking energy recovery control system based on tandem control comprises: a data acquisition and preprocessing module, a braking demand prediction module, a multi-objective dynamic optimization module, an adaptive slope adjustment module, and an energy recovery hierarchical control module: Data acquisition and preprocessing module: linear Hall sensors are used to monitor the brake pedal position, magnetoelectric wheel speed sensors are used to monitor the four-wheel speed, and the wheel speed difference is compensated in combination with the anti-lock braking system signal. The distributed battery status monitoring system is used to collect the battery's state of charge, health status and temperature status. Kalman filtering is used to process sensor data and battery status data, and abnormal value detection and data smoothing are performed; Braking demand prediction module: Extracts the vehicle speed change rate, brake pedal travel and its change rate, battery status data, and driver emergency braking frequency feature vectors from the processed sensor data, and uses LSTM and GNN combined with the adjacency matrix A to build a braking prediction model. LSTM processes time series features, and GNN analyzes the environment around the vehicle to output the prediction results; Multi-objective dynamic optimization module: The objectives are defined as minimizing hydraulic braking torque, maximizing motor efficiency, and minimizing acceleration change rate. Constraints are set including conservation of total braking force, motor temperature limit, and hydraulic response delay. The non-dominated sorting genetic algorithm II is used to solve the multi-objective optimization problem and generate the optimal solution set. The weights are adjusted in real time based on the current battery state of charge, battery temperature, vehicle speed, and road adhesion coefficient, and the optimal solution is selected accordingly. Adaptive slope adjustment module: an adaptive slope calculation model is established based on the real-time load and the real-time road friction coefficient to dynamically adjust the intervention and exit rates of the coasting brake. After the driver releases the accelerator pedal, the coasting brake torque is increased according to the calculated adaptive slope value during the coasting brake, and the coasting brake torque is reduced according to the calculated adaptive slope value during energy recovery. Energy recovery hierarchical control module: Sets energy recovery strategy based on bus speed, battery charge state, and braking demand factors, determines priority mechanism, and sets three energy recovery levels. Each level corresponds to different trigger conditions and power ranges. When multiple trigger conditions are met at the same time, the highest priority energy recovery control strategy is selected according to the pre-set priority. Emergency measures are taken to force exit from energy recovery mode based on activation of active safety system, battery temperature exceeding the upper limit, and vehicle speed dropping below the threshold.

Images