Detection system based on electric vehicle braking energy recovery device and control method

Abstract

The invention relates to the technical field of electric vehicle braking energy recovery, in particular to a detection system and a control method based on an electric vehicle braking energy recovery device, comprising the following steps: step 1, data acquisition: acquiring data from a vehicle-mounted sensor in real time, the sensor data including vehicle speed v (t), acceleration a (t), braking force F (t), voltage U (t) and current I (t), the sensor data are respectively collected by a vehicle-mounted vehicle speed sensor, an accelerometer, a braking force sensor and an electrical monitoring unit; 2, preprocessing state estimation is conducted on the collected data of the speed v (t), the acceleration a (t), the braking force F (t), the voltage U (t) and the current I (t) of the vehicle, and state estimation is conducted on the collected data through a Kalman filtering algorithm. The hybrid intelligent control module fuses the advantages of adaptive fuzzy PID control and reinforcement learning algorithm, and can dynamically adjust a control strategy according to real-time working conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle braking energy recovery, and specifically to a detection system and control method based on an electric vehicle braking energy recovery device. Background Technique

[0002] In the field of electric vehicles, the braking energy recovery technology has become one of the key means to improve the vehicle's endurance. In the prior art, the braking energy recovery system usually collects vehicle operation data through on-vehicle sensors, such as vehicle speed, acceleration, braking force, etc., and uses simple control algorithms to allocate the proportion of electric braking and mechanical braking;

[0003] However, these systems have some deficiencies. First, the data acquisition range is limited, and only basic vehicle operation parameters can be obtained, lacking comprehensive monitoring of key information such as battery state and motor speed. Second, the control strategy is relatively single, mostly based on fixed proportion allocation or simple threshold judgment, and it is difficult to adapt to complex and changeable driving conditions. In addition, the existing systems lack an effective safety verification mechanism and cannot monitor potential faults in real time during the energy recovery process. Once the system malfunctions, it may affect the normal braking performance of the vehicle, posing certain safety hazards;

[0004] The existing electric vehicle braking energy recovery system has problems such as incomplete data acquisition, single control strategy, and lack of effective safety verification. Therefore, a detection system and control method based on an electric vehicle braking energy recovery device are proposed for the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a detection system and control method based on an electric vehicle braking energy recovery device to solve the problems raised in the above background technique.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A detection system and control method based on an electric vehicle braking energy recovery device, including the following steps:

[0008] Step 1: Data acquisition: Real-time data is collected from on-vehicle sensors. The sensor data includes vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t), and current I(t). The sensor data is collected by an on-vehicle vehicle speed sensor, an accelerometer, a braking force sensor, and an electrical monitoring unit respectively;

[0009] Step 2: Perform preprocessing state estimation on the collected vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t), and current I(t) data: Use the Kalman filter algorithm to perform state estimation on the collected data to provide accurate state information to the subsequent intelligent control module;

[0010] Step 3: Intelligent control: Use the hybrid intelligent control module to generate a control signal u(t). The hybrid intelligent control module includes a low-level adaptive fuzzy PID controller and a high-level reinforcement learning control module;

[0011] Step 4: Execution and feedback: Drive the braking energy recovery system according to the control signal u(t) to achieve precise energy recovery during vehicle braking, and feedback the execution results and related data to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module.

[0012] As a further optimized content of the present invention, wherein: in the above step 1, further collect the braking pedal travel amount, motor speed, battery SOC value, and temperature parameters to provide basic data support for subsequent energy recovery control.

[0013] As a further optimized content of the present invention, wherein: the Kalman filter algorithm formula in the above step 2 is:

[0014]

[0015] In the formula, is the state estimate value at time t , P(t|t - 1) is the predicted error covariance matrix, K(t) is the Kalman gain, z(t) is the measurement vector, where the measurement vector data comes from the sensor data including v(t), a(t), F(t), U(t), and I(t), f(·) is the vehicle dynamics model, F is the state transition matrix, H is the observation matrix, Q and R are the process noise and measurement noise covariance matrices respectively, and h(·) is the state measurement function.

[0016] As a further optimized content of the present invention, wherein: the adaptive fuzzy PID control algorithm of the low-level adaptive fuzzy PID controller is:

[0017] e(t) = v set (t) - v actul (t),

[0018] Δe(t) = e(t) - e(t - 1),

[0019]

[0020] In the formula, v set(t) is the target vehicle speed preset by the dynamic test condition generation module, v actul (t) is the actual vehicle speed real-time collected by the on-vehicle vehicle speed sensor, K p (t), K i (t) and K d (t) and K

[0021] As a further optimized content of the present invention, wherein: the reinforcement learning control algorithm of the high-level reinforcement learning control module is:

[0022] For each time step t, a state vector s(t) is formed. The state vector s(t) is composed of the data processed in the first step and the second step, and the Q-learning algorithm is adopted, and its update formula is:

[0023]

[0024] In the formula, α is the learning rate, γ is the discount factor, (t) is the immediate reward, which is calculated based on the braking performance and the energy recovery efficiency, and its data comes from the sensor data real-time collected in the first step, a(t) is the control action;

[0025] The output u PID (t) of the adaptive fuzzy PID controller is fused with the adjustment result of the reinforcement learning module to generate the final control signal u(t).

[0026] As a further optimized content of the present invention, wherein: in the fourth step, it further includes:

[0027] When the energy recovery condition is satisfied, the ratio of the electric machine braking torque to the mechanical braking torque is dynamically allocated based on the maximum battery recovery power to optimize the energy recovery efficiency;

[0028] The voltage fluctuation, temperature change and braking efficiency during the energy recovery process are monitored in real time, and the braking mode is switched when the preset safety threshold is exceeded.

[0029] As a further optimized content of the present invention, wherein: it includes a sensor module, a state estimation module, an execution and feedback module and an execution and feedback module;

[0030] Sensor module: It includes an on-vehicle vehicle speed sensor, an accelerometer, a braking force sensor and an electrical monitoring unit, and is used to collect vehicle speed, acceleration, braking force, voltage and current data in real time;

[0031] State estimation module: The Kalman filtering algorithm is adopted to perform state estimation on the collected data, and is used to provide accurate state information to the subsequent intelligent control module;

[0032] Intelligent control module: It includes a low-level adaptive fuzzy PID controller and a high-level reinforcement learning control module, which are used to generate control signals;

[0033] Execution and feedback module: According to the control signal, it drives the braking energy recovery system to achieve precise energy recovery during vehicle braking, and feeds back the execution result and relevant data to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module.

[0034] As a further optimized content of the present invention, wherein: the sensor module further includes a brake pedal travel sensor, a motor speed sensor, a battery SOC sensor and a temperature sensor, which are used to collect the brake pedal travel amount, motor speed, battery SOC value and temperature parameters.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. In the present invention, by expanding the sensor module, the vehicle operation state data is comprehensively collected, providing a richer and more accurate information source for the intelligent control module. The hybrid intelligent control module combines the advantages of adaptive fuzzy PID control and reinforcement learning algorithms, and can dynamically adjust the control strategy according to the real-time working conditions. Under different road conditions and driving habits, the system can automatically optimize the distribution ratio of electric braking and mechanical braking, thus significantly improving the efficiency and stability of braking energy recovery and effectively extending the driving range of electric vehicles;

[0037] 2. In the present invention, a perfect safety verification and exception handling mechanism is designed. During the energy recovery process, key indicators such as voltage fluctuation, temperature change and braking efficiency are monitored in real time. Once exceeding the preset safety threshold, the braking mode is immediately switched to ensure driving safety. In addition, when a system fault is detected, the energy recovery circuit can be quickly disconnected, the redundant braking force of the mechanical braking system is maintained, and a fault warning is sent to the vehicle controller. These measures effectively reduce the risk of braking failure caused by system faults and improve the overall safety and reliability of the electric vehicle braking energy recovery system;

[0038] 3. In the present invention, based on the learning of the driver's braking habits and the real-time analysis of road conditions, the present invention can dynamically adjust the slope of the torque distribution curve and the smoothing coefficient of the transition interval to achieve a personalized energy recovery strategy. In the urban road conditions with frequent starts and stops, the system can optimize the recovery power, reduce braking impact and improve driving comfort; in the high-speed cruising or downhill sections, the electric braking ratio is reasonably increased to maximize the energy recovery efficiency. This adaptive energy recovery control method not only improves the energy utilization rate, but also provides a more stable and comfortable driving experience for the driver. Brief Description of the Drawings

[0039] Figure 1Flow chart of the detection system and control method based on the braking energy recovery device of an electric vehicle according to the present invention;

[0040] Figure 2 System block diagram of the detection system based on the braking energy recovery device of an electric vehicle according to the present invention. Specific embodiments

[0041] Please refer to Figure 1-2 The present invention provides a technical solution:

[0042] The detection system and control method based on the braking energy recovery device of an electric vehicle include the following steps:

[0043] Step 1: Data acquisition: Real-time data is acquired from in-vehicle sensors. The sensor data includes vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t), and current I(t). The sensor data is acquired by an in-vehicle vehicle speed sensor, an accelerometer, a braking force sensor, and an electrical monitoring unit respectively. Key vehicle operation parameters are acquired in real time to ensure the accuracy and timeliness of the data, providing precise data support for energy recovery;

[0044] Step 2: Preprocessing state estimation of the acquired vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t), and current I(t) data: The Kalman filter algorithm is used to perform state estimation on the acquired data to provide accurate state information to the subsequent intelligent control module. The Kalman filter algorithm is used for state estimation to improve the accuracy of the state information, thereby enhancing the response speed and precision of the control strategy;

[0045] Step 3: Intelligent control: A hybrid intelligent control module is used to generate a control signal u(t). The hybrid intelligent control module includes a low-level adaptive fuzzy PID controller and a high-level reinforcement learning control module. The hybrid intelligent control module is used to generate a control signal, combining the advantages of fuzzy PID control and reinforcement learning to improve the flexibility and adaptability of the control;

[0046] Step 4: Execution and feedback: The braking energy recovery system is driven according to the control signal u(t) to achieve precise energy recovery during vehicle braking, and the execution results and related data are fed back to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module. Through the execution and feedback mechanism, precise control of energy recovery is achieved, the energy utilization efficiency is improved, and at the same time, continuous optimization is performed through the online update platform to enhance the overall performance of the system.

[0047] As a further technical solution for the implementation of this solution, in the first step, the brake pedal travel, motor speed, battery SOC value, and temperature parameters are further collected to provide basic data support for subsequent energy recovery control. Further collecting the brake pedal travel, motor speed, battery SOC value, and temperature parameters provides more comprehensive basic data support for energy recovery control, which helps to more accurately control the energy recovery process and improve the energy recovery efficiency;

[0048] As a further technical solution for the implementation of this solution, the Kalman filter algorithm formula in the second step is as follows:

[0049]

[0050] In the formula, is the state estimate value at time t , P(t|t - 1) is the prediction error covariance matrix, K(t) is the Kalman gain, z(t) is the measurement vector, where the measurement vector data comes from the sensor data including v(t), a(t), F(t), U(t), and I(t), f(·) is the vehicle dynamics model, F is the state transition matrix, H is the observation matrix, Q and R are the process noise and measurement noise covariance matrices respectively, and h(·) is the state measurement function. Using the extended Kalman filter algorithm can better handle the state estimation problem of nonlinear systems, improve the accuracy of state estimation, and thus provide more reliable state information for the intelligent control module, further improving the control accuracy;

[0051] As a further technical solution for the implementation of this solution, the adaptive fuzzy PID control algorithm of the low-level adaptive fuzzy PID controller is as follows:

[0052] e(t) = v set (t) - v actul (t),

[0053] Δe(t) = e(t) - e(t - 1),

[0054]

[0055] In the formula, v set (t) is the target vehicle speed preset by the dynamic test condition generation module, v actul (t) is the actual vehicle speed real-time collected by the vehicle speed sensor, K p (t), K i (t), and K d(t) is dynamically adjusted according to the error e(t) and its change rate Δe(t) through a pre-constructed fuzzy rule base. The adaptive fuzzy PID control algorithm can dynamically adjust the control parameters according to the real-time error and its change rate, improve the adaptability and robustness of the control system, and enable the system to maintain optimal performance under different working conditions;

[0056] As a further technical solution of this scheme, the reinforcement learning control algorithm of the high-level reinforcement learning control module is:

[0057] For each time step t, a state vector s(t) is constructed. The state vector s(t) is composed of the data processed in Step 1 and Step 2, and the Q-learning algorithm is adopted. Its update formula is:

[0058]

[0059] In the formula, α is the learning rate, γ is the discount factor, (t) is the immediate reward, which is calculated based on the braking performance and energy recovery efficiency, and its data comes from the sensor data collected in real time in Step 1. a(t) is the control action;

[0060] Fuse the output u PID (t) of the adaptive fuzzy PID controller with the adjustment result of the reinforcement learning module to generate the final control signal u(t). The reinforcement learning control algorithm optimizes the control strategy through online learning, enabling the system to continuously self-improve according to the actual operating conditions, improve the braking performance and energy recovery efficiency, and enhance the intelligent level of the system;

[0061] As a further technical solution of this scheme, in Step 4, it further includes:

[0062] When the energy recovery condition is met, dynamically allocate the ratio of the electric motor braking torque to the mechanical braking torque based on the maximum battery recovery power to optimize the energy recovery efficiency;

[0063] Real-time monitor the voltage fluctuation, temperature change and braking effectiveness during the energy recovery process. When the preset safety threshold is exceeded, execute the braking mode switch, dynamically allocate the ratio of the electric motor braking torque to the mechanical braking torque, optimize the energy recovery efficiency, and at the same time ensure driving safety through real-time monitoring and safety verification, improving the reliability and safety of the system;

[0064] As a further technical solution of this scheme, it includes a sensor module, a state estimation module, an execution and feedback module, and an execution and feedback module;

[0065] Sensor module: It includes an on-vehicle vehicle speed sensor, an accelerometer, a braking force sensor, and an electrical monitoring unit, which are used to collect vehicle speed, acceleration, braking force, voltage, and current data in real time;

[0066] State estimation module: The Kalman filtering algorithm is used to perform state estimation on the collected data, providing accurate state information to the subsequent intelligent control module;

[0067] Intelligent control module: It includes a low-level adaptive fuzzy PID controller and a high-level reinforcement learning control module, which are used to generate control signals;

[0068] Execution and feedback module: It drives the braking energy recovery system according to the control signal, realizes accurate energy recovery during vehicle braking, and feeds back the execution results and relevant data to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module. It integrates multiple sensors and control modules to achieve comprehensive monitoring and precise control of the electric vehicle braking energy recovery process, improve energy recovery efficiency, reduce energy consumption, and enhance the economy and environmental friendliness of electric vehicles;

[0069] As a further technical solution for the implementation of this solution, the sensor module further includes a brake pedal travel sensor, a motor speed sensor, a battery SOC sensor, and a temperature sensor, which are used to collect the brake pedal travel amount, motor speed, battery SOC value, and temperature parameters. By adding a brake pedal travel sensor, a motor speed sensor, a battery SOC sensor, and a temperature sensor, the system can collect key operating parameters more comprehensively, provide more accurate data support for energy recovery control, and further improve energy recovery efficiency and system performance.

[0070] In this article, specific examples are used to elaborate on the principles and implementation methods of the present invention. The descriptions of the above examples are only used to help understand the method of the present invention and its core idea. The above is only the preferred implementation mode of the present invention. It should be noted that due to the limited nature of written expression and the objectively infinite specific structures, for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements, modifications, or changes can be made, or the above technical features can be combined in an appropriate manner; these improvements, modifications, changes, or combinations, or directly applying the concept and technical solution of the invention to other occasions without improvement, should all be regarded as the protection scope of the present invention.

Claims

1. A control method based on a detection system of an electric vehicle braking energy recovery device, characterized in that: The following steps are involved: Step 1: Data collection: collect data from vehicle-mounted sensors in real time. The sensor data includes vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t) and current I(t). The sensor data are collected by the vehicle-mounted speed sensor, accelerometer, braking force sensor and electrical monitoring unit respectively. Step 2: Pre-process the collected vehicle speed v(t), acceleration a(t), braking force F(t), voltage U(t) and current I(t) data for state estimation: Use the Kalman filter algorithm to perform state estimation on the collected data to provide accurate state information to the subsequent intelligent control module; Step 3: Intelligent control: Generate a control signal u(t) using a hybrid intelligent control module, wherein the hybrid intelligent control module includes a low-level adaptive fuzzy PID controller and a high-level reinforcement learning control module; Step 4: Execution and feedback: Drive the braking energy recovery system according to the control signal u(t) to achieve accurate energy recovery during vehicle braking, and feed back the execution results and related data to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module.

2. The control method of the detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: In the step 1, the brake pedal travel, motor speed, battery SOC value and temperature parameters are further collected to provide basic data support for subsequent energy recovery control.

3. The control method of the detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: The Kalman filter algorithm formula in step 2 is: In the formula, is the state estimate at time t, P(t|t-1) is the prediction error covariance matrix, K(t) is the Kalman gain, z(t) is the measurement vector, where the measurement vector data comes from the sensor data containing v(t), a(t), F(t), U(t) and I(t), f(·) is the vehicle dynamics model, F is the state transfer matrix, H is the observation matrix, Q and R are the process noise and measurement noise covariance matrices respectively, and h(·) is the state measurement function.

4. The control method of the detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: The adaptive fuzzy PID control algorithm of the low-level adaptive fuzzy PID controller is: e(t)=v set (t)-v actul (t), Δe(t)=e(t)-e(t-1), In the formula, v set (t) is the target vehicle speed preset by the dynamic test condition generation module, v actul (t) is the actual vehicle speed collected in real time by the vehicle speed sensor, K p (t), K i (t) and K d (t) are dynamically adjusted according to the error e(t) and its rate of change Δe(t) through a pre-built fuzzy rule base.

5. The control method of the detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: The reinforcement learning control algorithm of the high-level reinforcement learning control module is: For each time step t, a state vector s(t) is constructed. The state vector s(t) is composed of the data processed in steps 1 and 2, and the Q-learning algorithm is adopted. The update formula is: Where α is the learning rate, γ is the discount factor, (t) is the immediate reward, which is calculated based on the braking performance and energy recovery efficiency. Its data comes from the sensor data collected in real time in step 1, and a(t) is the control action; The adaptive fuzzy PID controller output u PID (t) is fused with the adjustment result of the reinforcement learning module to generate the final control signal u(t).

6. The control method of the detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: The step 4 further includes: When the energy recovery conditions are met, the ratio of the motor braking torque to the mechanical braking torque is dynamically allocated based on the maximum battery recovery power to optimize the energy recovery efficiency; Real-time monitoring of voltage fluctuations, temperature changes and braking performance during energy recovery, and switching of braking modes when the preset safety threshold is exceeded.

7. The detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: It includes a sensor module, a state estimation module, an execution and feedback module, and an execution and feedback module; Sensor module: including vehicle speed sensor, accelerometer, brake force sensor and electrical monitoring unit, used to collect vehicle speed, acceleration, brake force, voltage and current data in real time; State estimation module: uses Kalman filter algorithm to estimate the state of collected data, which is used to provide accurate state information to subsequent intelligent control modules; Intelligent control module: including low-level adaptive fuzzy PID controller and high-level reinforcement learning control module, used to generate control signals; Execution and feedback module: drives the brake energy recovery system according to the control signal to achieve accurate energy recovery during vehicle braking, and feeds back the execution results and related data to the online update platform for offline simulation and continuous optimization of the hybrid intelligent control module.

8. The detection system based on the electric vehicle braking energy recovery device according to claim 1 is characterized in that: The sensor module further includes a brake pedal travel sensor, a motor speed sensor, a battery SOC sensor and a temperature sensor, which are used to collect brake pedal travel, motor speed, battery SOC value and temperature parameters.

Images