Driver behavior-integrated adaptive energy recovery method for electric vehicle and system

Abstract

The present application relates to a driver behavior-integrated adaptive energy recovery method for an electric vehicle and a system. The method comprises the following steps: analyzing current traveling data and historical traveling data of a vehicle, determining a driving style, and predicting the probability of the driving style; estimating a vehicle mass on the basis of the current traveling data of the vehicle, and respectively calculating two values of the gradient by means of two different algorithms; setting a weight on the basis of a preset vehicle speed threshold, and weighting the two values to obtain a fused gradient value; and on the basis of the current traveling data of the vehicle, the vehicle mass, the fused gradient value, and the probability of the driving style, respectively determining corresponding correction factors, correcting an initial braking torque corresponding to a current motor speed to obtain a reference braking torque, comparing the reference braking torque with a preset motor torque range, obtaining an optimal braking torque, and performing braking, so as to achieve energy recovery. The present application effectively optimizes the distribution of regenerative braking energy and improves the energy recovery rate.

Description

An adaptive energy recovery method and system for electric vehicles that incorporates driver behavior Technical Field

[0001] This application relates to the field of vehicle energy recovery technology, specifically to an adaptive energy recovery method and system for electric vehicles that integrates driver behavior. Background Technology

[0002] Traditional vehicle energy recovery technologies mainly rely on electric motor braking to achieve energy recovery. They mostly adopt relatively simple control strategies and fail to fully consider various complex factors during vehicle operation, such as vehicle driving data and road conditions, resulting in low energy recovery efficiency.

[0003] To address this, some energy recovery technologies introduce correction coefficients for factors such as torque, speed, wheel angular velocity, gradient, and vehicle mass. These coefficients, along with two sets of torque diagrams, are used for interpolation to determine the optimal regenerative braking torque. While this approach considers various factors during vehicle operation and improves energy recovery efficiency to some extent, it still has some drawbacks: First, vehicle mass and gradient rely on sensor measurements, and inaccurate data may result from insufficient sensor precision, delayed response, or interference from external environments (such as severe weather or extreme temperatures). Second, the correction coefficients considered only include basic factors like vehicle speed, gradient, and vehicle mass, making them less adaptable to driver habits. Summary of the Invention

[0004] This application addresses the aforementioned problems in the current technology by providing an adaptive energy recovery method and system for electric vehicles that integrates driver behavior, thereby solving the technical problems of inaccurate energy recovery technology parameters and poor driver adaptability.

[0005] To achieve the above objectives, in a first aspect, this application provides a method for adaptive energy recovery of electric vehicles that incorporates driver behavior, the method comprising:

[0006] Analyze the vehicle's current and historical driving data to determine the driving style and predict the probability of that driving style occurring.

[0007] Based on the vehicle's current driving data, the overall vehicle mass is estimated, and two different values ​​of the slope are calculated using two different algorithms.

[0008] The weights are set by a preset vehicle speed threshold, and the fused slope is obtained by weighting the two values.

[0009] By combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles, corresponding correction factors are determined. The initial braking torque corresponding to the current motor speed is corrected to obtain the reference braking torque. The reference braking torque is compared with the preset motor torque range to obtain the optimal braking torque and perform braking to achieve energy recovery.

[0010] Furthermore, in one embodiment, analyzing the vehicle's current driving data and historical driving data to determine the driving style and predict the probability of that driving style occurring includes:

[0011] Preprocessing operations are performed on the driving data, including filtering and normalization.

[0012] The system analyzes current and historical driving data using a support vector machine algorithm to determine the driver's driving style, which includes aggressive, conservative, and economical driving styles. It then predicts the probability of the driver's driving style appearing among all driving styles.

[0013] Furthermore, in one embodiment, the vehicle mass is estimated based on the vehicle's current driving data, and two values ​​of the slope are calculated using two different algorithms, including:

[0014] Based on the vehicle's current driving data, the vehicle mass is estimated using the least squares estimation method, and the two values ​​of the slope are calculated using Kalman filtering and the least squares estimation method, respectively.

[0015] Furthermore, in one embodiment, a weight is set based on a preset vehicle speed threshold, and the fused slope is obtained by weighting the two values, including:

[0016] The vehicle speed threshold is determined based on calibration.

[0017] The weighting coefficient is determined by the critical speed value, and the two values ​​of the slope are weighted to obtain the merged slope.

[0018] Furthermore, in one embodiment, the step of determining corresponding correction factors by combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of the occurrence of a particular driving style includes:

[0019] Query the pre-set correction factor table to obtain the correction factors corresponding to the current vehicle speed, pedal opening, vehicle weight, blending gradient, and the probability of the driving style appearing.

[0020] The correction factor table includes: real-time vehicle speed, pedal opening, vehicle weight, blending gradient, and the probability of the occurrence of driving style, as well as the correction factor corresponding to each probability.

[0021] Furthermore, in one embodiment, the step of correcting the initial braking torque corresponding to the current motor speed to obtain the reference braking torque includes:

[0022] The initial braking torque is obtained from the current motor speed.

[0023] The initial braking torque is multiplied by the required correction factor to obtain the reference braking torque.

[0024] Furthermore, in one embodiment, comparing the reference braking torque with a preset motor torque range to obtain the optimal braking torque includes:

[0025] If the reference braking torque is within the preset motor torque range, the reference braking torque is taken as the optimal braking torque; if the reference braking torque is less than the minimum motor torque, the minimum motor torque is taken as the optimal braking torque. If the reference braking torque is greater than the maximum motor torque, the maximum motor torque is taken as the optimal braking torque.

[0026] Furthermore, in one embodiment, the method further includes limiting the output of the torque signal by controlling the slope change of the braking torque and using filtering techniques before performing braking.

[0027] Furthermore, in one embodiment, the vehicle's driving data includes: vehicle speed, acceleration, brake pedal opening, vehicle attitude information, and remaining battery power.

[0028] Secondly, based on the above-mentioned trolley adaptive energy recovery method that integrates driver behavior, this application provides an energy recovery system that integrates the trolley adaptive energy recovery method that integrates driver behavior, the energy recovery system comprising:

[0029] The analysis module is used to analyze the vehicle's current driving data and historical driving data to determine the driving style and predict the probability of that driving style occurring.

[0030] The estimation module is used to estimate the vehicle's mass based on the vehicle's current driving data and to calculate two values ​​of the slope using two different algorithms.

[0031] The weighting module is used to set weights based on preset vehicle speed thresholds and obtain the fused slope by weighting the two values.

[0032] The energy recovery module combines the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles to determine corresponding correction factors. It then corrects the initial braking torque corresponding to the current motor speed to obtain a reference braking torque. By comparing the reference braking torque with a preset motor torque range, it obtains the optimal braking torque and performs braking to achieve energy recovery.

[0033] The beneficial effects of the technical solutions provided in this application include:

[0034] This application analyzes the vehicle's current and historical driving data to determine the driving style and predict the probability of that driving style occurring. Based on the driving style and its probability of occurrence, the braking response is adjusted to match the individual's driving habits, thereby improving the adaptability of energy recovery technology to the driver's driving habits.

[0035] Based on the vehicle's current driving data, the overall vehicle mass is estimated, and two different slope values ​​are calculated using two different algorithms. The two values ​​are weighted by a preset vehicle speed threshold to obtain a fused slope, which reduces reliance on sensors and improves data accuracy and the reliability of the energy recovery system.

[0036] By combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles, corresponding correction factors are determined. The initial braking torque corresponding to the current motor speed is corrected to obtain the reference braking torque. The reference braking torque is compared with the preset motor torque range to obtain the optimal braking torque and perform braking to achieve energy recovery. This effectively optimizes the distribution of regenerative braking energy and improves the energy recovery rate by 5%-10%. Attached Figure Description

[0037] Figure 1 is a flowchart of the electric vehicle adaptive energy recovery method that incorporates driver behavior according to an embodiment of this application.

[0038] Figure 2 is a schematic diagram of the optimal braking torque determination process in an embodiment of this application.

[0039] Figure 3 is a block diagram of the energy recovery system that integrates driver behavior according to an embodiment of this application. Detailed Implementation

[0040] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0042] In a first aspect, embodiments of this application provide an adaptive energy recovery method for electric vehicles that incorporates driver behavior.

[0043] In one embodiment, referring to Figure 1, the above method includes:

[0044] S1. Analyze the vehicle's current driving data and historical driving data to determine the driving style and predict the probability of that driving style occurring.

[0045] S2. Based on the vehicle's current driving data, estimate the vehicle's mass and calculate the two values ​​of the slope using two different algorithms.

[0046] S3. Weights are set based on preset vehicle speed thresholds, and the fused slope is obtained by weighting the two values.

[0047] S4. Combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of the driving style occurring, determine the corresponding correction factors, correct the initial braking torque corresponding to the current motor speed to obtain the reference braking torque, compare the reference braking torque with the preset motor torque range to obtain the optimal braking torque and execute braking to achieve energy recovery.

[0048] This application embodiment analyzes the vehicle's current and historical driving data to determine the driving style and predict the probability of that style occurring. Based on the driving style and its probability, the braking response is adjusted to match individual driving habits, improving the adaptability of the energy recovery technology to the driver's driving habits. Based on the vehicle's current driving data, the vehicle's mass is estimated, and two different slope values ​​are calculated using two different algorithms. Weights are assigned by a preset vehicle speed threshold, and the two values ​​are weighted to obtain a fused slope. This reduces reliance on sensors and improves data accuracy and the reliability of the energy recovery system.

[0049] Furthermore, in one embodiment, step S1 above includes:

[0050] S101. Perform preprocessing operations on the driving data, including filtering and normalization.

[0051] Among them, driving data can be collected in advance by various types of sensors, such as vehicle speed sensor to collect vehicle speed, acceleration sensor to collect acceleration, pedal opening sensor to collect pedal opening, gyroscope to collect vehicle attitude information, and SOC (System on Chip) sensor to collect battery remaining power, etc.

[0052] S102. Analyze current and historical driving data using a support vector machine algorithm, such as the frequency of accelerator and brake use, steering amplitude, driving speed, and rate of change of acceleration, to determine the driver's driving style. Driving styles include: aggressive, conservative, and economical. Predict the probability of the driver's driving style occurring among all driving styles and transmit the driving style-related data to the brake energy recovery module.

[0053] The vehicle's driving data (including current driving data and historical driving data) includes: vehicle speed, acceleration, brake pedal opening, vehicle attitude information, and remaining battery power; other data can also be added as needed.

[0054] In this embodiment, by preprocessing the driving data, noise and erroneous signals can be removed, ensuring the accuracy and usability of the data. At the same time, different types of data can be effectively integrated in subsequent analysis. By judging the driver's driving style, the probability of the driver's corresponding driving style appearing in all driving styles is predicted, which improves the adaptability of energy recovery technology to the driver's driving habits and provides a more comfortable and personalized driving experience.

[0055] Furthermore, in one embodiment, the driving data collection and preparation in step S1 mainly involves collecting driving data such as speed, acceleration, braking intensity, and steering angle through various vehicle sensors. The driving data is cleaned, normalized, and balanced sampling processed to provide high-quality input for subsequent processes.

[0056] Step S1 above can be achieved through feature engineering and support vector machine training: key features, such as average speed and rate of change of acceleration, are extracted from driving data. A support vector machine with radial basis function kernels is then trained. Appropriate C and γ values ​​are selected, and the model is optimized through cross-validation to train a classifier that can accurately predict the driver's driving style. By analyzing the vehicle's current and historical driving data, the driver's driving style is accurately predicted, and the output of the support vector machine algorithm is set as a probability output, representing the probability of the driver's corresponding driving style appearing among all driving styles.

[0057] Furthermore, in one embodiment, step S2 above includes:

[0058] S201. Based on the real-time information collected by the gyroscope and acceleration in the vehicle's current driving data, acquire key signals such as X-axis acceleration, Y-axis acceleration, Z-axis angular acceleration, vehicle speed, pedal opening, and vehicle suspension system (such as suspension travel, shock absorber pressure, etc.), and perform signal filtering.

[0059] S202. Estimate the vehicle mass using the least squares estimation method.

[0060] S203. The two values ​​of the slope are calculated by Kalman filtering and least squares estimation respectively.

[0061] Furthermore, in one embodiment, in step S3 above, firstly, a critical speed value is determined based on calibration; then, a weighting coefficient is determined from the critical speed value, and the two slope values ​​are weighted to obtain the merged slope. Specifically:

[0062] A vehicle speed threshold value v1 is set. When the collected current vehicle speed v > v1, the formula for calculating the fused slope is:

[0063] The slope of the blending is equal to θ1*M + (1-M)*θ2.

[0064] Where θ1 is the slope value calculated by Kalman filtering, θ2 is the slope value calculated by least squares estimation, and M is the weighting coefficient, which is the maximum value of the slope fusion dynamic weighting coefficient.

[0065] When the current vehicle speed v ≤ v1, the formula for calculating the fused slope is:

[0066] Blending slope = θ1 * (av) 2 +bv)+[1-(av 2 +bv)」*θ2,

[0067] Among them, av 2 +bv is the weighting coefficient, and the values ​​of a and b are given by the formula v1 = -b / 2a and the formula... Calculated.

[0068] In this embodiment, by using algorithm-based gradient and mass estimation, reliance on sensors is reduced, improving the accuracy and efficiency of data processing. Furthermore, it enables gradual speed increase control when descending long slopes, maintaining the vehicle speed within a preset safe range. This significantly enhances driver confidence and peace of mind, allowing them to maintain greater driving confidence in complex road conditions, while also ensuring vehicle stability and comfort. Under the premise of ensuring safety, it fully utilizes regenerative braking to efficiently convert the kinetic energy generated during vehicle deceleration into electrical energy storage, improving energy utilization efficiency.

[0069] Furthermore, in one embodiment, in step S4 above, the corresponding correction factors are determined by combining the vehicle's current driving data, overall vehicle weight, fusion slope, and the probability of the driving style occurring, including:

[0070] S401 receives key control and status data such as accelerator and brake pedal opening, vehicle speed, mass gradient, SOC, motor speed, and motor torque from the vehicle's current driving data, as well as the probability of driving style occurrence.

[0071] S402. Query the pre-set correction factor table to obtain the correction factors corresponding to the current vehicle speed, pedal opening, vehicle weight, blending gradient, and the probability of the driving style appearing. The correction factor table includes: the probability of real-time vehicle speed, pedal opening, vehicle weight, blending gradient, and driving style appearing, and the correction factor corresponding to each probability.

[0072] Furthermore, in one embodiment, the step S4 above, which corrects the initial braking torque corresponding to the current motor speed to obtain the reference braking torque, includes:

[0073] S403. Obtain the initial braking torque from the current motor speed by querying a preset table of motor speed and target braking torque.

[0074] S404. Multiply the initial braking torque by the required correction factor (i.e., the correction factor corresponding to the current vehicle speed, pedal opening, vehicle weight, blending gradient, and the probability of the driving style occurring) to obtain the reference braking torque.

[0075] In this embodiment, the initial braking torque is adjusted by a correction factor to obtain the reference braking torque, so as to adapt to the current driving conditions and road conditions.

[0076] Furthermore, in one embodiment, referring to Figure 2, the step S4 above, which compares the reference braking torque with a preset motor torque range to obtain the optimal braking torque, includes:

[0077] S405, Preset motor torque range.

[0078] S406. Compare the reference braking torque with the preset motor torque range. If the reference braking torque is within the preset motor torque range (including the end of the range), proceed to S407. If the reference braking torque is less than the minimum motor torque, proceed to S408. If the reference braking torque is greater than the maximum motor torque, proceed to S409.

[0079] S407, take the reference braking torque as the optimal braking torque.

[0080] S408, take the minimum motor torque as the optimal braking torque.

[0081] S409. Take the maximum value of the motor torque as the optimal braking torque.

[0082] Furthermore, in one embodiment, step S4 above, before performing braking, further includes: limiting the output of the torque signal by controlling the slope change of the braking torque and using filtering techniques. This, to a certain extent, avoids shocks and abnormal noises during vehicle braking and improves the smoothness of the braking transition.

[0083] Furthermore, each correction factor value in the aforementioned correction factor table has a preset range, which is determined by calibration. Here, an example of a calibration strategy for the preset ranges of each correction factor in the correction factor table is provided, including vehicle speed correction factor, brake pedal opening correction factor, vehicle weight correction factor, gradient correction factor, and driving style correction factor, specifically:

[0084] Speed ​​correction factor: When the vehicle speed is low, the speed correction factor is small and the energy recovery efficiency is low; when driving at high speed, the kinetic energy increases and more energy can be recovered during braking.

[0085] Brake pedal opening correction factor: When the pedal opening is large, mechanical braking force dominates, electric braking force compensation decreases, and energy recovery decreases; when the opening is small, electric braking force plays a greater role, enhancing energy recovery.

[0086] Vehicle weight correction factor: A heavier vehicle converts more potential and kinetic energy into electrical energy during braking, thus improving energy recovery efficiency.

[0087] Slope correction factor: When driving downhill, the slope factor is increased to enhance energy recovery efficiency by utilizing the potential energy conversion during the downhill process.

[0088] Driving style correction factor: The magnitude of this factor depends on several factors. First, it depends on the probability of the driving style output by the support vector machine algorithm. Second, the driving style correction factor will also differ for drivers with the same driving style at different speeds, vehicle weights, and gradients. Specific data can be obtained by looking up tables. For example: First, the support vector machine algorithm determines that the driver is 70% aggressive, 20% conservative, and 10% economical. Then, based on the obtained speed, vehicle weight, and gradient, tables are looked up for each of the three styles: aggressive, conservative, and economical, to obtain the driving style correction factor A_F for 100% aggressive, C_F for 100% conservative, and E_F for 100% economical. Finally, a weighted calculation is performed to obtain the final driving style correction factor D_F = 0.7*A_F + 0.2*C_F + 0.1*E_F.

[0089] Secondly, based on the above-described embodiments of the trolley adaptive energy recovery method incorporating driver behavior, this application provides an embodiment of an energy recovery system for the trolley adaptive energy recovery method incorporating driver behavior. Referring to Figure 3, the energy recovery system includes an analysis module, an estimation module, a weighting module, and an energy recovery module, specifically:

[0090] The analysis module is used to analyze the vehicle's current driving data and historical driving data to determine the driving style and predict the probability of that driving style occurring.

[0091] The estimation module is used to estimate the vehicle's mass based on the vehicle's current driving data and to calculate two values ​​of the slope using two different algorithms.

[0092] The weighting module is used to set weights based on preset vehicle speed thresholds and obtain the fused slope by weighting the two values.

[0093] The energy recovery module combines the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles to determine corresponding correction factors. It then corrects the initial braking torque corresponding to the current motor speed to obtain a reference braking torque. By comparing the reference braking torque with a preset motor torque range, it obtains the optimal braking torque and performs braking to achieve energy recovery.

[0094] In this embodiment, the energy recovery system combines driver driving style prediction with factors such as vehicle weight, gradient, vehicle speed, and pedal opening, and optimizes the braking energy recovery efficiency of pure electric and hybrid vehicles in real time through intelligent algorithms. This energy recovery system utilizes support vector machine algorithms to analyze historical and real-time driving data, accurately predicts driver driving style, and dynamically adjusts braking torque based on the prediction results and the vehicle's real-time status, optimizing energy recovery and enhancing driving safety and comfort.

[0095] Furthermore, in this embodiment, based on the driving style prediction results and real-time vehicle status data (such as vehicle speed, slope, etc.), the energy recovery module calculates a correction factor and dynamically adjusts the braking torque to meet energy recovery requirements while maintaining the smoothness of the braking process. Compared to most current technologies that only consider vehicle physical parameters and ignore the influence of driving style on torque adjustment, the above-mentioned energy recovery system adopts a dynamic control strategy based on driver driving style and multiple factors, comprehensively considering factors such as driver driving style, slope, and weight, significantly improving the applicability and energy recovery efficiency of the energy recovery system.

[0096] Furthermore, in one embodiment, the energy recovery system further includes a transmission module, which is used to transmit the real-time estimated gradient and vehicle mass information, as well as the vehicle speed and brake pedal opening information collected by the sensors, to the subsequent brake energy recovery module to optimize the brake feedback torque and brake response.

[0097] Furthermore, in one embodiment, a user interface specifically designed for the regenerative braking system of an electric vehicle can be provided, following the principles of simplicity, intuitiveness, and ease of use. Through this interface, users can clearly and intuitively view the current regenerative braking status of the vehicle, including important information such as the real-time energy recovery rate and the distribution ratio of regenerative braking to mechanical braking. Users can also set and adjust relevant parameters of regenerative braking within a certain range, such as setting the intensity level of energy recovery (low, medium, high) and selecting different braking modes (economy mode, comfort mode, etc.), thereby customizing the regenerative braking strategy according to their own needs and driving habits. Simultaneously, the interface provides real-time feedback on the effects of user settings, such as displaying the expected change in the energy recovery rate after parameter adjustments and the impact on driving range, allowing users to clearly understand the impact of their operations on vehicle performance and thus participate more actively in the regenerative braking process.

[0098] This application combines the vehicle's current driving data, overall vehicle weight, gradient, and the probability of the driving style to determine the corresponding correction factors, correct the initial braking torque corresponding to the current motor speed to obtain the reference braking torque, compare the reference braking torque with the preset motor torque range to obtain the optimal braking torque and perform braking to achieve energy recovery. This effectively optimizes the distribution of regenerative braking energy and improves the energy recovery rate by 5%-10%.

[0099] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0100] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0101] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0102] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, S / B can mean S or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, S and / or B can mean: S exists alone, S and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0103] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RSM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0105] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for adaptive energy recovery of electric vehicles that integrates driver behavior, characterized in that, The method includes: Analyze the vehicle's current and historical driving data to determine the driving style and predict the probability of that driving style occurring. Based on the vehicle's current driving data, the total vehicle mass is estimated, and two values ​​of the slope are calculated using two different algorithms. The weights are set by a preset vehicle speed threshold, and the fused slope is obtained by weighting the two values. By combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles, corresponding correction factors are determined. The initial braking torque corresponding to the current motor speed is corrected to obtain the reference braking torque. The reference braking torque is compared with the preset motor torque range to obtain the optimal braking torque and perform braking to achieve energy recovery.

2. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, Analyze the vehicle's current and historical driving data to determine the driving style and predict the probability of that driving style occurring, including: Preprocessing operations are performed on the driving data, including filtering and normalization. The system analyzes current and historical driving data using a support vector machine algorithm to determine the driver's driving style, which includes aggressive, conservative, and economical driving styles. It then predicts the probability of the driver's driving style appearing among all driving styles.

3. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, Based on the vehicle's current driving data, the overall vehicle mass is estimated, and two values ​​for the gradient are calculated using two different algorithms: Based on the vehicle's current driving data, the vehicle mass is estimated using the least squares estimation method, and the two values ​​of the slope are calculated using Kalman filtering and the least squares estimation method, respectively.

4. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, The weights are set based on preset vehicle speed thresholds, and the fused slope is obtained by weighting the two values, including: The vehicle speed threshold value is determined based on calibration. The weighting coefficient is determined by the critical speed value, and the two values ​​of the slope are weighted to obtain the merged slope.

5. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, The correction factors are determined by combining the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles, including: Query the pre-set correction factor table to obtain the correction factors corresponding to the current vehicle speed, pedal opening, vehicle weight, blending gradient, and probability of the driving style. The correction factor table includes: real-time vehicle speed, pedal opening, vehicle weight, blending gradient, and the probability of the occurrence of driving style, as well as the correction factor corresponding to each probability.

6. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, The process of correcting the initial braking torque corresponding to the current motor speed to obtain the reference braking torque includes: The initial braking torque is obtained from the current motor speed; The initial braking torque is multiplied by the required correction factor to obtain the reference braking torque.

7. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, The step of comparing the reference braking torque with a preset motor torque range to obtain the optimal braking torque includes: If the reference braking torque is within the preset motor torque range, the reference braking torque is taken as the optimal braking torque; if the reference braking torque is less than the minimum motor torque, the minimum motor torque is taken as the optimal braking torque; if the reference braking torque is greater than the maximum motor torque, the maximum motor torque is taken as the optimal braking torque.

8. The trolley adaptive energy recovery method based on driver behavior as described in claim 1, characterized in that, Before braking is performed, the following steps are also taken: limiting the output of the torque signal by controlling the slope change of the braking torque and using filtering techniques.

9. The trolley adaptive energy recovery method based on driver behavior as described in any one of claims 1-8, characterized in that, Vehicle driving data includes: vehicle speed, acceleration, brake pedal opening, vehicle attitude information, and remaining battery power.

10. An energy recovery system based on the trolley adaptive energy recovery method for electric vehicles that integrates driver behavior as described in any one of claims 1-8, characterized in that, The energy recovery system includes: The analysis module is used to analyze the vehicle's current driving data and historical driving data to determine the driving style and predict the probability of that driving style occurring. The estimation module is used to estimate the vehicle's mass based on the vehicle's current driving data and to calculate two values ​​of the slope using two different algorithms. The weighting module is used to set weights based on preset vehicle speed thresholds and obtain the fused slope by weighting the two values. The energy recovery module combines the vehicle's current driving data, overall vehicle weight, gradient, and the probability of different driving styles to determine corresponding correction factors. It then corrects the initial braking torque corresponding to the current motor speed to obtain a reference braking torque. By comparing the reference braking torque with a preset motor torque range, it obtains the optimal braking torque and performs braking to achieve energy recovery.

Images