Multi-stage energy recovery control method and device for running vehicle

Abstract

The embodiment of the invention provides a multi-stage energy recovery control method and device for a running vehicle, and the method comprises the steps: obtaining vehicle running parameters of the running vehicle, and generating a dynamic working condition index; obtaining driving operation data and importing the driving operation data into the behavior prediction model to generate a braking demand probability grade; determining a corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level; and when the energy recovery grade is a middle grade, a dynamic torque change curve is generated to control the vehicle to execute energy recovery. And the balance problem of the energy recovery efficiency and the driving comfort is effectively solved. The generation of the dynamic working condition index ensures the accurate perception of the complex driving environment, and the introduction of the behavior prediction model realizes the pre-judgment of the driving intention. The energy recovery intensity is optimally matched with the real-time working condition through a two-factor linkage hierarchical control strategy, and the pause feeling generated by traditional stepped adjustment is remarkably reduced through application of a dynamic torque curve.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of vehicle energy recovery, and particularly to a multi-level energy recovery control method and device for an operating vehicle. Background Art

[0002] To maximize the driving range, new energy vehicles are equipped with energy recovery devices to recover some of the energy. The driving motor of the vehicle applies a positive torque to drive the vehicle forward and a negative torque to charge the drive battery, i.e., for energy recovery.

[0003] In the prior art, it is impossible to dynamically adjust according to the actual operating conditions of the vehicle and the driving behavior of the driver, resulting in low energy recovery efficiency. Traditional energy recovery methods usually adopt fixed recovery strategies, which cannot adapt to different driving environments and driver operating habits, thus affecting the energy recovery effect and resulting in a low energy recovery rate. Summary of the Invention

[0004] In view of this, the embodiments of this specification provide a multi-level energy recovery control method for an operating vehicle. One or more embodiments of this specification also relate to a multi-level energy recovery control device for an operating vehicle, a computing device, a computer-readable storage medium, and a computer program to solve the technical defects existing in the prior art.

[0005] According to the first aspect of the embodiments of this specification, a multi-level energy recovery control method for an operating vehicle is provided, including: Obtaining the vehicle operation parameters of the operating vehicle and performing weighted fusion processing to generate a dynamic working condition index; Obtaining the driving operation data of the current driver within a preset time window and importing it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained by the historical driving operation data of the current driver; Determining the corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level; When the energy recovery level is medium, generating a dynamic torque change curve based on the vehicle operation parameters and controlling the vehicle to perform energy recovery based on the dynamic torque change curve.

[0006] In some embodiments, performing weighted fusion processing on the vehicle operation parameters to generate a dynamic working condition index includes: Obtaining the vehicle speed, acceleration, slope angle, and battery load rate in the vehicle operation parameters; Performing fuzzy logic weighted fusion on the vehicle speed, acceleration, and slope angle to obtain a dynamic working condition index, where the product of the acceleration and the vehicle speed is assigned a preset first weight, the slope angle is assigned a preset second weight, and the battery load rate is assigned a preset third weight.

[0007] In some embodiments, obtaining the driving operation data of the current driver within a preset time window and importing it into a preset behavior prediction model to obtain a braking demand probability level includes: Obtaining the driving operation data of the current driver within a preset time window, where the driving operation data includes the accelerator pedal position, vehicle speed variance, and steering angle standard deviation; Importing the accelerator pedal position, vehicle speed variance, and steering angle standard deviation into the behavior prediction model to generate a braking demand probability level, where the prediction model is a long short-term memory neural network.

[0008] In some embodiments, the energy recovery levels include low, medium, and high. According to the dynamic working condition index and the braking demand probability level, determining the corresponding energy recovery level includes: When the dynamic working condition index is not greater than a preset first threshold and the braking demand probability level is low, the energy recovery level is low, generating a turn-off recovery instruction and outputting it; When the dynamic working condition index is greater than a preset second threshold and the braking demand probability level is high, the energy recovery level is high, generating a preloading energy storage capacitor instruction and a maximum recovery torque instruction and outputting them; When the dynamic working condition index is greater than the first threshold and the braking demand probability level is medium or high, the energy recovery level is medium; When the dynamic working condition index is not greater than the second threshold and the braking demand probability level is low or medium, the energy recovery level is medium; When the dynamic working condition index is not greater than the first threshold and less than the second threshold, the energy recovery level is medium.

[0009] In some embodiments, the curve optimization process includes: Calculating the current vehicle speed decrease rate; Based on the battery temperature parameter and the vehicle speed decrease rate, using an interpolation algorithm to generate a smoothly transitioning torque change curve, where the torque increase rate is positively correlated with the vehicle speed decrease rate.

[0010] In some embodiments, the method further includes: When the slope angle in the vehicle operating parameters exceeds a preset third threshold, reducing the maximum recovery torque based on a preset first ratio; When the yaw rate in the vehicle operating parameters exceeds a preset fourth threshold and the energy recovery level is high, downgrading the energy recovery level to medium.

[0011] In some embodiments, it further includes: Calculating the theoretical recovered energy; Performing a deviation calculation process on the obtained actual recovered energy and the theoretical recovered energy to obtain a weight correction coefficient; Dynamically adjust the first weight, second weight, and third weight ratios according to the weight correction coefficient.

[0012] According to the second aspect of the embodiments of this specification, a multi-level energy recovery control device for an operating vehicle is provided, including: A weighted fusion processing module configured to obtain vehicle operation parameters of the operating vehicle and perform weighted fusion processing to generate a dynamic working condition index; A first generation module configured to obtain driving operation data of the current driver within a preset time window and import it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained by historical driving operation data of the current driver; A determination module configured to determine a corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level; A second generation module configured to, when the energy recovery level is medium, generate a dynamic torque change curve based on the vehicle operation parameters and control the vehicle to which it belongs to perform energy recovery based on the dynamic torque change curve.

[0013] In some embodiments, performing weighted fusion processing on vehicle operation parameters to generate a dynamic working condition index includes: Obtain the vehicle speed, acceleration, slope angle, and battery load rate in the vehicle operation parameters; Perform fuzzy logic weighted fusion on the vehicle speed, acceleration, and slope angle to obtain a dynamic working condition index, where the product of the acceleration and the vehicle speed is assigned a preset first weight, the slope angle is assigned a preset second weight, and the battery load rate is assigned a preset third weight.

[0014] In some embodiments, obtaining driving operation data of the current driver within a preset time window and importing it into a preset behavior prediction model to obtain a braking demand probability level includes: Obtain driving operation data of the current driver within a preset time window, where the driving operation data includes the accelerator pedal position, vehicle speed variance, and steering angle standard deviation; Import the accelerator pedal position, vehicle speed variance, and steering angle standard deviation into the behavior prediction model to generate a braking demand probability level, where the prediction model is a long short-term memory neural network.

[0015] In some embodiments, the energy recovery levels include low, medium, and high. Determining the corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level includes: When the dynamic working condition index is not greater than a preset first threshold and the braking demand probability level is low, the energy recovery level is low, and a shutdown recovery instruction is generated and output; When the dynamic operating condition index is greater than a preset second threshold and the braking demand probability level is high, the energy recovery level is high, and a preloaded energy storage capacitor instruction and a maximum recovery torque instruction are generated and output; When the dynamic operating condition index is greater than the first threshold and the braking demand probability level is medium or high, the energy recovery level is medium; When the dynamic operating condition index is not greater than the second threshold and the braking demand probability level is low or medium, the energy recovery level is medium; When the dynamic operating condition index is not greater than the first threshold and less than the second threshold, the energy recovery level is medium.

[0016] In some embodiments, the curve optimization process includes: Calculating the current vehicle speed decrease rate; Based on the battery temperature parameter and the vehicle speed decrease rate, an interpolation algorithm is used to generate a smoothly transitioning torque change curve, where the torque increase rate is positively correlated with the vehicle speed decrease rate.

[0017] In some embodiments, the device further includes an adjustment module configured to: when the slope angle in the vehicle operating parameters exceeds a preset third threshold, reduce the maximum recovery torque based on a preset first ratio; When the yaw angular velocity in the vehicle operating parameters exceeds a preset fourth threshold and the energy recovery level is high, downgrade the energy recovery level to medium.

[0018] In some embodiments, the device further includes an adjustment module configured to: Calculate the theoretical recovered energy; Perform a deviation calculation process on the obtained actual recovered energy and the theoretical recovered energy to obtain a weight correction coefficient; Dynamically adjust the first weight, the second weight, and the third weight ratio according to the weight correction coefficient.

[0019] According to the third aspect of the embodiments of this specification, a vehicle is provided. The vehicle is provided with a control center and a hydraulic braking module AHB. The hydraulic braking module AHB is provided with a brake controller unit ECU, a supercharging unit PSU, and a hydraulic control unit PCU, where, The control center is configured to execute the steps of the multi-level energy recovery control method for the running vehicle described above; The brake controller unit ECU is respectively connected to the supercharging unit PSU and the hydraulic control unit PCU, and is configured to receive control instructions sent by the upper-level control center, and generate a first instruction and a second instruction according to the control instructions to control the supercharging unit PSU and the hydraulic control unit PCU to perform pressure control, where the first instruction is sent to the supercharging unit PSU, and the second instruction is sent to the supercharging unit PSU; The supercharging unit PSU is used to output hydraulic oil at a certain pressure to the hydraulic control unit PCU according to the received first instruction; The hydraulic control unit PCU is used to adjust, control, and distribute the output pressure according to the received second instruction to ensure different working pressures under different working conditions.

[0020] According to the fourth aspect of the embodiments of the present specification, a computing device is provided, including: A memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above multi-level energy recovery control method for an operating vehicle are implemented.

[0021] According to the fifth aspect of the embodiments of the present specification, a computer-readable storage medium is provided, which stores computer-executable instructions. When the instructions are executed by the processor, the steps of the above multi-level energy recovery control method for an operating vehicle are implemented.

[0022] According to the sixth aspect of the embodiments of the present specification, a computer program is provided. When the computer program is executed on a computer, the computer is made to execute the steps of the above multi-level energy recovery control method for an operating vehicle.

[0023] In at least one embodiment of the embodiments of the present specification, by obtaining the vehicle operation parameters of a running vehicle and performing weighted fusion processing, a dynamic working condition index is generated; obtaining the driving operation data of the current driver within a preset time window and importing it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained from the historical driving operation data of the current driver; determining the corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level; when the energy recovery level is medium, generating a dynamic torque change curve based on the vehicle operation parameters, and controlling the vehicle to which it belongs to perform energy recovery based on the dynamic torque change curve. Effectively solves the balance problem between energy recovery efficiency and driving comfort. The generation of the dynamic working condition index ensures accurate perception of complex driving environments, and the introduction of the behavior prediction model realizes the advance prediction of driving intentions. The hierarchical control strategy with two-factor linkage enables the energy recovery intensity to be optimally matched with the real-time working conditions, and the application of the dynamic torque curve significantly reduces the jerks generated by traditional stepwise regulation. Description of the Drawings

[0024] Figure 1 is a flowchart of some embodiments of a multi-level energy recovery control method for an operating vehicle provided by some embodiments of the present specification; Figure 2 is a flowchart of other embodiments of a multi-level energy recovery control method for an operating vehicle provided by some embodiments of the present specification; Figure 3 It is a schematic diagram of a simple structure of a multi-level energy recovery control device for an operating vehicle provided by some embodiments of this specification; Figure 4 It is a block diagram of the structure of a computing device provided by some embodiments of this specification. Detailed implementation manners

[0025] In the following description, many specific details are set forth in order to provide a thorough understanding of this specification. However, this specification can be implemented in many other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of this specification. Therefore, this specification is not limited by the specific implementations disclosed below.

[0026] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more of the associated listed items. The modifications of "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless clearly specified otherwise in the context, it should be understood as "one or more".

[0027] It should be understood that although the terms first, second, etc. may be used in one or more embodiments of this specification to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein can be interpreted as "when" or "while" or "in response to determining".

[0028] First, the noun terms related to one or more embodiments of this specification are explained.

[0029] AHB: Accumulator-based Electro-hydraulic Braking Module, an energy storage type electro-hydraulic braking module; ECU: Electronic Control Unit, a braking controller unit; PSU: Pressure Supply Unit, a pressurizing unit; PCU: Pressure Control Unit, a hydraulic control unit.

[0030] In the prior art, vehicle energy recovery systems usually perform single-mode control based on fixed operating condition parameters. Traditional methods only trigger energy recovery according to vehicle speed or brake pedal signals, and fail to fully integrate the dynamic driving environment and driving behavior characteristics. Especially in complex road conditions, fixed-threshold control is prone to cause a lag in the recovery timing, and frequent starts and stops result in energy losses. In a typical scenario, when driving on continuous curves on urban roads, the existing system cannot predict the driver's braking intention, resulting in the failure to start energy recovery in a downhill section in time, and premature intervention in a straight section, affecting driving smoothness.

[0031] To solve the above problems, it is first observed that the energy recovery efficiency is affected by both vehicle dynamic parameters and driver operating habits. By analyzing historical driving data, it is found that the braking operations of drivers under similar road conditions are predictable. It is further found that the dynamic combination of vehicle acceleration and slope can characterize the real-time operating condition load. Thus, the core idea is formed: to establish a joint decision-making mechanism for vehicle operating conditions and driver behavior, and to achieve dynamic adaptation of energy recovery through hierarchical control. The specific solution path is: to design a multi-dimensional parameter fusion algorithm to generate an operating condition index, to construct a driver behavior prediction model, and to develop a multi-level control strategy based on two-factor linkage.

[0032] Therefore, referring to Figure 1 , an embodiment of the present invention proposes a multi-level energy recovery control method for an operating vehicle, including: Step 101: Obtain the vehicle operating parameters of the operating vehicle and perform weighted fusion processing to generate a dynamic operating condition index.

[0033] Step 102: Obtain the driving operation data of the current driver within a preset time window and import it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained by the historical driving operation data of the current driver.

[0034] Step 103: Determine the corresponding energy recovery level according to the dynamic operating condition index and the braking demand probability level.

[0035] Step 104: When the energy recovery level is medium, generate a dynamic torque change curve based on the vehicle operating parameters, and control the vehicle to which it belongs to perform energy recovery based on the dynamic torque change curve.

[0036] Among them, weighted fusion processing can refer to fuzzy logic calculation and weight allocation of parameters such as vehicle speed, acceleration, and slope angle; for example, it can be implemented using membership function and weighted summation algorithm to quantify the current motion state of the vehicle. During the generation of the dynamic working condition index, the product of acceleration and vehicle speed is given a higher weight to accurately reflect the trend of kinetic energy changes. The behavior prediction model can be specifically constructed using a long short-term memory neural network, and the braking probability is predicted by analyzing time series data such as the accelerator pedal position and vehicle speed variance. When the dynamic torque change curve is generated, the vehicle speed drop rate and battery temperature parameters can be combined for interpolation calculation to ensure that the torque change is synchronized with the vehicle deceleration process.

[0037] Specifically, the real-time acquisition system of vehicle operating parameters continuously monitors data such as vehicle speed and acceleration, and calculates the membership value of each parameter through a fuzzy logic processor. The weighted fusion module synthesizes the dynamic operating condition index according to the preset weight ratio, and the index value range can be set to 0-100. At the same time, the driving behavior analysis unit extracts operating features such as the accelerator pedal position in a sliding time window manner, and inputs the pre-trained behavior prediction model to output low, medium and high three-level braking probabilities. The control decision module combines the operating condition index with the braking probability for judgment. When the intermediate energy recovery mode is activated, the torque curve generator calculates a smooth transition torque control instruction based on the real-time vehicle speed change rate and battery status parameters.

[0038] Compared with the existing technology, the traditional method only relies on a single vehicle speed threshold to trigger recovery, while this solution achieves dual-factor decision-making by integrating vehicle dynamic parameters and driver behavior characteristics. The existing technology uses a fixed recovery level, which leads to obvious frustration. This solution generates a dynamic torque curve in the intermediate mode to achieve progressive control. Compared with the conventional brake signal trigger mechanism, this solution predicts the braking demand in advance, so that the energy recovery system can enter the ready state in advance.

[0039] Through the above technical solutions, the embodiment of the present invention effectively solves the problem of balancing energy recovery efficiency and driving comfort. The generation of dynamic operating condition index ensures accurate perception of complex driving environments, and the introduction of behavior prediction models enables early prediction of driving intentions. The dual-factor linkage hierarchical control strategy enables the energy recovery intensity to achieve the best match with the real-time operating conditions, and the application of dynamic torque curves significantly reduces the sense of frustration caused by traditional step-by-step adjustment.

[0040] In some embodiments, the embodiments of the present invention further propose to obtain the vehicle speed, acceleration, slope angle, and battery load rate in the vehicle operating parameters, and perform fuzzy logic weighted fusion on the vehicle speed, acceleration and slope angle to obtain a dynamic operating condition index, where the product of acceleration and vehicle speed is assigned a first weight, the slope angle is assigned a second weight, and the battery load rate is assigned a third weight.

[0041] Among them, fuzzy logic weighted fusion refers to a data processing method that converts continuous parameters into fuzzy sets through membership functions and uses a preset rule base for reasoning and decision-making. Specifically, it can be implemented using triangular or trapezoidal membership functions and is used to process non-linear parameter relationships. The product of acceleration and vehicle speed refers to the composite variable obtained by multiplying the two, which can be specifically implemented using the product operation of the real-time data acquisition module and is used to characterize the rate of change of vehicle kinetic energy. Assigning the second weight to the slope angle means independently assigning an influence coefficient to the road inclination angle, which can be specifically implemented by multiplying the measured data of the angle sensor by a preset coefficient and is used to reflect the influence degree of terrain resistance on the energy recovery demand. Assigning the third weight to the battery load rate means setting an adjustment factor according to the current capacity state of the battery, which can be specifically implemented by multiplying the state of charge detection value by a preset proportional coefficient and is used to control the recovery intensity to prevent overcharging of the battery.

[0042] Specifically, the vehicle speed, acceleration, and slope angle data are collected in real time through in-vehicle sensors, and the load rate parameter provided by the battery management system is obtained synchronously. The product of acceleration and vehicle speed is used to obtain the kinetic energy change factor, and this product term is input into the fuzzy inference system after being adjusted by the first weight coefficient. The measured value of the slope angle forms the terrain resistance factor after being scaled by the second weight coefficient, and the battery load rate generates the safety constraint factor after being converted by the third weight coefficient. In the fuzzy logic processing layer, a three-dimensional membership function for vehicle speed, product term, and slope angle is set, a fuzzy rule base including battery load rate constraint conditions is established, and a dynamic working condition index is output through defuzzification calculation. This index comprehensively reflects the energy conversion characteristics of the vehicle motion state, the influence of road environment resistance, and the safety boundary of the energy storage device.

[0043] Compared with the prior art, traditional methods mostly use single-parameter threshold judgment or simple weighted average, which cannot accurately reflect the coupling effect of vehicle speed and acceleration and ignore the non-linear influence of slope angle on braking demand. This solution constructs a kinetic energy change characterization quantity through the product term, uses fuzzy logic to process multi-parameter interaction relationships, and while ensuring the real-time nature of the calculation, accurately quantifies the energy recovery demand under different working conditions. At the same time, the battery load rate is introduced as a dynamic constraint condition, overcoming the adaptability defect of the fixed weight allocation mode when the battery state changes.

[0044] Through the above technical solution, the embodiment of the present invention effectively solves the problem of working condition evaluation deviation caused by insufficient multi-dimensional parameter fusion, and realizes the coordinated optimization of vehicle kinetic energy state, road environment resistance, and battery safety boundary. The dynamic weight allocation mechanism takes into account the special needs of different driving scenarios, and the fuzzy logic processing improves the analysis accuracy of non-linear parameters, providing an accurate working condition judgment basis for energy recovery control. The dynamic constraint of the battery load rate avoids the overcharging risk and ensures the safety of system operation.

[0045] In some embodiments, the embodiments of the present invention further propose to obtain the driving operation data of the current driver within a preset time window, where the driving operation data includes the accelerator pedal position, vehicle speed variance, and steering angle standard deviation, and import these data into a behavior prediction model to generate a braking demand probability level, where the prediction model uses a long short-term memory neural network.

[0046] Among them, the preset time window refers to the time interval for continuously collecting driving operation data. For example, it can be set to 10 seconds to 30 seconds, and the data sequence is dynamically updated through a sliding window mechanism to capture the temporal correlation characteristics of driving behavior. The accelerator pedal position refers to the opening percentage data of the vehicle accelerator pedal, which is obtained by a sensor at a fixed sampling frequency and is used to characterize the driver's real-time acceleration intention. The vehicle speed variance refers to the square value of the standard deviation of the vehicle speed within the preset time window, which is calculated from continuous vehicle speed data and is used to reflect the degree of dispersion of vehicle speed changes. The steering angle standard deviation refers to the standard deviation value of the steering wheel angle within the preset time window, which is collected and calculated by a steering angle sensor and is used to reflect the stability of the steering operation during driving. The long short-term memory neural network refers to a time series processing model with a gating mechanism, which models the temporal characteristics of driving operation data through the structures of a forgetting gate, an input gate, and an output gate, and can capture long-term dependencies in driving behavior.

[0047] Specifically, during the vehicle operation, the accelerator pedal position, vehicle speed, and steering angle data are continuously collected at fixed time intervals, and the vehicle speed variance and steering angle standard deviation are calculated within a sliding time window. The combination of these three parameters can construct a driving behavior feature vector from three dimensions: acceleration intention, speed volatility, and steering stability. The long short-term memory neural network learns the mapping relationship between the driving operation mode and the braking demand by processing the driving behavior data sequence in consecutive time windows step by step. For example, when it is detected that the accelerator pedal position fluctuates frequently and the vehicle speed variance increases, the model can identify that the driver is in a state of frequent acceleration and deceleration, and then predict that the probability of a braking demand in the short term increases. The introduction of the steering angle standard deviation helps to identify driving behaviors under complex road conditions, such as the impact of changes in steering operation stability on braking demand when driving on a curve. By optimizing the network parameters using historical driving data during the training phase, the model can adapt to different driving styles and accurately output low, medium, and high-level braking demand probabilities.

[0048] Compared with the prior art, traditional methods usually rely only on a single parameter such as vehicle speed or acceleration for braking prediction, and cannot effectively capture the temporal characteristics and multi-dimensional correlations of driving behaviors. For example, some systems use a fixed threshold to judge the braking demand, and trigger energy recovery when the acceleration is lower than the set value. This method is prone to misjudgment when the driver maintains a constant speed or decelerates slightly. In contrast, this solution can more accurately identify the braking intention hidden in the driving behavior pattern by integrating multi-dimensional dynamic parameters such as acceleration intention, speed fluctuation, and steering stability, and combining the modeling ability of the long short-term memory neural network for temporal characteristics. Compared with traditional static models such as feedforward neural networks or support vector machines, the unique memory cell structure of the long short-term memory neural network can effectively process the temporal continuity of driving behavior data. For example, it can identify the behavior pattern of braking after three consecutive hard accelerations.

[0049] Through the above technical solution, the embodiments of the present invention can accurately predict the timing and intensity of the driver's braking demand, and avoid starting the energy recovery system too early or too late. When a high-probability braking demand is detected, the energy recovery level is adjusted in advance to ensure that the braking energy recovery is synchronized with the driver's operation intention, and reduce the energy waste caused by misjudgment of the braking demand in traditional methods. At the same time, through the collaborative analysis of multi-dimensional behavior parameters, the normal driving state and the emergency braking scenario can be effectively distinguished, and the triggering of high-level energy recovery in non-necessary situations is prevented, which affects the driving smoothness.

[0050] In some embodiments, the embodiments of the present invention further propose a control strategy for determining the energy recovery level by combining the dynamic working condition index and the braking demand probability level. The energy recovery level is divided into three levels: low level, medium level, and high level. The specific determination rules are as follows: when the dynamic working condition index does not exceed the first threshold and the braking demand probability level is low, the low-level mode of turning off the energy recovery is adopted; when the dynamic working condition index exceeds the second threshold and the braking demand probability level is high, the high-level mode including pre-loading energy storage and maximum torque recovery is started; in other combinations of working condition conditions, the medium-level recovery mode is adopted.

[0051] Among them, the dynamic working condition index refers to a comprehensive evaluation value generated by weighted fusion of vehicle parameters such as vehicle speed and acceleration. Specifically, it can be realized by using a fuzzy logic algorithm to fuse multi-source signals. This index is used to characterize the complexity of the current driving environment. The braking demand probability level refers to the grading of the braking demand intensity predicted based on the driver's operation behavior. Specifically, it can be realized by using a long short-term memory neural network to analyze parameters such as the accelerator pedal position and the standard deviation of the steering angle. This grading is used to reflect the driver's potential braking intention. The preset threshold refers to the judgment boundary value set according to the vehicle performance parameters. Specifically, it can be determined by using the statistical analysis method of real vehicle test data. For example, the first threshold can be the reference value in the vehicle coasting state, and the second threshold can be the critical value in the hard deceleration working condition.

[0052] Specifically, during the operation of the vehicle, by calculating the comparison relationship between the dynamic working condition index and the preset threshold in real time, and combining the multi-level evaluation of the braking demand probability level, five groups of conditional judgment logics are constructed. When it is detected that both the low-load working condition and the low braking demand exist simultaneously, the shutdown recovery operation is executed to avoid energy waste; when it is recognized that the high-load working condition and the forced braking demand occur synchronously, the energy storage preloading mechanism is activated and the maximum torque recovery is implemented; for other working condition combinations in the intermediate state, the medium recovery intensity is maintained. This judgment mechanism forms three working condition intervals through double-threshold division, and combines the three-level braking demand prediction to form a cross-verification judgment system, effectively avoiding misjudgment caused by the fluctuation of a single parameter.

[0053] Compared with the prior art, traditional methods usually control the recovery intensity only based on a single parameter such as vehicle speed or acceleration, and are prone to misjudgment under complex working conditions. For example, in a long downhill section, excessive recovery may occur due to continuous braking demand, or the recovery efficiency may be low due to frequent start-stop in urban congestion conditions. This solution combines vehicle dynamic parameters with driver behavior prediction, and adopts a multi-dimensional conditional combination judgment mechanism, which not only ensures the accuracy of basic working condition judgment, but also adjusts the recovery strategy in advance through behavior prediction, significantly improving the adaptability and judgment accuracy of the control system.

[0054] Through the above technical solutions, the embodiments of the present invention achieve precise hierarchical control of the energy recovery intensity, effectively solving the problem of low recovery efficiency caused by incomplete working condition recognition in traditional methods. It avoids system losses caused by ineffective energy recovery under low-demand working conditions, maximizes the energy capture efficiency under high-demand working conditions, and maintains the basic recovery ability under medium working conditions. This hierarchical control strategy is particularly suitable for energy management of hybrid vehicles under complex urban road conditions, and can automatically optimize the recovery strategy according to the real-time driving state and driver's operating habits.

[0055] In some embodiments, the embodiments of the present invention further propose curve optimization processing, including calculating the current vehicle speed decrease rate, and using an interpolation algorithm to generate a smooth torque change curve based on the battery temperature parameter and the vehicle speed decrease rate, where the torque increase rate is positively correlated with the vehicle speed decrease rate.

[0056] Among them, the vehicle speed decrease rate refers to the amount of vehicle speed change per unit time, and can be specifically realized by differential calculation of the real-time vehicle speed data collected by the on-vehicle speed sensor. This parameter is used to reflect the urgency of the vehicle deceleration trend and provides a basic judgment basis for the dynamic adjustment of the torque curve.

[0057] Among them, the battery temperature parameter refers to the real-time temperature monitoring value of the power battery module, and can be specifically obtained by temperature sensors arranged in the battery pack. This parameter is used to restrict the torque adjustment amplitude and prevent safety risks caused by battery overheating.

[0058] Among them, the interpolation algorithm refers to a mathematical method for generating a continuous curve based on discrete data points, and specifically, the cubic spline interpolation method can be used to implement it. This algorithm can dynamically construct a torque change curve according to the working condition parameters that change in real time, avoiding stepwise mutations.

[0059] Among them, the positive correlation relationship means that the growth amplitude of the torque rising rate and the growth amplitude of the vehicle speed decreasing rate change in the same direction, and specifically, a mapping relationship can be established through a linear proportional function or an exponential function. This design ensures that the energy recovery intensity is dynamically matched with the actual deceleration demand of the vehicle.

[0060] Specifically, when the vehicle decelerates, the decreasing rate of the current vehicle speed is calculated in real time, and at the same time, the power battery temperature data is read. The reference torque rising rate is determined according to the decreasing rate of the vehicle speed, and the reference value is safely corrected in combination with the battery temperature. The cubic spline interpolation algorithm is used to generate a smooth torque transition curve between adjacent control cycles, so that the rising slope of the torque control command maintains a preset positive mapping relationship with the decreasing rate of the vehicle speed. For example, when it is detected that the vehicle speed drops by more than 20 km / h within 0.5 seconds, the system will automatically increase the torque rising rate according to the proportional coefficient, and at the same time, dynamically limit the target torque value according to whether the battery temperature exceeds 55°C. Thus, the continuous adjustment of the energy recovery torque is realized, and the energy loss caused by mechanical impact is eliminated.

[0061] Compared with the prior art, the traditional energy recovery system usually directly switches fixed torque values according to the vehicle speed threshold, resulting in torque command mutations that cause mechanical vibrations. However, the present solution can eliminate stepwise torque changes through the smooth curve generated by the dynamic interpolation algorithm, and simultaneously consider the limitation of the battery thermal state on the recovery power during the adjustment process, improving the energy conversion efficiency on the premise of ensuring system safety.

[0062] Through the above technical solution, the embodiment of the present invention effectively solves the problem of mechanical energy loss caused by torque mutations during the energy recovery process. By establishing a dynamic correlation mechanism between the vehicle speed change rate and the torque adjustment rate, the energy recovery intensity can adapt to the actual deceleration demand of the vehicle, avoiding the impact on the power transmission system while maximizing the kinetic energy conversion efficiency during the braking process.

[0063] In some embodiments, generating a dynamic torque change curve based on vehicle operating parameters includes: Obtaining the optimal operating temperature, the limit operating temperature, and the current operating temperature of the vehicle battery, combining the dynamic operating condition index and the braking demand probability level, the current vehicle speed, the current slope angle, the preset time constant, the time cumulative amount starting from the braking trigger moment, and the preset first calculation formula in the vehicle operating parameters to generate a dynamic torque change curve, where the first calculation formula includes:

[0064] Among them, T ( t ) represents the torque varying with t The torque varies with DCI represents the dynamic operating condition index, ‌P brake represents the braking demand probability level, v represents the current vehicle speed, t represents the time accumulation, represents the preset time constant, ‌T bat represents the current operating temperature of the battery, represents the preset limit operating temperature of the battery, θ represents the current slope angle, ‌T opt represents the preset optimal operating temperature of the battery.

[0065] The above calculation formula realizes multi-objective collaborative control while ensuring the curve smoothness through the cross-validation of physical constraints and statistical prediction, and can further reduce the operation load of the controller compared with the traditional cubic spline interpolation.

[0066] In some embodiments, the embodiments of the present invention further propose that when the slope angle in the vehicle operating parameters exceeds a preset third threshold, the maximum recovery torque is reduced based on a preset first ratio; when the yaw angular velocity in the vehicle operating parameters exceeds a preset fourth threshold and the energy recovery level is high, the energy recovery level is downgraded to medium.

[0067] Among them, the slope angle refers to the inclination angle of the road where the vehicle is currently located, which can be specifically collected in real time by an in-vehicle inertial measurement unit or a slope sensor, and is used to judge whether the vehicle is in a steep slope condition. The yaw angular velocity refers to the angular velocity of the vehicle rotating around the vertical axis, which can be specifically measured by a gyroscope sensor and is used to characterize the vehicle's steering or side-slip dynamics. The third threshold is set according to the vehicle type and terrain adaptability, for example, it can be set to 15 degrees. The fourth threshold is dynamically adjusted according to the vehicle dynamics stability parameters, for example, it can be set to 30 degrees per second. The first ratio refers to the reduction ratio of the maximum recovery torque, which can be specifically calculated dynamically according to the difference between the slope angle and the third threshold. For example, when the slope angle exceeds the threshold, the torque is reduced by 2% for each additional 1 degree.

[0068] Specifically, when the slope angle exceeds the third threshold, the vehicle is in a steep downhill state. At this time, directly performing high-intensity energy recovery may cause the vehicle's center of gravity to shift or the braking system to be overloaded. By linearly reducing the maximum recovery torque at a preset first ratio, a dynamic balance can be formed between the braking torque and the slope resistance. At the same time, when the yaw rate exceeds the fourth threshold, it indicates that the vehicle is performing an emergency turn or there is a risk of sideslip. If the reverse torque generated by maintaining the advanced energy recovery is maintained at this time, it may exacerbate the vehicle's yaw motion. By forcibly downgrading the energy recovery level from advanced to intermediate, the impact of torque mutation on vehicle stability can be reduced. This dual adjustment mechanism, through the real-time coupling of vehicle dynamic parameters and energy recovery strategies, while maintaining the recovery efficiency under normal working conditions, gives priority to ensuring driving safety under special working conditions.

[0069] Compared with the prior art, traditional energy recovery systems usually only control according to vehicle speed or braking signals, without considering the influence of complex terrain and vehicle dynamic stability on the recovery process. For example, in the patent with the publication number CN112744189A, only the braking pedal travel is used to judge the recovery intensity, without involving the comprehensive judgment of slope and yaw angle parameters. And this solution, by introducing the dual threshold judgment of slope angle and yaw rate, establishes a direct relationship between the energy recovery intensity and the vehicle dynamic state, and solves the problem of insufficient system robustness caused by single-parameter control.

[0070] Through the above technical solutions, the embodiments of the present invention effectively avoid the risk of vehicle center of gravity shift caused by excessive recovery torque under steep slope conditions, and prevent the interference of high-intensity energy recovery on the vehicle dynamic balance in the state of high yaw rate. On the basis of maintaining the energy recovery efficiency, through the real-time monitoring of physical parameters and the dynamic adjustment of the recovery level, the safety control priority under special working conditions is improved, and the vehicle driving stability and braking system reliability are ensured.

[0071] In some embodiments, the embodiments of the present invention further propose to calculate the theoretical recovered energy; perform deviation calculation processing on the obtained actual recovered energy and the theoretical recovered energy to obtain a weight correction coefficient; perform dynamic adjustment processing on the first weight, the second weight, and the third weight ratio according to the weight correction coefficient.

[0072] Among them, theoretical recovered energy refers to the theoretical recoverable energy value based on the current operating state of the vehicle. It can be calculated by combining the vehicle dynamics model with the battery characteristic curve, and is used to establish a benchmark reference for the energy recovery effect. Actual recovered energy refers to the electric energy data output by the energy recovery system collected in real time by sensors. It can be obtained by combining current sensors and voltage sensors for measurement and integral calculation, and is used to reflect the energy recovery efficiency under actual working conditions. The weight correction coefficient refers to the deviation ratio between the actual recovered energy and the theoretical recovered energy. It can be calculated by the root mean square error after normalization, and is used to quantify the adjustment needs of the current weight allocation strategy. Dynamic adjustment processing refers to adjusting the weight ratio of each parameter in the weighted fusion according to deviation feedback. It can be implemented by the proportional-integral control algorithm, and a new weight allocation ratio is generated by inputting the correction coefficient into the control model.

[0073] Specifically, during the operation of the vehicle, the theoretical recoverable energy value is first calculated through the theoretical energy model based on the current vehicle speed, acceleration and battery status parameters. At the same time, the actual recovered energy data is collected through the on-board power metering device, and both are input into the deviation calculation module. When the actual recovered energy is continuously lower than the theoretical value, it indicates that the weight distribution of vehicle speed or acceleration may be too high in the current weighted fusion process, causing the dynamic working condition index to deviate from the actual recovery potential. At this time, the correction coefficient is generated by the deviation ratio, which acts on the fuzzy logic weighted fusion module to gradually reduce the weight ratio of vehicle speed and acceleration, and at the same time increase the weight ratio of the slope angle parameter. For example, when the vehicle is in a long downhill condition, the theoretical model can identify the significant impact of the slope angle on the braking demand, and if the actual recovered energy is lower than expected, the weight of the slope angle is increased by adjusting the coefficient, so that the dynamic working condition index more accurately reflects the real energy recovery demand, thereby optimizing the energy recovery level judgment logic.

[0074] Compared with the existing technology, the traditional method usually adopts fixed weights to fuse vehicle operating parameters, which cannot adapt to complex and changeable driving conditions. For example, when the slope angle changes suddenly or the battery load fluctuates, the fixed weight distribution will cause the dynamic operating index to be distorted, thereby causing misjudgment of the energy recovery level. However, this solution can identify the deviation between the weight distribution and the working condition matching degree in real time by establishing a closed-loop feedback mechanism of theoretical-actual energy recovery, and dynamically optimize the contribution ratio of each parameter through an adaptive algorithm. This reverse weight adjustment mechanism based on energy recovery efficiency effectively solves the problem of insufficient control accuracy caused by the solidification of parameter fusion rules in traditional methods.

[0075] Through the above technical solutions, the embodiments of the present invention can reversely optimize the weighted fusion rule of vehicle operating parameters according to the real-time energy recovery effect, improving the matching degree between the dynamic condition index and the real braking demand. Among them, the deviation calculation based on the theoretical model can accurately identify the failure scenarios of the weight allocation strategy, while the dynamic adjustment mechanism ensures that the weighted fusion process adapts to changes in different road conditions and driving behaviors. This adaptive weight optimization method can improve the determination accuracy of the energy recovery level, thereby maximizing the energy recovery efficiency on the premise of ensuring driving safety.

[0076] The following combination of attached Figure 2 , shows the processing flowchart of a multi-level energy recovery control method for an operating vehicle provided by some other embodiments of this specification, specifically including the following steps.

[0077] Step 201: Obtain the vehicle speed, acceleration, slope angle, and battery load rate in the vehicle operating parameters.

[0078] Step 202: Perform fuzzy logic weighted fusion on the vehicle speed, acceleration, and slope angle to obtain a dynamic condition index. Among them, the product of the acceleration and the vehicle speed is assigned a preset first weight, the slope angle is assigned a preset second weight, and the battery load rate is assigned a preset third weight.

[0079] Step 203: Obtain the driving operation data of the current driver within a preset time window and import it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained by the historical driving operation data of the current driver.

[0080] Step 204: Determine the corresponding energy recovery level according to the dynamic condition index and the braking demand probability level.

[0081] Step 205: When the energy recovery level is medium, calculate the current vehicle speed reduction rate.

[0082] Step 206: Based on the battery temperature parameter and the vehicle speed reduction rate, use the interpolation algorithm to generate a smooth torque change curve, where the torque increase rate is positively correlated with the vehicle speed reduction rate.

[0083] Step 207: And control the vehicle to perform energy recovery based on the dynamic torque change curve.

[0084] In some embodiments, the specific implementation and the technical effects brought by the corresponding steps in the embodiments corresponding to steps 20aa - 20bb can refer to Figure 1 the corresponding steps in Figure 1 , and will not be elaborated here.

[0085] Corresponding to the above method embodiments, this specification also provides embodiments of a multi-level energy recovery control device for an operating vehicle.Figure 3 The figure shows a schematic structural diagram of a multi-level energy recovery control device for an operating vehicle provided by some embodiments of this specification. As Figure 3 shown, the device includes: A weighted fusion processing module 301, configured to obtain vehicle operation parameters of the operating vehicle and perform weighted fusion processing to generate a dynamic working condition index; A first generation module 302, configured to obtain driving operation data of the current driver within a preset time window and import it into a preset behavior prediction model to generate a braking demand probability level, where the behavior prediction model is trained by historical driving operation data of the current driver; A determination module 303, configured to determine a corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level; A second generation module 304, configured to, when the energy recovery level is medium, generate a dynamic torque change curve based on the vehicle operation parameters and control the vehicle to which it belongs to perform energy recovery based on the dynamic torque change curve.

[0086] In some embodiments, performing weighted fusion processing on vehicle operation parameters to generate a dynamic working condition index includes: Obtaining the vehicle speed, acceleration, slope angle, and battery load rate in the vehicle operation parameters; Performing fuzzy logic weighted fusion on the vehicle speed, acceleration, and slope angle to obtain a dynamic working condition index, where the product of the acceleration and the vehicle speed is assigned a preset first weight, the slope angle is assigned a preset second weight, and the battery load rate is assigned a preset third weight.

[0087] In some embodiments, obtaining driving operation data of the current driver within a preset time window and importing it into a preset behavior prediction model to obtain a braking demand probability level includes: Obtaining driving operation data of the current driver within a preset time window, where the driving operation data includes the accelerator pedal position, vehicle speed variance, and steering angle standard deviation; Importing the accelerator pedal position, vehicle speed variance, and steering angle standard deviation into the behavior prediction model to generate a braking demand probability level, where the prediction model is a long short-term memory neural network.

[0088] In some embodiments, the energy recovery levels include low, medium, and high. Determining the corresponding energy recovery level according to the dynamic working condition index and the braking demand probability level includes, When the dynamic working condition index is not greater than a preset first threshold and the braking demand probability level is low, the energy recovery level is low, and a shutdown recovery instruction is generated and output; When the dynamic working condition index is greater than a preset second threshold and the braking demand probability level is high, the energy recovery level is high, and a preloading energy storage capacitor instruction and a maximum recovery torque instruction are generated and output; When the dynamic working condition index is greater than the first threshold and the braking demand probability level is medium or high, the energy recovery level is medium; When the dynamic working condition index is not greater than the second threshold and the braking demand probability level is low or medium, the energy recovery level is medium; When the dynamic working condition index is not greater than the first threshold and less than the second threshold, the energy recovery level is medium.

[0089] In some embodiments, the curve optimization process includes: Calculate the current vehicle speed decrease rate; Based on the battery temperature parameter and the vehicle speed decrease rate, use an interpolation algorithm to generate a smoothly transitioning torque change curve, where the torque increase rate is positively correlated with the vehicle speed decrease rate.

[0090] In some embodiments, the device further includes an adjustment module configured to: when the slope angle in the vehicle operating parameters exceeds a preset third threshold, reduce the maximum recovery torque based on a preset first ratio; When the yaw angular velocity in the vehicle operating parameters exceeds a preset fourth threshold and the energy recovery level is high, downgrade the energy recovery level to medium.

[0091] In some embodiments, the device further includes an adjustment module configured to: Calculate the theoretical recovered energy; Perform a deviation calculation process on the obtained actual recovered energy and the theoretical recovered energy to obtain a weight correction coefficient; Dynamically adjust the first weight, the second weight, and the third weight ratio according to the weight correction coefficient.

[0092] The above is a schematic solution of a multi-level energy recovery control device for an operating vehicle in this embodiment. It should be noted that the technical solution of the multi-level energy recovery control device for the operating vehicle belongs to the same concept as the technical solution of the above-mentioned multi-level energy recovery control method for the operating vehicle. For the details not described in the technical solution of the multi-level energy recovery control device for the operating vehicle, reference can be made to the description of the technical solution of the above-mentioned multi-level energy recovery control method for the operating vehicle.

[0093] In some embodiments, according to the third aspect of the embodiments of the present invention, a vehicle is provided. The vehicle is provided with a control center and a hydraulic braking module AHB. The hydraulic braking module AHB is provided with a brake controller unit ECU, a supercharging unit PSU, and a hydraulic control unit PCU, where, The control center is used to execute the steps of the multi-level energy recovery control method for operating a vehicle according to any one of claims 1 to 7; The brake controller unit ECU is respectively connected to the supercharging unit PSU and the hydraulic control unit PCU, and is used to receive the control instructions sent by the upper-level control center, and generate a first instruction and a second instruction according to the control instructions to control the supercharging unit PSU and the hydraulic control unit PCU to execute pressure control. Among them, the first instruction is sent to the supercharging unit PSU, and the second instruction is sent to the supercharging unit PSU; The supercharging unit PSU is used to output hydraulic oil with a certain pressure to the hydraulic control unit PCU according to the received first instruction; The hydraulic control unit PCU is used to adjust, control, and distribute the output pressure according to the received second instruction to ensure different working pressures under different working conditions.

[0094] Figure 4 The structural block diagram of a computing device 400 provided according to some embodiments of this specification is shown. The components of the computing device 400 include but are not limited to a memory 401 and a processor 402. The processor 402 is connected to the memory 401 through a bus 403, and the database 405 is used to store data.

[0095] The computing device 400 further includes an access device 404, and the access device 404 enables the computing device 400 to communicate via one or more networks 406. Examples of these networks include the Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device 404 may include one or more of any type of wired or wireless network interfaces (for example, a network interface card (NIC)), such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC).

[0096] In one embodiment of this specification, the above components of the computing device 400, as well as Figure 4 other components not shown in Figure 4 the computing device structure block diagram shown are also connected to each other, for example, through a bus. It should be understood that

[0097] the computing device 400 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.) or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC, Personal Computer). The computing device 400 can also be a mobile or stationary server.

[0098] Among them, the processor 402 is used to execute the following computer-executable instructions, and when the computer-executable instructions are executed by the processor, the steps of the above-mentioned multi-level energy recovery control method for an operating vehicle are implemented. The above is a schematic solution of a computing device in this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the above-mentioned multi-level energy recovery control method for an operating vehicle belong to the same concept. For the details not described in detail in the technical solution of the computing device, reference can be made to the description of the technical solution of the above-mentioned multi-level energy recovery control method for an operating vehicle.

[0099] One embodiment of this specification also provides a computer-readable storage medium, which stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the above-mentioned multi-level energy recovery control method for an operating vehicle are implemented.

[0100] The above is a schematic solution of a computer-readable storage medium in this embodiment. It should be noted that the technical solution of this storage medium and the technical solution of the above-mentioned multi-level energy recovery control method for an operating vehicle belong to the same concept. For the details not described in detail in the technical solution of the storage medium, reference can be made to the description of the technical solution of the above-mentioned multi-level energy recovery control method for an operating vehicle.

[0101] One embodiment of this specification also provides a computer program, wherein when the computer program is executed on a computer, the computer is made to execute the steps of the above-mentioned multi-level energy recovery control method for an operating vehicle.

[0102] The above is a schematic solution of a computer program according to this embodiment. It should be noted that the technical solution of this computer program and the technical solution of the above multi-level energy recovery control method for a running vehicle belong to the same concept. For the details not described in detail in the technical solution of the computer program, reference can be made to the description of the technical solution of the above multi-level energy recovery control method for a running vehicle.

[0103] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0104] Computer instructions include computer program code, which can be in the form of source code, object code, executable files, or some intermediate form, etc. A computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, external hard drives, magnetic disks, optical discs, computer memories, read-only memories (ROMs), random access memories (RAMs), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0105] It should be noted that for the foregoing method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of this specification are not limited by the described order of actions, because according to the embodiments of this specification, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments of this specification.

[0106] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0107] The preferred embodiments of the present specification disclosed above are only used to help explain the present specification. The alternative embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation manners. Obviously, many modifications and variations can be made according to the content of the embodiments of the present specification. These embodiments are selected and specifically described in the present specification in order to better explain the principles and practical applications of the embodiments of the present specification, so that those skilled in the art can well understand and utilize the present specification. The present specification is only limited by the claims and their full scope and equivalents.

Claims

1. A multi-stage energy recovery control method for a running vehicle, characterized in that: include: Obtain vehicle operating parameters of the running vehicle and perform weighted fusion processing to generate a dynamic operating condition index; Acquire the driving operation data of the current driver within a preset time window and import it into a preset behavior prediction model to generate a braking demand probability level, wherein the behavior prediction model is trained by the historical driving operation data of the current driver; Determining a corresponding energy recovery level according to the dynamic operating condition index and the braking demand probability level; When the energy recovery level is medium, a dynamic torque variation curve is generated based on the vehicle operating parameters, and the vehicle is controlled to perform energy recovery based on the dynamic torque variation curve.

2. The method according to claim 1, characterized in that The vehicle operating parameters are weighted and fused to generate a dynamic operating condition index, including: Obtaining vehicle speed, acceleration, slope angle, and battery load rate from the vehicle operating parameters; The vehicle speed, acceleration and slope angle are subjected to fuzzy logic weighted fusion to obtain the dynamic operating condition index, wherein the product of the acceleration and the vehicle speed is assigned a preset first weight, the slope angle is assigned a preset second weight, and the battery load rate is assigned a preset third weight.

3. The method according to claim 1, characterized in that The step of obtaining the driving operation data of the current driver within a preset time window and importing the preset behavior prediction model to obtain the braking demand probability level includes: Acquiring driving operation data of the current driver within a preset time window, wherein the driving operation data includes accelerator pedal position, vehicle speed variance, and steering angle standard deviation; The accelerator pedal position, vehicle speed variance and steering angle standard deviation are introduced into the behavior prediction model to generate the braking demand probability level, wherein the prediction model is a long short-term memory neural network.

4. The method according to claim 1, characterized in that The energy recovery level includes low, medium and high levels. According to the dynamic operating condition index and the braking demand probability level, the corresponding energy recovery level is determined, including: When the dynamic operating condition index is not greater than a preset first threshold value and the braking demand probability level is low, the energy recovery level is low, and a closing recovery instruction is generated and output; When the dynamic operating condition index is greater than a preset second threshold value and the braking demand probability level is high, the energy recovery level is high, and a preload energy storage capacitor instruction and a maximum recovery torque instruction are generated and output; When the dynamic operating condition index is greater than the first threshold and the braking demand probability level is medium or high, the energy recovery level is medium; When the dynamic operating condition index is not greater than the second threshold and the braking demand probability level is low or medium, the energy recovery level is medium; When the dynamic operating condition index is not greater than the first threshold and less than the second threshold, the energy recovery level is medium.

5. The method according to claim 1, characterized in that The curve optimization process comprises: Calculate the current vehicle speed decrease rate; Based on the battery temperature parameters and the vehicle speed decrease rate, an interpolation algorithm is used to generate a smooth transition torque change curve, in which the torque increase rate is positively correlated with the vehicle speed decrease rate.

6. The method according to claim 1, characterized in that The method further comprises: When the slope angle in the vehicle operating parameter exceeds a preset third threshold, the maximum recovery torque is reduced based on a preset first ratio; When the yaw rate in the vehicle operating parameters exceeds a preset fourth threshold and the energy recovery level is high, the energy recovery level is downgraded to medium.

7. The method according to claim 2, characterized in that: Also includes: Calculate theoretical recovery energy; Performing deviation calculation processing on the actual recovered energy obtained and the theoretical recovered energy to obtain a weight correction coefficient; The ratio of the first weight, the second weight and the third weight is dynamically adjusted according to the weight correction coefficient.

8. A multi-stage energy recovery control device for a running vehicle, characterized in that: include: A weighted fusion processing module is configured to obtain vehicle operating parameters of the running vehicle and perform weighted fusion processing to generate a dynamic operating condition index; A first generating module is configured to obtain driving operation data of the current driver within a preset time window and import a preset behavior prediction model to generate a braking demand probability level, wherein the behavior prediction model is trained by the historical driving operation data of the current driver; a determination module configured to determine a corresponding energy recovery level according to the dynamic operating condition index and the braking demand probability level; The second generating module is configured to generate a dynamic torque change curve based on the vehicle operating parameters when the energy recovery level is medium, and control the vehicle to perform energy recovery based on the dynamic torque change curve.

9. A vehicle, characterized in that: The vehicle is provided with a control center and a hydraulic brake module AHB, wherein the hydraulic brake module AHB is provided with a brake controller unit ECU, a boost unit PSU and a hydraulic control unit PCU, wherein: The control center is used to execute the steps of the multi-stage energy recovery control method for operating a vehicle according to any one of claims 1 to 7; The brake controller unit ECU is connected to the boost unit PSU and the hydraulic control unit PCU respectively, and is used to receive the control command sent by the superior control center, and generate the first command and the second command according to the control command to control the boost unit PSU and the hydraulic control unit PCU to perform pressure control, wherein the first command is sent to the boost unit PSU, and the second command is sent to the boost unit PSU; The pressure boosting unit PSU is used to output hydraulic oil of a certain pressure to the hydraulic control unit PCU according to the received first instruction; The hydraulic control unit PCU is used to adjust, control and distribute the output pressure according to the received second instruction to ensure that different working pressures are achieved under different working conditions.

10. A computer-readable storage medium storing computer-executable instructions, characterized in that: When the computer executable instructions are executed by a processor, the steps of the multi-stage energy recovery control method for operating a vehicle as described in any one of claims 1 to 7 are implemented.

Images