Range-extending type automobile braking energy recovery optimization method based on Internet of Vehicles

Abstract

The invention provides an extended-range automobile braking energy recovery optimization method based on the Internet of Vehicles, and belongs to the technical field of new energy automobiles. The method comprises the following steps: S1, acquiring road traffic information by using the Internet of Vehicles, and performing feature extraction to serve as input of a vehicle speed prediction model; s2, building an extended range type automobile energy management system model; s3, performing feature fusion on the obtained road traffic information by using an LSTM network, constructing a vehicle speed prediction model, and predicting the vehicle speed; and S4, according to the predicted vehicle speed, through a prediction-decision-execution closed-loop control framework, brake energy recovery is optimized, and the optimized energy distribution relation is fed back to the extended range type vehicle energy management system model. By the adoption of the method, the braking energy recovery efficiency in the extended range driving mode can be improved, the battery overcharge risk is avoided, the service life of a power battery is prolonged, and fuel consumption and carbon emission are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and particularly to an optimization method for regenerative braking energy recovery of an extended-range vehicle based on the vehicle Internet of Things. Background Art

[0002] Currently, the global energy and environmental problems are becoming increasingly serious. The emissions of vehicle greenhouse gases have a significant impact on the environment, which has led automobile manufacturers to turn their attention to the research of hybrid vehicles, electric vehicles, and fuel cell vehicles. As a transitional model from traditional vehicles to future clean vehicles, hybrid vehicles, with the rapid development of new energy vehicle technologies, extended-range electric vehicles (EREVs) have become an important choice in the current new energy vehicle market due to their advantages of both the high efficiency of electric drive and the extended range of the engine.

[0003] Regenerative braking technology is an important energy recovery technology in electric vehicles and hybrid vehicles. Through the reverse action of the motor, regenerative braking technology converts kinetic energy into electrical energy and stores it in the battery for subsequent acceleration, thereby reducing energy waste and brake system wear. The regenerative braking energy management strategy is to reasonably distribute the braking force between the motor and the traditional mechanical brake according to the operating conditions of the vehicle, with the goal of maximizing energy recovery and optimizing braking performance. This strategy plays a core role in the energy management of extended-range vehicles and is of great significance for improving the energy utilization efficiency and environmental protection performance of vehicles. However, during the braking process in the extended-range driving mode of extended-range vehicles, the range extender starts to charge the power battery, resulting in limited regenerative braking energy recovery ability due to the charging power limit of the power battery, which affects the overall vehicle energy utilization efficiency. Therefore, there is an urgent need for a new regenerative braking energy recovery control strategy for extended-range vehicles to improve the efficiency of regenerative braking energy recovery in the extended-range model. Summary of the Invention

[0004] The purpose of the present invention is to provide an optimization method for regenerative braking energy recovery of an extended-range vehicle based on the vehicle Internet of Things, which uses the vehicle Internet of Things to obtain road traffic information to optimize the control strategy, improve the energy recovery efficiency during vehicle braking, and extend the service life of the power battery.

[0005] To achieve the above purpose, the present invention provides an optimization method for regenerative braking energy recovery of an extended-range vehicle based on the vehicle Internet of Things, including the following steps:

[0006] Step S1: Use the vehicle Internet of Things to obtain road traffic information and perform feature extraction, which is used as the input of the vehicle speed prediction model;

[0007] Step S2: Build an energy management system model for the extended-range vehicle;

[0008] Step S3: Use the LSTM network to perform feature fusion on the obtained road traffic information, construct a vehicle speed prediction model, and predict the vehicle speed;

[0009] Step S4: According to the predicted vehicle speed, optimize the braking energy recovery through a prediction-decision-execution closed-loop control architecture, and feedback the optimized energy distribution relationship to the range extender vehicle energy management system model.

[0010] Preferably, in step S1, the road traffic information includes surrounding environment vehicle information and the vehicle's own historical information, specifically the vehicle's own historical speed, acceleration, the historical speed and acceleration of the vehicle in front, the distance between the vehicle and the vehicle in front, and the average traffic flow speed.

[0011] Preferably, in step S2, build a range extender vehicle energy management system model, including the following steps:

[0012] Step S21: Build a vehicle model;

[0013] Step S22: Build an energy management controller model.

[0014] Preferably, in step S21, the vehicle model includes: a demand power calculation module, a range extender power output module, and a battery state of charge calculation module;

[0015] The demand power calculation module is as follows:

[0016]

[0017] Among them, P req represents the demand power, v represents the speed, η represents the motor efficiency, m represents the vehicle weight, F w represents the air resistance, F f represents the rolling friction, F i represents the gradient resistance;

[0018] The range extender power output module is as follows:

[0019]

[0020] Among them, P g represents the range extender output power, P b represents the power battery power, T g , n g , η g respectively represent the torque, speed, and efficiency of the range extender;

[0021] The battery state of charge calculation module is as follows:

[0022] The output of the range extender is multiple fixed power output points pg1 , p g2 , …, p gn , the output power of the range extender will switch between multiple operating points;

[0023]

[0024] Among them, SOC t represents the battery charge state at time t, SOC0 represents the initial battery charge state, I b represents the output current of the power battery, and C represents the total capacity of the battery.

[0025] Preferably, the regenerative braking torque of the energy management controller model is limited by the following conditions:

[0026]

[0027] T reg_max = min{f d (n d ), T reg_cmax};

[0028] Among them, P reg_cmax , P cmax and P c respectively represent the maximum regenerative braking power limited by the battery, the maximum battery charging power, and the current battery charging power; T reg_cmax represents the maximum regenerative braking torque under the charging power limit; n d represents the current rotational speed of the drive motor; i0 represents the transmission ratio; r represents the wheel radius; f d (n d ) represents the maximum regenerative braking torque limited by the external characteristic curve of the drive motor; T reg_max represents the maximum regenerative braking torque after limitation.

[0029] Preferably, in step S3, the following steps are included:

[0030] According to the information collected by the vehicle network, taking the road traffic information as the input, select the features of the vehicle's historical speed V n = [v1, v2, v3,..., v n , the vehicle's historical acceleration A n = [a1, a2, a3,..., a n , the leading vehicle's historical speed V n_f = [v 1_f , v 2_f , v 3_f ,..., v n_f , the leading vehicle's historical acceleration A n_f = [a 1_f , a2_f , a 3_f ,..., a n_f , the distance D between this vehicle and the vehicle ahead n = [d1, d2, d3,..., d n and the average vehicle speed V of the traffic flow n_a = [v 1_a , v 2_a , v 3_a ,..., v n_a are used as the inputs of the LSTM vehicle speed prediction model, and feature fusion is performed on the time series data to output the predicted vehicle speed sequence V n_pre = [v 1_pre , v 2_pre , v 3_pre ,..., v n_pre , where n represents the nth data.

[0031] Preferably, in step S4, the following steps are included:

[0032] Step S41: Dynamically control the start and stop of the range extender according to the state of charge SOC of the power battery. When the SOC is lower than the first threshold, start the range extender; when the SOC is higher than the second threshold, turn off the range extender. The power output point of the basic layer range extender is determined by the current vehicle speed. In different vehicle speed ranges V j [v j1 , v j2 , the power output point of the range extender is P gj ;

[0033] Step S42: Estimate the future braking intensity based on the predicted vehicle speed sequence and judge the braking intention;

[0034] Step S43: Optimize the power ratio between the range extender and the regenerative braking according to the braking intensity and the battery charging power limit.

[0035] Preferably, in step S41, the first threshold is SOC < 0.3, and the second threshold is SOC ≥ 0.5.

[0036] Preferably, in step S42, the calculation formula for the braking intensity is:

[0037]

[0038] where Z pre represents the predicted braking intensity. When Z pre < 0, it indicates an intention to accelerate; g represents the acceleration due to gravity, and v pre represents the predicted vehicle speed.

[0039] Preferably, in step S43, the optimization strategy is as follows: under the limitation of the battery charging power, the braking energy recovery power is preferentially increased, and the output power of the range extender is dynamically adjusted according to the vehicle speed prediction error.

[0040] Therefore, the present invention adopts the above-mentioned method for optimizing the braking energy recovery of a range-extended electric vehicle based on vehicle networking, and the beneficial technical effects are as follows:

[0041] (1) Through forward-looking energy management, adjusting the power ratio between the range extender and the braking energy recovery can significantly improve the braking energy recovery efficiency in the range-extended driving mode;

[0042] (2) Avoid the saturation of the battery charging power during braking and reduce the risk of overcharging the battery;

[0043] (3) Adopt a hierarchical control architecture, where the basic layer ensures real-time performance and the prediction layer provides optimization, enhancing the responsiveness and stability of the system;

[0044] (4) It has significant environmental benefits. By improving the energy utilization efficiency, it reduces fuel consumption and carbon emissions. Description of the Drawings

[0045] Figure 1 It is a flowchart of a method for optimizing the braking energy recovery of a range-extended electric vehicle based on vehicle networking according to the present invention;

[0046] Figure 2 It is a control flowchart of an energy management controller model;

[0047] Figure 3 It is a flowchart of the start-up rule of the braking energy recovery system;

[0048] Figure 4 It is a comparison chart of the braking energy recovery in the range-extended driving mode before and after the optimization of the braking energy recovery of a range-extended electric vehicle;

[0049] Figure 5 It is a graph of the change in the battery SOC before and after the optimization of the braking energy recovery of a range-extended electric vehicle. Detailed Embodiments

[0050] The technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0051] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the art to which the present invention belongs.

[0052] Embodiment 1

[0053] In this embodiment, a certain range-extended electric vehicle is taken as the prototype, and its power system parameters are used as the experimental data of this embodiment. Part of the data in Lankershim Boulevard and Peachtree Street in the NGSIM dataset is used as the training data for the vehicle speed prediction model in this embodiment, and part of the vehicle data is used as the verification data for this embodiment.

[0054] As Figure 1 shown, the present invention proposes an optimization method for braking energy recovery of a range-extended vehicle based on the vehicle networking, which takes into account both the braking recovery efficiency and the battery life by integrating real-time traffic information and vehicle speed prediction technology. The method specifically includes the following steps:

[0055] Step S1: Obtain road traffic information through the vehicle networking and perform feature extraction, including: the historical speed and acceleration of the vehicle itself, the historical speed and acceleration of the vehicle in front, the distance between the vehicle itself and the vehicle in front, and the average traffic flow speed.

[0056] Step S2: Build a model of the energy management system for the range-extended vehicle;

[0057] Step S21: Build a vehicle model, which includes: a demand power calculation module, a range extender power output module, and a battery state of charge (SOC) calculation module;

[0058] The demand power calculation module is as follows:

[0059]

[0060] Among them, P req represents the demand power, v represents the speed, η represents the motor efficiency, m represents the weight of the vehicle, F w represents the air resistance, F f represents the rolling friction, F i represents the gradient resistance;

[0061] The range extender power output module is as follows:

[0062]

[0063] Among them, P g represents the range extender output power, P b represents the power of the power battery, T g , n g , η g respectively represent the torque, speed, and efficiency of the range extender.

[0064] The battery state of charge calculation module is as follows:

[0065] The output of the range extender is multiple fixed power output points p g1 , p g2 , …, p gn , and the output power of the range extender will switch between multiple operating points;

[0066]

[0067] Among them, SOC0 represents the initial battery charge state; SOC t represents the battery charge state at time t; C represents the total capacity of the battery; I b represents the output current of the power battery.

[0068] Step S22, build an energy management controller model

[0069] As Figure 2 shown is the control flow chart of the energy management controller model. The controller calculates the current vehicle demand power based on the pedal signal and the current vehicle speed; if the demand power P req is greater than 0, it means the vehicle needs to accelerate or maintain the vehicle speed. At this time, it is decided whether to use the power battery to provide the required power (pure electric mode) or start the range extender according to the current SOC state of the power battery. If the SOC state of the power battery is lower than the lower limit value SOC min (minimum charge state), start the range extender, and then judge whether the demand power exceeds the maximum power that the range extender can provide to decide whether to use the hybrid drive mode or the range extender mode; if the demand power is less than 0, it means the vehicle needs to brake. At this time, it is necessary to distribute the front and rear axle braking forces according to the vehicle driving state, and whether to use the motor to recover braking energy. If the regenerative braking condition is met, switch between pure electric braking and hybrid braking according to the maximum regenerative braking power P reg_max , otherwise use mechanical braking.

[0070] As Figure 3 shown is the start rule flow chart of the braking energy recovery system. When the SOC of the power battery exceeds 90%, no energy recovery is performed to prevent the SOC of the power battery from being too high and damaging the battery life; no energy recovery is performed when the vehicle speed is too low. When the vehicle speed v is lower than 10 km / h, the reverse electromotive force provided by the drive motor is not enough to provide the minimum charging power of the power battery, and no energy recovery will be performed; when the vehicle is in emergency braking and the braking intensity Z > 0.7, in order to prevent the braking response speed of the motor from not meeting the emergency braking, all mechanical braking will be used, and hybrid braking will be used when P reg_max is less than P req .

[0071] More specifically, first judge whether the battery charge state is greater than 0.9.

[0072] If the state of charge (SOC) > 0.9, to prevent overcharging of the battery, mechanical braking is directly adopted. Otherwise, it continues to judge whether the braking intensity is greater than 0.7.

[0073] If Z > 0.7, that is, the vehicle is in emergency braking. At this time, all mechanical braking is adopted because electric braking may not be able to meet the emergency braking requirements. Otherwise, it continues to judge whether the vehicle speed is less than 10 km / h.

[0074] If v < 10 km / h, since the back electromotive force of the drive motor may not be able to meet the minimum charging power requirement of the power battery at this time, mechanical braking is adopted. Otherwise, it continues to judge the actual required braking torque T r whether it is greater than the maximum regenerative braking torque that the motor can provide.

[0075] If T r > T reg_max , relying only on electric braking cannot meet the braking requirements, and hybrid braking is required. Otherwise, electric braking is adopted.

[0076] The regenerative braking torque should be limited by the motor state and the charging power of the power battery:

[0077]

[0078] T reg_max = min{f d (n d ), T reg_cmax};

[0079] Among them, P reg_cmax , P cmax and P c respectively represent the maximum regenerative braking power limited by the battery, the maximum charging power of the battery, and the current charging power of the battery; T reg_cmax represents the maximum regenerative braking torque under the charging power limit; n d represents the current rotational speed of the drive motor; i0 represents the transmission ratio; r represents the wheel radius; f d (n d ) represents the maximum regenerative braking torque limited by the external characteristic curve of the drive motor; T reg_max represents the maximum regenerative braking torque after limitation.

[0080] Step S3: Use the LSTM network to perform feature fusion on the obtained road traffic information, construct a vehicle speed prediction model, and predict the vehicle speed within the time steps of the next 3 - 10 s, specifically including:

[0081] According to the information collected by the vehicle network, taking the surrounding road traffic characteristics and historical information as inputs, the selected features are the vehicle's historical speed V n = [v1, v2, v3,..., vn 1. The historical acceleration A of this vehicle n = [a1, a2, a3,..., a n , the historical speed V of the vehicle ahead n_f = [v 1_f , v 2_f , v 3_f ,..., v n_f , the historical acceleration A of the vehicle ahead n_f = [a 1_f , a 2_f , a 3_f ,..., a n_f , the distance D between this vehicle and the vehicle ahead n = [d1, d2, d3,..., d n and the average speed V of the traffic flow n_a = [v 1_a , v 2_a , v 3_a ,..., v n_a are used as the inputs of the LSTM vehicle speed prediction model, and the time series data are subjected to feature fusion to output the predicted vehicle speed sequence V n_pre = [v 1_pre , v 2_pre , v 3_pre ,..., v n_pre , where n represents the nth data.

[0082] Step S4. According to the predicted vehicle speed, optimize the braking energy recovery through the prediction - decision - execution closed - loop control architecture, and feedback the optimized energy distribution relationship to the extended - range electric vehicle energy management system model, specifically including:

[0083] Step S41. Basic layer control strategy: Dynamically control the start and stop of the range extender according to the state of charge SOC of the power battery. Start the range extender when SOC is lower than the first threshold, and stop the range extender when SOC is higher than the second threshold. The first threshold is SOC < 0.3, and the second threshold is SOC ≥ 0.5.

[0084] The power output point of the basic layer range extender is determined by the current vehicle speed. The control layer divides the vehicle speed into different intervals V1[0, 50], V2[50, 80], V3[80, 120]. The basic layer selects the power output point P j [v j1 , v j2 at different vehicle speed intervals V gj , v j1 is the lower limit of the starting speed of this interval, and v j2 is the upper limit of the starting speed of this interval. Specifically, select the output power point as P g2 at V1, and P at V2g3 , it is P under V3 g4 .

[0085] Step S42, regenerative braking power prediction: Estimate the future braking intensity based on the predicted vehicle speed sequence, and judge the braking intention. The calculation formula of the braking intensity is as follows:

[0086]

[0087] where, Z pre represents the predicted braking intensity. When Z pre <0 indicates an acceleration intention; g represents the acceleration due to gravity, and v pre represents the predicted vehicle speed.

[0088] Step S43, regenerative braking optimization: Adjust the power ratio of the range extender and braking recovery based on the predicted braking intensity and the power point of the range extender selected by the basic layer. When Z pre >0:

[0089]

[0090] where, F f_pre represents the predicted front wheel braking force, L, L b and h g respectively represent the wheelbase of the vehicle, the rear wheelbase, and the center of mass height, T f_pre represents the predicted front wheel braking torque, and P f_pre represents the predicted braking recovery energy power.

[0091] P tmp =P cmax -P f_pre ;

[0092] where, P tmp represents the target power of the range extender;

[0093] Perform error feedback on the result according to the vehicle speed prediction error:

[0094]

[0095] where, e v represents the vehicle speed prediction error, α represents the dynamic weight, represents the target power of the optimization strategy, P g represents the current output power of the range extender, e p represents the power error reference value, and P gj represents the output power point of the range extender determined according to the current vehicle speed range.

[0096] When e p ≥0.4(P gj -Pgj-1 ) When adjusting the power point of the range extender to P gj-1 .

[0097] Among them, in this embodiment, the operating conditions of three vehicles in the Peachtree section of the NGSIM dataset are used as verification operating conditions. The braking recovery energy and battery SOC under the range-extended driving mode before and after optimization are compared. The results are as follows Figure 4 and Figure 5 shown. Both the braking recovery energy and charging efficiency under the range-extended driving mode have been improved.

[0098] Therefore, by adopting the above-mentioned method for optimizing the braking energy recovery of a range-extended electric vehicle based on the vehicle network, the present invention can significantly improve the braking energy recovery efficiency under the range-extended driving mode, avoid the risk of overcharging the battery, extend the service life of the power battery, and reduce fuel consumption and carbon emissions, having significant environmental benefits and economic value.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking, characterized in that: The following steps are involved: Step S1, using the Internet of Vehicles to obtain road traffic information and perform feature extraction as input to a vehicle speed prediction model; Step S2, building an extended-range vehicle energy management system model; Step S3: Using the LSTM network to perform feature fusion on the acquired road traffic information, construct a vehicle speed prediction model, and predict the vehicle speed; Step S4: According to the predicted vehicle speed, the braking energy recovery is optimized through a prediction-decision-execution closed-loop control architecture, and the optimized energy distribution relationship is fed back to the extended-range vehicle energy management system model.

2. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 1 is characterized in that: In step S1, the road traffic information includes surrounding vehicle information and vehicle history information, specifically the historical speed and acceleration of the vehicle, the historical speed and acceleration of the preceding vehicle, the distance between the vehicle and the preceding vehicle, and the average speed of the traffic flow.

3. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 1 is characterized in that: In step S2, building a range-extended vehicle energy management system model includes the following steps: Step S21, building a whole vehicle model; Step S22: building an energy management controller model.

4. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 3 is characterized in that: In step S21, the vehicle model includes: a required power calculation module, a range extender power output module and a battery charge state calculation module; The required power calculation module is as follows: Among them, P req represents the required power, v represents the speed, η represents the motor efficiency, m represents the weight of the vehicle, F w Indicates air resistance, F f Denotes rolling friction, F i represents slope resistance; The range extender power output module is as follows: Among them, P g Indicates the output power of the range extender, P b Indicates the power of the power battery, T g 、n g , η g They represent the torque, speed and efficiency of the range extender respectively; The battery charge state calculation module is as follows: The output of the range extender is multiple fixed power output points p g1 , p g2 ,…,p gn , the output power of the range extender will switch between multiple operating points; Among them, SOC t represents the battery charge state at time t, SOC0 represents the initial battery charge state, I b It represents the output current of the power battery, and C represents the total capacity of the battery.

5. According to the method for optimizing the braking energy recovery of an extended-range vehicle based on the Internet of Vehicles according to claim 4, in step S22, the regenerative braking torque of the energy management controller model is limited by the following conditions: T reg_max =min{f d (n d ),T reg_cmax }; in, P reg_cmax , P cmax and P c They represent the maximum regenerative braking power limited by the battery, the maximum charging power of the battery and the current charging power of the battery respectively; T reg_cmax Indicates the maximum regenerative braking torque under charging power limitation; n d represents the current speed of the drive motor; i0 represents the transmission ratio; r represents the wheel radius; f d (n d ) represents the maximum regenerative braking torque under the limitation of the external characteristic curve of the drive motor; T reg_max Indicates the maximum regenerative braking torque after limitation.

6. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 1 is characterized in that: Step S3 includes the following steps: According to the information collected by the Internet of Vehicles, the road traffic information is used as input, and the selected features are the historical speed V of the vehicle n =[v1,v2,v3,...,v n ], the historical acceleration of the vehicle A n =[a1,a2,a3,...,a n ]、The previous vehicle’s historical speed V n_f =[v 1_f ,v 2_f ,v 3_f ,...,v n_f ]、History acceleration of the preceding vehicle A n_f =[a 1_f ,a 2_f ,a 3_f ,...,a n_f ]、The distance between this vehicle and the vehicle in front is D n =[d1,d2,d3,...,d n ] and the average speed of traffic flow V n_a =[v 1_a ,v 2_a ,v 3_a ,...,v n_a ] as the input of the LSTM vehicle speed prediction model, and perform feature fusion on the time series data to output the predicted vehicle speed sequence V n_pre =[v 1_pre ,v 2_pre ,v 3_pre ,...,v n_pre ], where n represents the nth data.

7. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41, dynamically controlling the start and stop of the range extender according to the state of charge SOC of the power battery, starting the range extender when the SOC is lower than a first threshold, and shutting down the range extender when the SOC is higher than a second threshold; Step S42: estimating the future braking intensity based on the predicted vehicle speed sequence and determining the braking intention; Step S43: Optimize the power ratio of the range extender and the braking recovery according to the braking intensity and the battery charging power limit.

8. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 7 is characterized in that: In step S41, the first threshold is SOC<0.3, and the second threshold is SOC≥0.

5.

9. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 8, characterized in that: In step S42, the calculation formula of the braking intensity is: Among them, Z pre Indicates the predicted braking intensity. When Z pre <0 indicates an acceleration intention; g indicates the acceleration due to gravity, v pre Indicates the predicted vehicle speed.

10. The method for optimizing braking energy recovery of an extended-range vehicle based on vehicle networking according to claim 9, characterized in that: In step S43, the optimization strategy is: give priority to increasing the braking recovery power under the battery charging power limit, and dynamically adjust the range extender output power according to the vehicle speed prediction error.

Images