Vehicle queue anti-drag following driving control method considering wake flow influence of external flow field

Abstract

The invention discloses a vehicle queue anti-drag following driving control method considering the wake flow influence of an external flow field. The method comprises the following steps: 1) obtaining an equivalent wind resistance coefficient Cd < eq > of a rear vehicle during following driving; 2) constructing an equivalent wind resistance coefficient estimation model under the wake flow influence; 3) constructing a car-following driving control model based on an equivalent wind resistance coefficient estimation model under wake flow influence; and 4) solving the car-following driving control model to obtain a car-following driving control scheme. According to the drag reduction car-following strategy provided by the invention, the car-following safety can be ensured, the aerodynamic resistance of the car-following vehicle can be effectively reduced, and theoretical and technical supports are provided for realizing the energy-saving driving technology of the intelligent networked vehicle.

Description

Technical Field

[0001] The invention relates to the technical field of vehicle following control, in particular to a vehicle queue drag reduction following driving control method taking into account the influence of external flow field wake. Background Art

[0002] With the consumption of a large amount of non-renewable energy and the rapid growth of electricity demand, energy consumption and environmental issues have become important issues that all mankind faces and urgently needs to solve. The rapid development of the vehicle industry in recent years has made the contradiction between energy supply and consumption very acute, so this issue has received more and more attention.

[0003] When vehicles are platooning, a smaller distance between vehicles can greatly reduce air resistance and improve vehicle fuel economy. This is because the wake of the leading vehicle acts on the front of the following vehicle to reduce the air resistance of the following vehicle. However, traditional vehicles are limited by autonomous driving technology and cannot achieve safe following driving with a small distance. Summary of the invention

[0004] The object of the present invention is to provide a vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake, comprising the following steps:

[0005] 1) Obtain the equivalent drag coefficient Cd of the following vehicle during car-following driving eq ;

[0006] 2) Construct an estimation model of equivalent drag coefficient under the influence of wake;

[0007] 3) Based on the equivalent drag coefficient estimation model under the influence of wake turbulence, a car-following driving control model is constructed;

[0008] 4) Solve the car-following driving control model and obtain the car-following driving control scheme.

[0009] Furthermore, the equivalent drag coefficient Cd of the following vehicle during car-following driving is eq As shown below:

[0010] Cd eq =2F follow / ρ×S×v 2 (1)

[0011] In the formula, F follow is the following wind resistance of the following vehicle at the same speed as the single vehicle, ρ is the air density, S is the frontal area, and v is the driving speed.

[0012] Furthermore, the equivalent drag coefficient estimation model under the influence of the wake is as follows:

[0013]

[0014] Where α, β, and γ are influencing factors; d is the distance between vehicle queues.

[0015] Furthermore, the impact factors α, β, and γ are as follows:

[0016]

[0017] Where v is the driving speed.

[0018] Furthermore, the objective function of the car-following driving control model is as follows:

[0019]

[0020] In the formula, J opt (k) is the total cost function; λ, δ, μ is the weighting factor; J L (k) is the security cost function, J Cd (k) is the low-resistance cost function, J v (k) is the follow-up cost function, J a (k) is the comfort cost function; N p is the prediction time range; k is the time step.

[0021] Furthermore, the security cost function J L (k) is as follows:

[0022] L min (k) = 0.9 + 0.15v (k) + 0.0325v 2 (k) (7)

[0023] L max (k) = 7 + v (k) + 0.05v 2 (k) (8)

[0024] J L (k)=||L max (k)-L(k)|-|L min (k)-L(k)|| (9)

[0025] Where, L max , L min are the upper and lower limits of the distance between the front and rear vehicles; v(k) is the driving speed corresponding to time step k; L(k) is the distance between the front and rear vehicles corresponding to time step k.

[0026] Furthermore, the follow-up cost function J v (k) is as follows:

[0027] J V (k)=|V f (k)-Ve (k)| (10)

[0028] Where V f (k), V e (k) represents the speed of the front and rear vehicles respectively.

[0029] Furthermore, the comfort cost function J a (k) is as follows:

[0030] J a (k)=|a(k-1)-a(k)| (11)

[0031] Where a(k-1) and a(k) are the accelerations corresponding to time step k-1 and time step k.

[0032] Furthermore, the low-resistance cost function J Cd (k) is as follows:

[0033] J Cd (k) = Cd eq (k) / Cd (12)

[0034] In the formula, Cd eq (k) is the equivalent drag coefficient of the rear vehicle; Cd is the physical drag coefficient of a single vehicle.

[0035] Furthermore, the equivalent drag coefficient of the rear vehicle Cd eq The upper limit of (k) is a preset fixed value, and the lower limit is the change in slack as the speed of the preceding vehicle changes.

[0036] Equivalent drag coefficient of the rear vehicle Cd eq (k) The functional relationship between speed v and vehicle spacing d is as follows:

[0037]

[0038] The technical effect of the present invention is undoubted, and the beneficial effects of the present invention are as follows:

[0039] 1) This invention discusses the influence of the speed of the leading vehicle and the distance between the vehicle queues on the equivalent drag coefficient of the following vehicle, and based on this estimation model, integrates the vehicle following control strategy, and proposes a drag reduction following control strategy based on the influence of the leading vehicle's wake. The drag reduction following strategy proposed in this invention can not only ensure the safety of following, but also effectively reduce the aerodynamic resistance of the following vehicle, providing theoretical and technical support for the realization of energy-saving driving technology for intelligent connected vehicles.

[0040] 2) In order to optimize the safety and low resistance of following driving under actual driving conditions, the present invention introduces a soft constraint on the equivalent drag coefficient. Compared with the traditional hard constraint control strategy, the total cost function optimization process is improved to pursue the minimum value of the total cost function. When the following distance exceeds the lower limit of the safe distance, the queue distance is shortened to obtain better low resistance, and the better low resistance makes up for the lack of safety in the total cost function.

[0041] 3) The present invention optimizes the wind resistance experienced by the following vehicle while ensuring the safety of the vehicle, thereby improving the fuel economy of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of a vehicle queue drag reduction and following driving control method considering the influence of the external flow field wake provided by the present invention;

[0043] Figure 2 This is the result of fitting the equivalent drag coefficient function under the influence of wake;

[0044] Figure 3 is a flow chart of a car-following driving control method based on an equivalent drag coefficient estimation model;

[0045] Figure 4 This is the result diagram based on drag reduction and following control. DETAILED DESCRIPTION

[0046] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0047] Embodiment 1:

[0048] See also Figures 1 to 4 The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake includes the following steps:

[0049] 1) Obtain the equivalent drag coefficient Cd of the following vehicle during car-following driving eq ;

[0050] 2) Construct an estimation model of equivalent drag coefficient under the influence of wake;

[0051] 3) Based on the equivalent drag coefficient estimation model under the influence of wake turbulence, a car-following driving control model is constructed;

[0052] 4) Solve the car-following driving control model and obtain the car-following driving control scheme.

[0053] Equivalent drag coefficient Cd of the rear vehicle during following driving eq As shown below:

[0054] Cd eq =2F follow / ρ×S×v 2 (1)

[0055] In the formula, F follow is the following wind resistance of the following vehicle at the same speed as the single vehicle, ρ is the air density, S is the frontal area, and v is the driving speed.

[0056] The equivalent drag coefficient estimation model under the influence of the wake is as follows:

[0057]

[0058] Where α, β, and γ are influencing factors; d is the distance between vehicle queues.

[0059] The impact factors α, β, and γ are as follows:

[0060]

[0061] Where v is the driving speed.

[0062] The objective function of the car-following driving control model is as follows:

[0063]

[0064] In the formula, J opt (k) is the total cost function; λ, δ, μ is the weighting factor; J L (k) is the security cost function, J Cd (k) is the low-resistance cost function, J v (k) is the follow-up cost function, J a (k) is the comfort cost function; N p is the prediction time range; k is the time step.

[0065] Security cost function J L (k) is as follows:

[0066] L min (k) = 0.9 + 0.15v (k) + 0.0325v 2 (k) (7)

[0067] L max (k) = 7 + v (k) + 0.05v 2 (k) (8)

[0068] J L (k)=||L max (k)-L(k)|-|Lmin (k)-L(k)|| (9)

[0069] Where, L max , L min are the upper and lower limits of the distance between the front and rear vehicles; v(k) is the driving speed corresponding to time step k; L(k) is the distance between the front and rear vehicles corresponding to time step k.

[0070] Follow-up cost function J v (k) is as follows:

[0071] J V (k)=|V f (k)-V e (k)| (10)

[0072] Where V f (k), V e (k) represents the speed of the front and rear vehicles respectively.

[0073] Comfort cost function J a (k) is as follows:

[0074] J a (k)=|a(k-1)-a(k)| (11)

[0075] Where a(k-1) and a(k) are the accelerations corresponding to time step k-1 and time step k.

[0076] Low resistance cost function J Cd (k) is as follows:

[0077] J Cd (k) = Cd eq (k) / Cd(12)

[0078] In the formula, Cd eq (k) is the equivalent drag coefficient of the rear vehicle; Cd is the physical drag coefficient of a single vehicle.

[0079] Equivalent drag coefficient of the rear vehicle Cd eq The upper limit of (k) is a preset fixed value, and the lower limit is the change in slack as the speed of the preceding vehicle changes.

[0080] Equivalent drag coefficient of the rear vehicle Cd eq (k) The functional relationship between speed v and vehicle spacing d is as follows:

[0081]

[0082] Embodiment 2:

[0083] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake includes the following steps:

[0084] 1) Obtain the equivalent drag coefficient Cd of the following vehicle during car-following driving eq ;

[0085] 2) Construct an estimation model of equivalent drag coefficient under the influence of wake;

[0086] 3) Based on the equivalent drag coefficient estimation model under the influence of wake turbulence, a car-following driving control model is constructed;

[0087] 4) Solve the car-following driving control model and obtain the car-following driving control scheme.

[0088] Embodiment 3:

[0089] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as that of Example 2. Furthermore, the equivalent wind resistance coefficient Cd of the following vehicle during following driving is eq As shown below:

[0090] Cd eq =2F follow / ρ×S×v 2 (1)

[0091] In the formula, F follow is the following wind resistance of the following vehicle at the same speed as the single vehicle, ρ is the air density, S is the frontal area, and v is the driving speed.

[0092] Embodiment 4:

[0093] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as any one of Embodiments 2-3. Furthermore, the equivalent drag coefficient estimation model under the influence of the wake is as follows:

[0094]

[0095] Where α, β, and γ are influencing factors; d is the distance between vehicle queues.

[0096] Embodiment 5:

[0097] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as any one of Embodiments 2-4. Furthermore, the influencing factors α, β, and γ are respectively as follows:

[0098]

[0099] Where v is the driving speed.

[0100] Embodiment 6:

[0101] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as any one of Embodiments 2-5. Furthermore, the objective function of the following driving control model is as follows:

[0102]

[0103] In the formula, J opt (k) is the total cost function; λ, δ, μ is the weighting factor; J L (k) is the security cost function, J Cd (k) is the low-resistance cost function, J v (k) is the follow-up cost function, J a (k) is the comfort cost function; N p is the prediction time range; k is the time step.

[0104] Embodiment 7:

[0105] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as any one of embodiments 2-6. Furthermore, the safety cost function J L (k) is as follows:

[0106] L min (k) = 0.9 + 0.15v (k) + 0.0325v 2 (k) (7)

[0107] L max (k) = 7 + v (k) + 0.05v 2 (k) (8)

[0108] J L (k)=||L max (k)-L(k)|-|L min (k)-L(k)|| (9)

[0109] Where, L max , L min are the upper and lower limits of the distance between the front and rear vehicles; v(k) is the driving speed corresponding to time step k; L(k) is the distance between the front and rear vehicles corresponding to time step k.

[0110] Embodiment 8:

[0111] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake has the same technical content as any one of the embodiments 2-7. Further, the following cost function J v (k) is as follows:

[0112] J V (k)=|Vf (k)-V e (k)|(10)

[0113] Where V f (k), V e (k) represents the speed of the front and rear vehicles respectively.

[0114] Embodiment 9:

[0115] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake, the technical content is the same as any one of the embodiments 2-8, further, the comfort cost function J a (k) is as follows:

[0116] J a (k)=|a(k-1)-a(k)|(11)

[0117] Where a(k-1) and a(k) are the accelerations corresponding to time step k-1 and time step k.

[0118] Embodiment 10:

[0119] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake, the technical content is the same as any one of the embodiments 2-9, further, the low resistance cost function J Cd (k) is as follows:

[0120] J Cd (k) = Cd eq (k) / Cd(12)

[0121] In the formula, Cd eq (k) is the equivalent drag coefficient of the rear vehicle; Cd is the physical drag coefficient of a single vehicle.

[0122] Embodiment 11:

[0123] The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake, the technical content is the same as any one of the embodiments 2-10, further, the equivalent wind resistance coefficient Cd of the following vehicle eq The upper limit of (k) is a preset fixed value, and the lower limit is the change in slack as the speed of the preceding vehicle changes.

[0124] Equivalent drag coefficient of the rear vehicle Cd eq (k) The functional relationship between speed v and vehicle spacing d is as follows:

[0125]

[0126] Embodiment 12:

[0127] A vehicle platoon drag reduction and following driving control method considering the influence of external flow field wake, comprising the following steps:

[0128] 1. Obtain the equivalent drag coefficient of the vehicle behind during car-following driving

[0129] The physical drag coefficient Cd of a single vehicle at different speeds was obtained experimentally. The average drag coefficient under different speed conditions was finally measured to be 0.27201. This is used as the actual physical drag coefficient of the vehicle to compare the wind resistance change law of the following car under the condition of the front car being affected by the wake. In order to conveniently describe the drag reduction effect of the following car affected by the wake of the front car, the equivalent drag coefficient Cd is introduced. eq The equivalent drag coefficient can be considered as the drag coefficient calculated using the air resistance affected by the wake of the vehicle in front.

[0130] 2. Construct an estimation model for the equivalent drag coefficient under the influence of wake turbulence

[0131] The distance between the queues and the driving speed affect each other and couple with each other to jointly determine the aerodynamic resistance of the following vehicle. It can be observed that the equivalent drag coefficient of the following vehicle has a certain linear relationship with the speed and the distance between the vehicle queues, and the change trend of its equivalent drag coefficient is closest to the Langmuir function through experiments. Therefore, the Langmuir function is used to fit and analyze it. Therefore, the formula about d can be used to express Cd eq .

[0132] 3. Determine the influencing factors of the equivalent drag coefficient estimation model under the influence of wake turbulence

[0133] Since the obtained Cd eq While introducing the function expression, three influencing factors α, β, and γ are introduced to describe the equivalent drag coefficient of the following vehicle. Based on the previous experimental data, α, β, and γ used to describe the equivalent drag coefficient of the experimental object can be derived.

[0134] Since the curvature of each equivalent drag coefficient curve is different, the fitted influencing factors α, β, and γ are also different. However, through the arrangement of the data, it is found that the three influencing factors all show a certain linear relationship with the speed v. When ensuring Cd eq Calculation formula linear correlation coefficient R 2 When it is always greater than 0.99, the three influencing factors α, β, and γ are linearly related to the speed v.

[0135] 4. Design a car-following driving control method based on the equivalent drag coefficient estimation model

[0136] Define the total cost function J opt (k) Used to evaluate the safety, low resistance, followability, and comfort during car-following driving.

[0137] 5. Safety restrictions for car-following driving

[0138] During the following driving process, a smaller distance can bring a better equivalent drag coefficient, but at the same time there will be a high risk of rear-end collision. Although a larger queue distance can fully ensure driving safety, too large a queue distance will make the equivalent drag coefficient tend to the actual drag coefficient of a single vehicle. Therefore, based on the above following driving requirements, L min and L max is a function related to vehicle speed v, J L (k) is defined to describe the safety during car-following driving.

[0139] 6. Followability constraints for car-following driving

[0140] Regarding the followability constraint of car-following driving, it can be seen from the equivalent drag coefficient diagram in the previous article that the equivalent drag coefficient is very sensitive to the speed of car-following driving and the distance between vehicles in the queue. Therefore, the stronger the followability of the speed of the following vehicle to the speed of the leading vehicle, the smaller the distance error change at different speeds with the same equivalent drag coefficient, the smaller the fluctuation of the equivalent drag coefficient, and the better the low resistance. Therefore, J is defined v (k) Describe the low resistance during car-following driving.

[0141] 7. Comfort constraints for car-following driving

[0142] The comfort of car-following driving can be characterized by the acceleration during car-following driving. The more gradual the acceleration change, the higher the riding comfort. However, due to the consideration of strong following performance, low drag coefficient and safety, the acceleration may fluctuate within a certain range. Therefore, J is defined as a (k) Describes the degree of change in acceleration from the previous moment to the next moment during car-following driving.

[0143] 8. Low resistance constraints for car-following driving

[0144] Regarding the constraint of the equivalent wind resistance of car-following driving, it can be seen from the previous article that in the car-following driving process, the equivalent wind resistance coefficient of the rear vehicle is jointly determined by the distance between the vehicle queues and the driving speed of the front vehicle. In the real car-following driving environment, the driving speeds of the front and rear vehicles are difficult to keep consistent and the driving distance varies with the speed difference. In order to ensure the feasibility of the calculation, the entire car-following driving process is discretized. At each moment, the vehicle queue can be considered to be relatively static. The distance between the vehicle queues and the driving speed of the front vehicle at each moment are used to calculate the equivalent wind resistance coefficient of the rear vehicle. Therefore, JC is defined d (k) Describe the low resistance during car-following driving.

[0145] 9. Optimization of low-resistance soft constraints for car-following control

[0146] In the present invention, the vehicle spacing d is controlled by two different hard constraints that jointly affect the spacing d. The safety constraint and the low resistance constraint are both controlled by changing the vehicle queue spacing d. As a result, in the multi-objective optimization process of car-following driving, the vehicle queue spacing d is affected by two different constraints, which may cause the queue driving spacing to fail to meet one of the constraints when the total optimization cost is minimized. Therefore, the prediction optimization problem of the car-following control strategy based on the tail flow of the leading vehicle is computationally infeasible.

[0147] The present invention converts the equivalent drag coefficient Cd eq The lower limit is a hard constraint consisting of a fixed value, which is replaced by a variable that can change the relaxation degree with the speed of the front vehicle, so that its equivalent drag coefficient Cd eq The lower limit range is transformed from a hard constraint to a soft constraint, while the upper limit does not have this problem and the hard constraint design is retained. According to the speed corresponding to the sudden change of the safety distance-speed ratio, it can be known that the speed requirement for the safety distance at that time suddenly changes. These data feature points are matched to different lower limits of the equivalent drag coefficient to achieve soft constraints on the equivalent drag coefficient.

[0148] The present invention aims to improve the low-drag following driving problem of intelligent networked vehicles during following driving. Based on the study of the variation law of the equivalent drag coefficient of the following vehicle under the influence of the wake of the preceding vehicle, an equivalent drag coefficient estimation model is established, and an aerodynamic equivalent drag coefficient estimation method of the following vehicle under the influence of the wake of the preceding vehicle is proposed. Based on the estimation model, a drag reduction following control strategy based on the influence of the wake of the preceding vehicle is built, and the effectiveness of the following control strategy designed by integrating the equivalent drag coefficient estimation model is explained. The results show that the established drag reduction following strategy can not only ensure the safety of following but also has good performance in reducing the aerodynamic resistance of the following vehicle, providing theoretical and technical support for realizing the energy-saving driving technology of intelligent networked vehicles.

[0149] Embodiment 13:

[0150] A vehicle platoon drag reduction and following driving control method considering the influence of external flow field wake, comprising the following steps:

[0151] 1. Obtain the equivalent drag coefficient of the vehicle behind during car-following driving

[0152] The specific method is:

[0153] 1) In order to facilitate the calculation of the wind resistance reduction effect of the following car under the wake of the preceding car, the physical drag coefficient Cd of a single vehicle at different speeds is experimentally obtained. It can be concluded that the physical drag coefficient of the vehicle fluctuates slightly under different speed conditions, but the variation is less than 2%, which is consistent with the fact that the actual drag coefficient is only related to the aerodynamic layout and has nothing to do with speed, and also indirectly verifies that the model has achieved the required accuracy. Finally, the average drag coefficient under different speed conditions was measured to be 0.27201. The present invention will use this as the actual physical drag coefficient of the vehicle to compare the change law of the wind resistance of the following car under the wake of the preceding car.

[0154] 2) In order to conveniently describe the drag reduction effect of the following car affected by the wake of the preceding car, the equivalent drag coefficient Cd is introduced eq The equivalent drag coefficient can be considered as the drag coefficient calculated using the air resistance affected by the wake of the preceding vehicle, and its equivalent drag coefficient can be explained using the traditional equivalent drag coefficient calculation formula.

[0155] Cd=2F / ρ×S×v 2

[0156] Cd eq =2F follow / ρ×S×v 2

[0157] Where F is the wind resistance of the bicycle. follow is the following wind resistance of the following vehicle at the same speed as the single vehicle, ρ is the air density, S is the frontal area, and v 2 is the driving speed.

[0158] 2. Construct an estimation model for the equivalent drag coefficient under the influence of wake turbulence

[0159] The specific method is:

[0160] 1) Through the analysis of the simulation results in the previous section, it can be concluded that the equivalent drag coefficient of the following car is jointly determined by the distance between the vehicles in the convoy and the driving speed. The distance between the convoy and the driving speed affect each other and couple with each other to jointly determine the aerodynamic resistance of the following car. It can be observed that the equivalent drag coefficient of the following car has a certain linear relationship with the speed and the distance between the vehicles in the convoy, and the change trend of the equivalent drag coefficient is closest to the Langmuir function through experiments, so the Langmuir function is used to fit and analyze it.

[0161] 2) The Langmuir function has a good fitting effect on the equivalent drag coefficient, and the correlation coefficient of each curve is R 2 Both are greater than or equal to 0.99, showing a strong correlation, so Cd can be expressed using the formula about d eq .

[0162]

[0163] 3) Due to the obtained Cd eq While formulating the function expression, three influencing factors α, β, and y are introduced to describe the equivalent drag coefficient of the following vehicle. Based on the experimental data in the early stage of this study, α, β, and γ used to describe the equivalent drag coefficient of the experimental object can be derived.

[0164] Since the curvature of each equivalent drag coefficient curve is different, the fitted influencing factors α, β, and γ are also different. However, through the arrangement of the data, it is found that the three influencing factors have a certain linear relationship with the speed v. The three influencing factors of each curve are optimally addressed to ensure that Cd eq Calculation formula linear correlation coefficient R 2 When it is always greater than 0.99, the linear correlation points of the three influencing factors α, β, and γ with respect to the speed v are obtained, as shown in Table 1.

[0165] Table 1 The linear correlation points between the influencing factors α, β, γ and speed under various speed conditions

[0166]

[0167] 4) Finally, the equivalent drag coefficient calculation formula and the three influencing factor calculation formulas can be used to associate the equivalent drag coefficient of the following vehicle in this experiment with the distance between the vehicle queues and the driving speed, and finally a function group of the equivalent drag coefficient that can be described by mathematical methods can be obtained:

[0168]

[0169]

[0170]

[0171]

[0172] 3. Design a car-following driving control method based on the equivalent drag coefficient estimation model

[0173] The specific method is:

[0174] 1) Multi-objective optimization problem of car-following driving

[0175] Define the total cost function J opt (k) is used to evaluate the safety, low resistance, followability, and comfort during car-following driving. Its cost function for the time step k is:

[0176]

[0177] Among them JL (k) is the security cost function, J cd (k) is the low-resistance cost function, J v (k) is the follow-up cost function, J a (k) is the comfort cost function, λ, δ, μ is the weighting factor for adjusting safety, low resistance, speed following and comfort in the total cost function. The total cost of optimization within each prediction range, the optimization problem at time step k is expressed as:

[0178]

[0179] Among them, N p is the forecast range.

[0180] 2) Safety constraints for car-following driving

[0181] The allowable distance between the front and rear vehicles should be limited to a certain range [L min , L max ], because during the following driving process, a smaller distance can bring a better equivalent drag coefficient, but at the same time there will be a high risk of rear-end collision. Although a larger queue distance can fully ensure driving safety, too large a queue distance will make the equivalent drag coefficient tend to the actual drag coefficient when a single vehicle is driving. Therefore, based on the above following driving requirements, L min and L max is defined as a function of vehicle speed v:

[0182] L min (k) = 0.9 + 0.15v (k) + 0.0325v 2 (k)

[0183] L max (k) = 7 + v (k) + 0.05v 2 (k)

[0184] J L (k)=||L max (k)-L(k)|-|L min (k)-L(k)||

[0185] In this paper, driving safety is achieved by ensuring that the following vehicle drives within the safe driving constraints. Note that the future speed and position of the leading vehicle are estimated based on its current and last speeds, that is, for simplicity, it is assumed that the leading vehicle has a constant acceleration within each prediction range. In order to obtain a more accurate future speed, a data-driven approach can be used. Therefore, J is defined L (k) Describe the safety of car-following driving.

[0186] 3) Followability constraints for car-following driving

[0187] Regarding the followability constraint of car-following driving, it can be seen from the equivalent drag coefficient diagram in the previous article that the equivalent drag coefficient is very sensitive to the speed of car-following driving and the distance between vehicles in the queue. Therefore, the stronger the followability of the speed of the following vehicle to the speed of the leading vehicle, the smaller the distance error change at different speeds with the same equivalent drag coefficient, the smaller the fluctuation of the equivalent drag coefficient, and the better the low resistance. Therefore, J is defined V (k) Describe the low resistance during car-following driving.

[0188] J V (k)=|V f (k)-V e (k)|

[0189] 4) Comfort constraints for car-following driving

[0190] The comfort of car-following driving can be characterized by the acceleration during car-following driving. The more gradual the acceleration change, the higher the riding comfort. However, due to the consideration of strong following performance, low drag coefficient and safety, the acceleration may fluctuate within a certain range. Therefore, J is defined as a (k) Describes the degree of change in acceleration from the previous moment to the next moment during car-following driving.

[0191] J a (k)=|a(k-1)-a(k)|

[0192] 4) Low resistance constraints for car-following driving

[0193] Regarding the constraint of the equivalent wind resistance of car-following driving, it can be seen from the previous article that in the car-following driving process, the equivalent wind resistance coefficient of the rear vehicle is jointly determined by the distance between the vehicle queues and the driving speed of the front vehicle. In the real car-following driving environment, the driving speeds of the front and rear vehicles are difficult to keep consistent and the driving distance varies with the speed difference. In order to ensure the feasibility of the calculation, the entire car-following driving process is discretized. At each moment, the vehicle queue can be considered to be relatively static. The distance between the vehicle queues and the driving speed of the front vehicle at each moment are used to calculate the equivalent wind resistance coefficient of the rear vehicle. Therefore, J is defined Cd (k) Describe the low resistance during car-following driving.

[0194] J Cd (k) = Cd eq (k) / Cd

[0195] In order to ensure the low resistance of the following control strategy, Cd eqLimited to [0.235, 0.25]. It is worth noting that in the process of following driving, v in the equivalent wind resistance calculation formula can be regarded as a known quantity in disguise, because v in the equivalent wind resistance calculation formula is the speed of the leading vehicle and has nothing to do with the speed of the following vehicle. The wake generated by the leading vehicle during driving acts on the front of the following vehicle. The following vehicle can only change the influence of the wake of the leading vehicle on its own vehicle by changing the distance d between the vehicle queues, thereby changing the equivalent wind resistance coefficient of the following vehicle.

[0196] 4. Optimization of low-resistance soft constraints for car-following control

[0197] The specific method is:

[0198] In the present invention, the vehicle spacing d has two different hard constraints that jointly affect the spacing d. The safety constraint and the low resistance constraint are both controlled by changing the vehicle queue spacing d. Safety is achieved by directly limiting the queue spacing d in [L min , L max ] to ensure the safety of its following driving; and the low resistance is achieved by giving Cd eq The restricted range and the current speed of the preceding vehicle can limit the queue spacing d in disguise, resulting in the interaction of two different constraints in the multi-objective optimization process of following driving, which may cause the queue driving spacing to fail to meet one of the constraints when the total optimization cost is minimized. Therefore, the predictive optimization problem of the following control strategy based on the wake of the preceding vehicle is computationally infeasible.

[0199] In order to improve this situation, the equivalent drag coefficient Cd eq Introduce soft constraints to distinguish the equivalent drag coefficient Cd proposed in the previous section eq The hard constraint is [0.235, 0.25]. To optimize the problem of vehicle queue spacing overflow constraint in the following scenario. The low-resistance vehicle queue spacing under the hard constraint condition can be expressed as:

[0200]

[0201] Although the lower limit of the equivalent drag coefficient has good low resistance, it may be excessively pursued during the driving of the leading vehicle, resulting in the output of the following control strategy not meeting the lower limit of safety requirements. eq The lower limit is a hard constraint consisting of a fixed value, which is replaced by a variable that can change the relaxation degree with the speed of the front vehicle, so that its equivalent drag coefficient Cd eqThe lower limit range is transformed from a hard constraint to a soft constraint, while the upper limit does not have this problem and the hard constraint design is retained. According to the speed corresponding to the sudden change in the safety distance-speed ratio, it can be known that the speed requirement for the safety distance at that time suddenly changed. These data feature points correspond to different lower limits of the equivalent drag coefficient to achieve soft constraints on the equivalent drag coefficient. Under the soft constraint, the distance between low-resistance vehicles in the queue can be expressed as:

[0202]

[0203] This achieves an equivalent drag coefficient Cd eq The transition from hard constraints to soft constraints is to improve the decision-making process in the following driving control strategy of excessively pursuing low resistance within a specific speed range and selectively giving up some safety, resulting in the phenomenon of vehicle driving distance overflowing constraints.

[0204] When conducting simulation experiments on the vehicle queue drag reduction and following driving control technology considering the influence of the external flow field wake, the NEDC standard working condition is used as a case analysis. Since the experimental results of the present invention focus on the comparison of the degree of decrease in its equivalent drag coefficient. Since the cycle working condition is directly applied to the cycle working condition of the front vehicle, and one of the factors affecting the equivalent drag coefficient of the rear vehicle is the speed of the front vehicle, after the front vehicle stops, there will be no wake influence of the front vehicle's wake on the rear vehicle, causing the rear vehicle's equivalent drag coefficient to collapse to the physical drag coefficient of the vehicle, affecting the comparison of its drag reduction characteristics. Therefore, the NEDC and WLTC working conditions are partially corrected, and all zero speed points within the cycle are deleted to more intuitively compare the changes in the equivalent drag coefficient of the rear vehicle under the influence of the front vehicle's wake.

[0205] Depend on Figure 4 It can be seen that when the speed of the front vehicle is greater than a certain range, the hard constraint will sacrifice some safety and pursue better low resistance. Under the soft constraint condition, the safety and low resistance constraints will be re-decided to weigh the weight between low resistance and safety. Its equivalent drag coefficient can change with the speed in the constraint to improve the weight coefficient of safety and low resistance in the total cost function, which is directly reflected in the fact that the equivalent drag coefficient of the soft constraint control strategy is better than the equivalent drag coefficient of the hard constraint at some speeds. It can make up for the low resistance performance sacrificed by lengthening the queue spacing to ensure safety when the speed is greater than a certain range, so that in the entire working condition, the soft constraint equivalent drag coefficient decreases by 10.68% compared with the hard constraint equivalent drag coefficient decrease of 10.77%, and the performance is almost the same. It can be considered that the soft constraint has almost no difference in drag reduction performance with the hard constraint under the premise of ensuring safety. It can effectively reduce the driving resistance of the vehicle during driving and improve fuel economy, providing a new idea for reducing wind resistance and optimizing fuel economy for vehicle driving.

Claims

1. A drag-reducing and following driving control method for vehicle platoons considering the influence of external flow field wake. It is characterized in that The following steps are involved: 1) Obtain the equivalent drag coefficient Cd of the following vehicle during car-following driving eq . 2) Construct an estimation model of equivalent drag coefficient under the influence of wake; 3) Based on the equivalent drag coefficient estimation model under the influence of wake turbulence, a car-following driving control model is constructed; 4) Solve the car-following driving control model and obtain the car-following driving control scheme.

2. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 1, It is characterized in that Equivalent drag coefficient Cd of the rear vehicle during following driving eq As shown below: Contd eq =2F follow / ρ×S×v 2 (1) In the formula, F follow is the following wind resistance of the following vehicle at the same speed as the single vehicle, ρ is the air density, S is the frontal area, and v is the driving speed.

3. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 1, It is characterized in that The equivalent drag coefficient estimation model under the influence of the wake is as follows: Where α, β, and γ are influencing factors; d is the distance between vehicle queues.

4. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 1, It is characterized in that The impact factors α, β, and γ are as follows: In the formula, v is the driving speed.

5. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 1, It is characterized in that The objective function of the car-following driving control model is as follows: In the formula, J opt (k) is the total cost function; λ, δ, μ is the weighting factor; J L (k) is the security cost function, J Cd (k) is the low-resistance cost function, J v (k) is the follow-up cost function, J a (k) is the comfort cost function; N p is the prediction time range; k is the time step.

6. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 5, It is characterized in that Security cost function J L (k) is as follows: L min (k)=0.9+0.15v(k)+0.0325v 2 (k) (7) L max (k)=7+v(k)+0.05v 2 (k) (8) I L (k)=||L max (k)-L(k)|-|L min (k)-L(k)|| (9) Where, L max , L min are the upper and lower limits of the distance between the front and rear vehicles; v(k) is the driving speed corresponding to time step k; L(k) is the distance between the front and rear vehicles corresponding to time step k.

7. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 5, It is characterized in that Follow-up cost function J v (k) is as follows: J V (k)=|V f (k)-V e (k)| (10) Where V f (k), V e (k) represents the speed of the front and rear vehicles respectively.

8. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 5, It is characterized in that Comfort cost function J a (k) is as follows: J a (k)=|a(k-1)-a(k)| (11) Where a(k-1) and a(k) are the accelerations corresponding to time step k-1 and time step k.

9. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 5, It is characterized in that Low resistance cost function J Cd (k) is as follows: J Cd (k)=Cont eq (k) / Cont (12) In the formula, Cd eq (k) is the equivalent drag coefficient of the rear vehicle; Cd is the physical drag coefficient of a single vehicle.

10. The vehicle platoon drag reduction and following driving control method considering the influence of the external flow field wake according to claim 9, It is characterized in that Equivalent drag coefficient of the rear vehicle Cd eq The upper limit of (k) is a preset fixed value, and the lower limit is the change in relaxation as the speed of the preceding vehicle changes; Equivalent drag coefficient of the rear vehicle Cd eq (k) The functional relationship between speed v and vehicle spacing d is as follows:

Images