A heterogeneous vehicle platoon cooperative control method based on a hierarchical control architecture

Abstract

The present application relates to the field of vehicle platoon cooperative control, and particularly relates to a heterogeneous vehicle platoon cooperative control method based on a hierarchical control architecture. The heterogeneous vehicle platoon includes a pure electric vehicle, a hybrid vehicle and a fuel cell vehicle. The cooperative control method includes establishing a heterogeneous vehicle platoon model, a vehicle platoon cooperative control based on a top-level centralized controller and a vehicle platoon energy-saving control based on a bottom-level distributed controller. The present application realizes hierarchical decoupling of platoon control and energy management by focusing on vehicle platoon longitudinal following error at the top level and focusing on energy management optimization at the bottom level, improves system real-time performance and scalability. Independent design of the top level and the bottom level makes the algorithm structure clearer, easy to implement and expand in an actual vehicle control system, and has high real-time performance and applicability in different scales and complexities of vehicle platoon environment, is compatible with pure electric vehicles, hybrid vehicles and fuel cell vehicles, and realizes optimal energy-saving control.

Description

Technical Field

[0001] This invention relates to the field of vehicle queuing cooperative control, and in particular to a heterogeneous vehicle queuing cooperative control method based on a hierarchical control architecture. Background Technology

[0002] The rapid development of new energy vehicle technology has driven the widespread application of intelligent connected transportation systems, among which pure electric vehicles, hybrid electric vehicles, and fuel cell vehicles have become important components of intelligent transportation systems. However, in a heterogeneous vehicle fleet composed of these vehicles, the differences in power response, energy compensation, range, and control strategies among different power systems make achieving stable and coordinated control and optimal energy allocation a key technical challenge.

[0003] Currently, vehicle platoon control methods mainly include distance control, adaptive cruise control, and cooperative adaptive cruise control. Among these, cooperative adaptive cruise control incorporates vehicle-to-everything (V2X) technology, enabling information sharing and collaborative optimization. However, existing methods are mostly designed for homogeneous vehicle platoons, assuming all vehicles have the same powertrain. In practical applications, however, different types of new energy vehicles often travel in mixed platoons. Existing control strategies struggle to simultaneously consider the dynamic response characteristics, energy management requirements, and collaborative control methods of heterogeneous powertrains, leading to mismatched responses and significant following errors among vehicles within the platoon, thus affecting platoon stability. Furthermore, traditional energy management strategies primarily optimize for individual vehicles, lacking a global energy-optimal control mechanism, resulting in excessively high overall platoon energy consumption. Considering these factors, achieving globally optimal energy management while ensuring stable following of vehicles with different powertrains remains a critical technical challenge in the field of intelligent connected new energy vehicles. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a heterogeneous vehicle platoon cooperative control method based on a hierarchical control architecture. The heterogeneous vehicle platoon includes at least two of the following: pure electric vehicles, hybrid electric vehicles, and fuel cell vehicles. The cooperative control method includes the following steps:

[0005] Step 1: Establish a heterogeneous vehicle queue model:

[0006] The heterogeneous vehicle queue model includes a vehicle queue dynamics model, a pure electric vehicle power system model, a hybrid electric vehicle power system model, and a fuel cell vehicle power system model.

[0007] The method for establishing the vehicle platoon dynamics model is as follows:

[0008] For the i-th vehicle, its longitudinal dynamic equation is:

[0009]

[0010] Where, m i Let v be the mass of the i-th vehicle. i Let F be the longitudinal velocity of the i-th vehicle. T,i F is the driving force of the i-th vehicle. R,i Let be the total driving resistance of the i-th vehicle, which includes rolling resistance, air resistance, and gradient resistance.

[0011]

[0012] Among them, C r C is the rolling resistance coefficient. d ρ is the air resistance coefficient, A is the front surface area of ​​the vehicle, ρ is the air density, g is the acceleration due to gravity, and θ is the slope angle.

[0013] The distance error and speed error between the i-th vehicle and the vehicle in front are defined as follows:

[0014]

[0015] Among them, e s,i e represents the distance error. v,i For speed error, s i Let h be the position of the i-th vehicle, and h be the headway.

[0016] The pure electric vehicle powertrain model is as follows:

[0017]

[0018] Among them, F T,BEV For the driving force of pure electric vehicles, T mot η is the motor torque. gear For the efficiency of the transmission system, r w Let P be the radius of the wheel. mot ω is the output power of the motor. mot P is the motor speed. bat η is the net discharge power of the battery. mot For motor efficiency.

[0019] The driving power P of a pure electric vehicle drive,BEV for:

[0020] P drive,BEV =P mot

[0021] The hybrid electric vehicle powertrain model is as follows:

[0022]

[0023] Among them, F T,HEVFor the driving force of hybrid vehicles, P eng ω is the power output of the engine through the generator. eng T is the engine speed. eng For engine torque, η gen For generator efficiency, P bat This represents the net discharge power of the battery.

[0024] The driving power of a hybrid electric vehicle is provided by an electric motor; the driving power P of a hybrid electric vehicle is... drive,HEV for:

[0025] P drive,HEV =P mot

[0026] The aforementioned fuel cell vehicle powertrain model is as follows:

[0027]

[0028] Among them, F T,FCEV For the driving force of fuel cell vehicles, P FC For fuel cell output power, LHV is the hydrogen consumption rate. H Because hydrogen has a low calorific value, η FC For fuel cell efficiency.

[0029] The drive power P of fuel cell vehicles drive,FCEV for:

[0030] P drive,FCEV =P mot

[0031] Step 2: Vehicle queuing cooperative control based on the top-level centralized controller:

[0032] The aforementioned top-level centralized controller focuses on reducing following errors between vehicles and ensuring the cooperative stability of the vehicle platoon. The state vector x is defined as:

[0033]

[0034] Where s is position, v is velocity, and a is acceleration.

[0035] Define the control input u as acceleration, that is:

[0036] u = a

[0037] Define the objective function J of the top-level control as:

[0038]

[0039] Among them, Q s As the weight of the distance error, Qv The speed error weight is determined based on the tendency of the distance error and speed error in the objective function. T is the control time domain.

[0040] The Hamiltonian function constructed based on Pontryagin's minimum principle is:

[0041] H(x,u,λ)=Q s (e s ) 2 +Q v (e v ) 2 +λ s v+λ v a+λ a u

[0042] In the formula, H(x,u,λ) is the Hamiltonian function, and λ=[λ s ,λ v ,λ a ] T Let λ be the costate vector, where λ is the costate vector. s , λ v , λ a These are the costate quantities of position, velocity, and acceleration in the Hamiltonian function, respectively.

[0043] The costate vector satisfies the costate update equation:

[0044]

[0045] Therefore, optimal control input for:

[0046]

[0047] Step 3: Energy-saving control of vehicle queuing based on the underlying distributed controller:

[0048] The goal of the underlying distributed controller is to minimize the overall energy consumption of the fleet while satisfying the power requirements and operational constraints of each vehicle model. For the i-th vehicle, a state vector X is defined. i for:

[0049] X i =SOC i

[0050] Among them, SOC i Let represent the state of charge of the battery of vehicle i.

[0051] For the i-th vehicle, define the control input U. BDC,i for:

[0052]

[0053] Among them, P bat,i Let P be the net battery discharge power of the i-th vehicle. FC,i Let P be the fuel cell output power of the i-th vehicle. eng,i Let be the power output of the engine of the i-th vehicle through the generator;

[0054] For the i-th vehicle, the objective function J of the underlying control is... ECMS,i for:

[0055]

[0056] Among them, w bat As a weight for battery energy consumption, w FC For the hydrogen consumption weight of fuel cell vehicles, w f The weights for fuel consumption of hybrid electric vehicles, battery energy consumption, hydrogen consumption of fuel cell vehicles, and fuel consumption of hybrid electric vehicles are determined based on their respective propensities in the objective function. P bat,i Let C be the net battery discharge power of the i-th vehicle. H,i For the hydrogen consumption cost of fuel cell vehicles, C f,i This refers to the fuel consumption cost of hybrid vehicles.

[0057] The local Hamiltonian function constructed based on Pontryagin's minimum principle is:

[0058] H E,i (X i U BDC,i ,μ i ) = w bat P bat,i +w FC C H,i +w f C f,i +μ i T g i (X i U BDC,i )

[0059] Among them, H E,i (X i U BDC,i ,μ i ) is a local Hamiltonian function, g i (X i U BDC,i Let μ be a function describing the dynamic state of the i-th vehicle.i Let be the local costate vector of the i-th vehicle. The local costate vector satisfies the costate update equation:

[0060]

[0061] The constraints for pure electric vehicles are:

[0062]

[0063] The constraints for fuel cell vehicles are as follows:

[0064]

[0065] The constraints for hybrid vehicles are:

[0066]

[0067] Among them, P drive,i P is the driving force of the i-th vehicle. mot,i For the output power of the motor of the i-th vehicle, SOC i (t) represents the battery state of charge of the i-th vehicle, P FC,i Let P be the output power of the fuel cell in the i-th vehicle. eng,i For the engine output power of the i-th vehicle, SOC min The lower limit of the battery's state of charge (SOC) max This represents the upper limit of the battery's state of charge. Let be the lower limit of the fuel cell output power of the i-th vehicle. Let be the upper limit of the fuel cell output power of the i-th vehicle. This represents the lower limit of the engine output power of the i-th vehicle. This represents the upper limit of the engine output power of the i-th vehicle;

[0068] Local optimal control input for:

[0069]

[0070] The beneficial effects of this invention are:

[0071] 1. This invention achieves layered decoupling between queue control and energy management by focusing on longitudinal following error in the vehicle queue at the top level and energy management optimization at the bottom level, thereby improving the system's real-time performance and scalability. The top level only considers vehicle following performance and does not consider energy-saving control, while the bottom level controller realizes personalized energy-saving control for vehicles with multiple energy sources (heterogeneous vehicles). The independent design of the top and bottom levels makes the algorithm structure clearer, easier to implement and extend in actual vehicle control systems, and has high real-time performance and applicability in fleet environments of different sizes and complexities.

[0072] 2. The top-level centralized controller constructed in this invention acquires and integrates dynamic information between vehicles in real time, and adopts advanced prediction and optimal control algorithms to achieve accurate modeling and optimized control of platoon following error. This enables high coordination and rapid response among vehicles, effectively breaking through the limitations of traditional single-vehicle control strategies in heterogeneous platoon collaborative control, and improving the overall stability and safety of the platoon.

[0073] 3. This invention is compatible with heterogeneous vehicle platoons, including pure electric vehicles, hybrid electric vehicles, and fuel cell vehicles, achieving optimal energy scheduling. The constructed underlying distributed controller enables each vehicle to independently solve local energy optimization problems and achieves global coordination through real-time information exchange, thereby significantly improving the system's flexibility, scalability, and energy efficiency under multiple scenarios and demand conditions. Attached Figure Description

[0074] Figure 1 This is a schematic diagram of a heterogeneous vehicle queue according to an embodiment of the present invention;

[0075] Figure 2 This is a schematic diagram of the framework of the heterogeneous vehicle queue cooperative control method based on a hierarchical control architecture according to the present invention. Detailed Implementation

[0076] This embodiment provides a heterogeneous vehicle platoon cooperative control method based on a hierarchical control architecture. The leading vehicle in the heterogeneous vehicle platoon is a pure electric vehicle, the first following vehicle is a hybrid electric vehicle, and the second following vehicle is a fuel cell vehicle. Figure 1 As shown;

[0077] First, establish a heterogeneous vehicle queue model:

[0078] The heterogeneous vehicle queue model includes a vehicle queue dynamics model, a pure electric vehicle power system model, a hybrid electric vehicle power system model, and a fuel cell vehicle power system model.

[0079] The method for establishing the vehicle platoon dynamics model is as follows:

[0080] For the i-th vehicle, its longitudinal dynamic equation is:

[0081]

[0082] Where, m i Let v be the mass (kg) of the i-th vehicle. i Let F be the longitudinal velocity (m / s) of the i-th vehicle. T,i Let F be the driving force (N) of the i-th vehicle. R,i Let N be the total running resistance of the i-th vehicle. The total running resistance includes rolling resistance, air resistance, and gradient resistance.

[0083]

[0084] Among them, C r C is the rolling resistance coefficient. d Where A is the air resistance coefficient and A is the front surface area of ​​the vehicle (m²). 2 ), where ρ is the air density (kg / m³) 3 ), g is the acceleration due to gravity (9.81 m / s²). 2 ), where θ is the ramp angle (rad).

[0085] To describe the following performance in a vehicle platoon, the distance error and speed error between the i-th vehicle and the vehicle in front are defined as follows:

[0086]

[0087] Among them, e s,i e represents the distance error (m). v,i Let s be the velocity error (m / s). i Let be the position (m) of the i-th vehicle, and h be the headway (s).

[0088] The pure electric vehicle powertrain model is as follows:

[0089]

[0090] Among them, F T,BEV For the driving force (N) of a pure electric vehicle, T mot η is the motor torque (Nm). gear For the efficiency of the transmission system, r w Let P be the radius of the wheel (m). mot ω is the motor output power (W). mot P is the motor speed (rad / s). bat η is the net discharge power of the battery (W). mot For motor efficiency.

[0091] Therefore, the driving power of a pure electric vehicle is:

[0092] P drive,BEV =P mot

[0093] Among them, P drive,BEV This represents the driving power (W) of a pure electric vehicle.

[0094] The hybrid electric vehicle powertrain model is as follows:

[0095]

[0096] Among them, FT,HEV For the driving force (N) of a hybrid electric vehicle, P eng ω represents the power output (W) of the engine through the generator. eng T is the engine speed (rad / s). eng For engine torque (Nm), η gen For generator efficiency, P bat This represents the net discharge power of the battery (W).

[0097] Hybrid electric vehicles are powered by an electric motor, that is:

[0098] P drive,HEV =P mot

[0099] Among them, P drive,HEV The driving power (W) of a hybrid electric vehicle.

[0100] The aforementioned fuel cell vehicle powertrain model is as follows:

[0101]

[0102] Among them, F T,FCEV For the driving force (N) of fuel cell vehicles, P FC The output power (W) of the fuel cell. LHV is the hydrogen consumption rate (kg / s). H For hydrogen, the lower calorific value (J / kg) is η. FC For fuel cell efficiency.

[0103] Therefore, the drive power of a fuel cell vehicle is:

[0104] P drive,FCEV =P mot

[0105] Among them, P drive,FCEV This represents the drive power (W) of a fuel cell vehicle.

[0106] like Figure 2 As shown, the collaborative control method includes a vehicle queue collaborative control method based on a top-level centralized controller and a vehicle queue energy-saving control method based on a bottom-level distributed controller.

[0107] In the vehicle platoon cooperative control method based on a top-level centralized controller, the top-level centralized controller focuses on reducing the following error between vehicles and ensuring the cooperative stability of the vehicle platoon. The state vector x is defined as:

[0108]

[0109] Where s is position (m), v is velocity (m / s), and a is acceleration (m / s²). 2 ).

[0110] Define the control input u as acceleration, that is:

[0111] u = a

[0112] Define the objective function J of the top-level control as:

[0113]

[0114] Among them, Q s The distance error weight (N) 2 / m 2 ), Q v The speed error weight is ((m / s)2), and the distance error weight and speed error weight are determined based on the tendency of the distance error and speed error in the objective function. T is the control time domain (s).

[0115] The Hamiltonian function constructed based on Pontryagin's minimum principle is:

[0116] H(x,u,λ)=Q s (e s ) 2 +Q v (e v ) 2 +λ s v+λ v a+λ a u

[0117] In the formula, H(x,u,λ) is the Hamiltonian function, and λ=[λ s ,λ v ,λ a ] T Let λ be the costate vector, where λ is the costate vector. s , λ v , λ a These are the costate variables for position, velocity, and acceleration, respectively.

[0118] Define constraints: The costate vector satisfies the costate update equation:

[0119]

[0120] Therefore, optimal control input for:

[0121]

[0122] In the aforementioned vehicle platoon energy-saving control method based on a low-level distributed controller, the goal of the low-level distributed controller is to minimize the overall energy consumption of the platoon while simultaneously satisfying the power requirements and operational constraints of each vehicle type. For the i-th vehicle, a state vector X is defined. i for:

[0123] X i =SOC i

[0124] Among them, SOC i Let represent the state of charge of the battery of vehicle i.

[0125] For the i-th vehicle, define the control input U. BDC,i for:

[0126]

[0127] Among them, P bat,i Let P be the net battery discharge power (W) of the i-th vehicle. FC,i Let P be the fuel cell output power (W) of the i-th vehicle. eng,i Let be the power output (W) of the engine of the i-th vehicle through the generator;

[0128] For the i-th vehicle, the objective function J of the underlying control is... ECMS,i for:

[0129]

[0130] Among them, w bat Battery energy consumption weight (W) -1 ), w FC Hydrogen consumption weight (kg) for fuel cell vehicles -1 / s -1 ), w f Fuel consumption weighting for hybrid vehicles (kg) -1 / s -1 The weights for battery energy consumption, hydrogen consumption of fuel cell vehicles, and fuel consumption of hybrid vehicles are determined based on their respective propensities in the objective function. bat,i Let C be the net battery discharge power (W) of the i-th vehicle. H,i The cost of hydrogen consumption (J / s) for fuel cell vehicles, C f,i Fuel consumption cost (J / s) for hybrid vehicles.

[0131] The local Hamiltonian function constructed based on Pontryagin's minimum principle is:

[0132] HE,i (X i U BDC,i ,μ i ) = w bat P bat,i +w FC C H,i +w f C f,i +μ i T g i (X i U BDC,i )

[0133] Among them, H E,i (X i U BDC,i ,μ i ) is a local Hamiltonian function, g i (X i U BDC,i Let μ be a function describing the dynamic state of the i-th vehicle. i Let be the local costate vector of the i-th vehicle. The local costate vector satisfies the costate update equation:

[0134]

[0135] The constraints for pure electric vehicles are:

[0136]

[0137] The constraints for fuel cell vehicles are as follows:

[0138]

[0139] The constraints for hybrid vehicles are:

[0140]

[0141] Among them, P drive,i Let P be the driving force (N) of the i-th vehicle. mot,i Let the output power (W) of the motor of the i-th vehicle be , and the SOC be . i (t) represents the battery state of charge of the i-th vehicle, P FC,i Let P be the fuel cell output power (W) of the i-th vehicle. eng,i Let SOC be the engine output power (W) of the i-th vehicle. min The lower limit of the battery's state of charge (SOC) max This represents the upper limit of the battery's state of charge. Let be the lower limit of the fuel cell output power of the i-th vehicle. Let be the upper limit of the fuel cell output power of the i-th vehicle. This represents the lower limit of the engine output power of the i-th vehicle. This represents the upper limit of the engine output power of the i-th vehicle.

[0142] Therefore, the local optimal control input for:

[0143]

Claims

1. A heterogeneous vehicle platoon cooperative control method based on a hierarchical control architecture, wherein the heterogeneous vehicle platoon includes at least two of the following: pure electric vehicles, hybrid electric vehicles, and fuel cell vehicles; characterized in that: The aforementioned collaborative control method includes the following steps: Step 1: Establish a heterogeneous vehicle queue model: The heterogeneous vehicle queuing model includes a vehicle queuing dynamics model, a pure electric vehicle powertrain model, a hybrid electric vehicle powertrain model, and a fuel cell vehicle powertrain model. The method for establishing the vehicle platoon dynamics model is as follows: For the The longitudinal dynamic equation of the vehicle is: in, For the first The quality of the vehicle For the first The longitudinal speed of the vehicle, For the first The driving force of the car For the first The total resistance of a vehicle includes rolling resistance, air resistance, and gradient resistance. in, The rolling resistance coefficient, The air drag coefficient, This refers to the front surface area of ​​the vehicle. air density, It is the acceleration due to gravity. The slope angle; No. The distance error and speed error between a vehicle and the vehicle in front are defined as follows: in, For vehicle distance error, For speed error, For the first The location of the car This refers to the headway of the train. The pure electric vehicle powertrain model is as follows: Among them, F T,BEV For the driving force of pure electric vehicles, This is the motor torque. For the efficiency of the transmission system, For the wheel radius, This refers to the motor's output power. This refers to the motor speed. η is the net discharge power of the battery. mot For motor efficiency; Drive power of pure electric vehicles for: ； The hybrid electric vehicle powertrain model is as follows: Among them, F T,HEV For the driving force of hybrid vehicles, This refers to the power output of the engine through the generator. Engine speed, For engine torque, For generator efficiency, This refers to the net discharge power of the battery. Drive power of hybrid vehicles for: The fuel cell vehicle powertrain model is as follows: Among them, F T,FCEV For the driving force of fuel cell vehicles, For fuel cell output power, Hydrogen consumption rate, Hydrogen has a low calorific value. For fuel cell efficiency; Drive power of fuel cell vehicles for: ； Step 2: Vehicle queuing cooperative control based on the top-level centralized controller: The top-level centralized controller defines the state vector x as follows: in, For location, For speed, For acceleration; Define the control input u as acceleration, that is: Define the objective function J of the top-level control as: in, As the weight of the distance error, As the speed error weight, To control the time domain, For vehicle distance error, For speed error; The Hamiltonian function constructed based on Pontryagin's minimum principle is: In the formula, H(x,u,λ) is the Hamiltonian function. Let λ be the costate vector, where λ is the costate vector. s , λ v , λ a These are the costate variables for position, velocity, and acceleration, respectively. The costate vector satisfies the costate update equation: Optimal control input for: Step 3: Energy-saving control of vehicle queuing based on the underlying distributed controller: The underlying distributed controller is for the first Vehicle, define state vector X i for: in, For the first The battery state of charge of the vehicle; Regarding the first Vehicle, define control input U BDC,i for: Among them, P bat,i For the first The net discharge power of the vehicle's battery, P FC,i For the first The fuel cell output power of the vehicle, P eng,i For the first The power output of a vehicle's engine through a generator; Regarding the first Vehicle, the objective function J of the underlying control ECMS,i for: in, Weighted by battery energy consumption. The hydrogen consumption weight for fuel cell vehicles For the fuel consumption weighting of hybrid vehicles, For the first The vehicle's net battery discharge power, The cost of hydrogen consumption for fuel cell vehicles, Fuel consumption costs for hybrid vehicles; The local Hamiltonian function constructed based on Pontryagin's minimum principle is: Among them, H E,i (X i U BDC,i , μ i ) is a local Hamiltonian function. To describe the first A function for the dynamic state of the vehicle. For the first The vehicle's local costate vector satisfies the costate update equation: The constraints for pure electric vehicles are: The constraints for fuel cell vehicles are as follows: The constraints for hybrid vehicles are: Among them, P drive,i For the first The driving force of the car For the first Vehicle motor output power, For the first The vehicle's battery state of charge. For the first Vehicle fuel cell output power, For the first Vehicle engine output power, SOC min The lower limit of the battery's state of charge (SOC) max This represents the upper limit of the battery's state of charge. For the first The lower limit of the output power of a vehicle's fuel cell. For the first The upper limit of the output power of a vehicle's fuel cell. For the first The lower limit of the engine output power of a vehicle. For the first The upper limit of the engine output power of a vehicle; Local optimal control input for: 。

Images