Fuel cell vehicle queue hierarchical collaborative energy management control method and system

Abstract

The invention discloses a fuel cell vehicle queue hierarchical cooperative energy management control method and system, and the method comprises the steps: building a hierarchical cooperative control frame and a fuel cell electric vehicle model based on a network connection fuel cell electric vehicle queue; based on an upper-layer longitudinal motion coordination controller, distributed model prediction control is carried out on the fuel cell electric vehicle model, and the expected torque and speed of the fuel cell electric vehicle are obtained; based on a lower-layer energy management controller, energy management is conducted according to the expected torque and speed of the fuel cell electric vehicle, and the optimal output power of a fuel cell and the optimal output power of a lithium battery are obtained; and fuel cell vehicle queue layered collaborative energy management control is realized. According to the invention, SoC adjustment and equivalent hydrogen energy consumption minimization can be realized, and reasonable power distribution and energy-saving optimization are realized. The fuel cell vehicle queue hierarchical collaborative energy management control method and system can be widely applied to the technical field of battery energy management.

Description

Technical Field

[0001] The present invention relates to the technical field of battery energy management, and in particular to a method and system for controlling tiered collaborative energy management of fuel cell vehicle platoons. Background Art

[0002] With increasing environmental concerns and the growing demand for sustainable green energy, fuel cell electric vehicles (FCEVs) have become a powerful solution for sustainable transportation due to their high energy efficiency, zero emissions, and long driving range. However, multiple power supply systems increase system operational complexity. Therefore, ensuring stable, efficient, and reliable operation of multiple power supplies has become a key technology for FCEVs. Research on energy management strategies is crucial for FCEVs. Energy management strategies directly influence the operating points of each energy source and effectively allocate power while meeting required power, thereby reducing energy consumption, improving system efficiency, and extending the service life of FCEVs. However, rule-based energy management relies heavily on the calibration of various parameters and thresholds, and therefore cannot guarantee optimal performance under real-world driving conditions. Optimization-based approaches include global optimization and local optimization. Global optimization energy management methods, such as dynamic programming and game theory, optimize control sequences based on static data of specific states. Because these methods require prior knowledge of the global state, online optimal calculation is difficult. Local optimization can be used to approximate global optimization. Local optimization methods calculate control sequences online based on the vehicle's real-time state, ensuring local or instantaneous optimality. Learning-based approaches use data-driven methods, such as reinforcement learning, to learn from dynamic driving conditions to adjust and optimize energy management strategies. However, these approaches are limited by training scenarios and difficult to apply in real time. Model predictive control (MPC)-based approaches have been widely studied due to their inherent robustness and rolling optimization accuracy. However, MPC-based approaches are computationally intensive, and most existing methods focus on optimizing a single power source. Optimizing multiple power sources simultaneously while ensuring real-time performance remains a worthy research question. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a fuel cell vehicle fleet tiered collaborative energy management control method and system, which can achieve SoC adjustment and minimization of equivalent hydrogen energy consumption, and realize reasonable power allocation and energy-saving optimization.

[0004] The first technical solution adopted by the present invention is: a fuel cell vehicle platoon stratified collaborative energy management control method, comprising the following steps:

[0005] Based on a platoon of connected fuel cell electric vehicles, a hierarchical collaborative control framework and a fuel cell electric vehicle model are constructed. The hierarchical collaborative control framework includes an upper-layer longitudinal motion coordination controller and a lower-layer energy management controller.

[0006] Based on the upper-level longitudinal motion coordination controller, a distributed model predictive control is performed on the fuel cell electric vehicle model to obtain the desired torque and speed of the fuel cell electric vehicle;

[0007] Based on the lower-level energy management controller, energy management is performed according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery;

[0008] By combining the optimal output power of the fuel cell with the optimal output power of the lithium battery, hierarchical collaborative energy management control of fuel cell vehicle platoons can be achieved.

[0009] Furthermore, the step of constructing a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a networked fuel cell electric vehicle platoon specifically includes:

[0010] Based on distributed model predictive control, an upper-level longitudinal motion coordination controller is constructed. The upper-level longitudinal motion coordination controller is used to obtain the position, speed, and driving torque information of the lead vehicle and calculate the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle.

[0011] Based on model predictive control, a lower-level energy management controller is constructed to obtain the output power of the fuel cell based on the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle;

[0012] Combining the upper-level longitudinal motion coordination controller with the lower-level energy management controller, a hierarchical collaborative control framework is constructed;

[0013] Equivalently treat the lithium-ion battery as a RINT equivalent model to obtain the open-loop voltage and output power of the lithium battery;

[0014] Determine the state of charge of the lithium battery based on the open-loop voltage and output power of the lithium battery;

[0015] Obtain the output voltage of the fuel cell and combine it with the output power of the lithium battery to determine the total power of the fuel cell electric vehicle;

[0016] According to the total power of the fuel cell electric vehicle, the hydrogen consumption rate of the fuel cell is determined and a fuel cell electric vehicle model is constructed.

[0017] Furthermore, the step of performing distributed model predictive control on the fuel cell electric vehicle model based on the upper-layer longitudinal motion coordination controller to obtain the desired torque and speed of the fuel cell electric vehicle specifically includes:

[0018] Obtain information on the resistance that fuel cell electric vehicles need to overcome during driving and construct the longitudinal dynamic equation of fuel cell electric vehicles;

[0019] Discretizing the longitudinal dynamic equation of the fuel cell electric vehicle to obtain the discretized longitudinal dynamic equation of the fuel cell electric vehicle;

[0020] Based on the constant headway strategy, the desired relative distance between fuel cell electric vehicles is determined and the control target of the fuel cell vehicle platoon is constructed;

[0021] Based on the need for the following vehicle to track the speed and position of the leading vehicle, the reconstructed state space equations and state matrix in the distributed model predictive control are determined;

[0022] Designing a cost function for the control objective of the fuel cell vehicle platoon based on the reconstructed state space equation and the state matrix and converting it into a quadratic programming form to obtain a cost function in a quadratic programming form;

[0023] According to the cost function in the form of quadratic programming, a state following term, a state smoothing term and a state constraint term are added to determine the error matrix, and an optimal control input is obtained through rolling optimization. The optimal control input includes the desired torque and speed of the fuel cell electric vehicle.

[0024] Furthermore, the cost function of the control objective of the fuel cell vehicle fleet is specifically expressed as follows:

[0025]

[0026] In the above formula, J i represents the cost function, Q1, q2, R1, R2, H1, H2, G represent the weight matrix, s p 、v p Indicates the position and speed of the vehicle at the last moment, s h 、v h Indicates the position and speed of the preceding vehicle (s i,j -s des ) 2 and (v i,j -v des ) 2 Indicates the state following item, (s i,j -s p ) 2 and (ν i,j -v p ) 2 represents the state smoothing term, (s i,j -s h -d h,i ) 2 and (v i,j -v h ) 2 represents the state constraint, u i,j Indicates the control quantity.

[0027] Furthermore, the expression of the optimal control input is specifically as follows:

[0028]

[0029] In the above formula, represents the optimal control input, Ξ and Ω represent the inequality matrix, U min Indicates the minimum value of the control quantity, U max Indicates the maximum value of the control amount, J i (k) represents the cost function, U pi (k) represents the control quantity sequence, represents the matrix, U i (k) represents the calculated control quantity, Θ p , Γ p 、F p 、W pi represents the reconstruction matrix, Q block represents the weight matrix, R represents the weight matrix of the control amount, e represents the relaxation factor, x pi (k) represents the state vector, W pi Represents the reconstruction matrix, Y ref Represents the reference value, and k represents the kth moment in the prediction domain.

[0030] Furthermore, the step of performing energy management based on the lower-layer energy management controller according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery specifically includes:

[0031] Determine the rotation speed of the fuel cell electric vehicle and the expected total power of the drive motor according to the expected torque and speed of the fuel cell electric vehicle;

[0032] The state of charge of the lithium battery and the hydrogen consumption rate of the fuel cell are combined and discretized to construct a discretized state space matrix;

[0033] Apply constraints to the discretized state space to obtain the reconstructed state space matrix;

[0034] Based on the reconstructed state space matrix, the upper and lower limits of the lithium battery state of charge, the upper and lower limits of the lithium battery output power, the upper and lower limits of the fuel cell output power, and the upper and lower limits of the fuel cell output power fluctuation are determined to construct the cost function of the lower-level optimization control;

[0035] The desired torque and speed of the fuel cell electric vehicle are optimized according to the cost function of the lower-level optimization control to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery.

[0036] Furthermore, the cost function of the lower-level optimization control is specifically expressed as follows:

[0037]

[0038] SoC min ≤SoC(k+i|k)≤SoC max

[0039] P bat_min ≤P bat (k+i|k)≤P bat_max

[0040] P fc_min ≤P fc (k+i|k)≤P fc_max

[0041] ΔP fc_min ≤ΔP fc (k+i|k)≤ΔP fc_max

[0042] In the above formula, represents the optimal control sequence, α1 and α2 represent weight coefficients, and m total Indicates the total amount of hydrogen on board the vehicle, SoC min , SoC max Indicates the upper and lower limits of the battery state of charge, P bat_min 、P bat_max Indicates the upper and lower limits of lithium-ion battery output power, P fc_min 、P fc_max Indicates the upper and lower limits of the fuel cell output power, ΔP fc_min , ΔP fc_max Indicates the upper and lower limits of fuel cell output power fluctuation, SoC ref Indicates reference value, m eq represents the equivalent hydrogen consumption, i represents the i-th vehicle, and k represents the k-th moment in the prediction time domain.

[0043] The second technical solution adopted by the present invention is: a fuel cell vehicle platoon hierarchical collaborative energy management and control system, comprising:

[0044] The first module is used to build a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a platoon of connected fuel cell electric vehicles. The hierarchical collaborative control framework includes an upper-level longitudinal motion coordination controller and a lower-level energy management controller.

[0045] The second module is used to perform distributed model predictive control on the fuel cell electric vehicle model based on the upper-level longitudinal motion coordination controller to obtain the expected torque and speed of the fuel cell electric vehicle;

[0046] The third module is used to perform energy management based on the lower-level energy management controller according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery;

[0047] The fourth module is used to combine the optimal output power of the fuel cell with the optimal output power of the lithium battery to achieve hierarchical collaborative energy management control of fuel cell vehicle queues.

[0048] The beneficial effects of the method and system of the present invention are as follows: the present invention constructs a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a networked fuel cell electric vehicle fleet, wherein the upper layer in the hierarchical collaborative control framework is responsible for longitudinal motion coordination to ensure the safety and stability of the system, and the lower layer reasonably distributes power according to the output instructions of the upper layer to optimize the energy use of the system, and then based on the upper layer longitudinal motion coordination controller, the fuel cell electric vehicle model is subjected to distributed model predictive control to obtain the desired torque and speed of the fuel cell electric vehicle, and based on the lower layer energy management controller, energy management is performed according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the lithium battery. The optimal output power of the battery can calculate in real time the expected torque required for the following vehicle in the queue to track the lead vehicle, and add state following terms, state smoothing terms, and state constraint terms to the algorithm objective function to improve the overall following performance of the queue. At the same time, it realizes the predictive optimization of the vehicle spacing and speed, ensuring safe, adaptive, and energy-saving following performance under system constraints. Finally, the optimal output power of the fuel cell is combined with the optimal output power of the lithium battery to realize the hierarchical collaborative energy management control of the fuel cell vehicle queue. An MPC-based energy management strategy is developed. Through rolling optimization and constraint processing, SoC adjustment and minimization of equivalent hydrogen energy consumption are realized, and reasonable power distribution and energy-saving optimization are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flowchart of the steps of a fuel cell vehicle platoon stratified collaborative energy management control method of the present invention;

[0050] Figure 2 This is a structural block diagram of a fuel cell vehicle platoon hierarchical collaborative energy management control system according to the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are provided for ease of description only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted based on the understanding of those skilled in the art.

[0052] First of all, it should be noted that currently, energy management strategies can be divided into three categories: rule-based strategies, optimization-based strategies, and learning-based strategies. The first two strategies are more widely used. The model predictive control-based method transforms the global optimal control problem into a local optimal control problem within a limited prediction range. With the improvement of prediction accuracy, model predictive control can approach the global optimal solution. At the same time, this method has strong robustness and is suitable for power distribution in multi-power systems. Based on the concept of non-uniform sampling time, Gomozov developed a real-time management strategy based on model predictive control to keep the key parameters of batteries and supercapacitors within a safe range. Kofler used dynamic programming to provide a predictive energy management for fuel cell electric vehicles. This strategy calculates the optimal driving cost offline based on static path information, and calculates the model predictive control online based on the linear driving cost to minimize the fuel consumption of the remaining journey.

[0053] Deploying connected fuel cell electric vehicles in fleets can further reduce aerodynamic drag, improve traffic flow, and increase overall energy savings. However, this requires coordinated control strategies to ensure vehicle-level safety and system-level energy optimization. Fleet control and energy management have traditionally been viewed as separate issues, and a lack of integration between the two layers often leads to poor performance, especially under dynamic and uncertain driving conditions.

[0054] Collaborative optimization methods can significantly improve the performance of electric vehicle platoons. It is foreseeable that connected fuel cell electric vehicles will adopt higher-level connectivity technologies such as V2V, V2I, and V2X to improve performance. Real-time data sharing within connected fuel cell electric vehicles will be achieved through V2V communication. This will help establish a comprehensive framework to guide fuel cell electric vehicle platoons towards more coordinated, intelligent, and safer systems. This can improve safety and reduce wind resistance, thereby reducing hydrogen consumption and costs. To integrate fuel cell electric vehicles and fleet control technologies, Hu Xiaosong and others from Chongqing University used the consistent alternating direction multiplier method to create a hierarchical control strategy for homogeneous heavy-duty modular fuel cell vehicle fleets.

[0055] Eco-driving for connected fuel cell electric vehicles (FCEVs) involves a coupled problem of speed planning and energy management. Connected FCEVs can leverage traffic information to optimize cruising and power allocation. However, in complex traffic scenarios, given the complex internal system dynamics and constraints of EV fleets, integrating vehicle-level safety and system-level energy efficiency into a unified prediction framework while considering multiple objectives such as vehicle speed, relative distance, and power limits remains a current research challenge.

[0056] Based on this, embodiments of the present invention aim to coordinate the longitudinal motion of fuel cell vehicle platoons, improve energy efficiency, and optimize ecological coordination. To maintain safe inter-vehicle spacing through coordinated longitudinal control, the upper-level controller is designed based on distributed model predictive control and vehicle-to-vehicle communication. Simultaneously, the lower-level energy management strategy optimizes power distribution based on the speed coordinated by the upper-level controller, reduces hydrogen consumption, and regulates the battery's state of charge.

[0057] Reference Figure 1 The present invention provides a fuel cell vehicle platoon hierarchical collaborative energy management control method, the method comprising the following steps:

[0058] S100, constructing a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a platoon of networked fuel cell electric vehicles, wherein the hierarchical collaborative control framework includes an upper-layer longitudinal motion coordination controller and a lower-layer energy management controller;

[0059] S110. Based on distributed model predictive control, construct an upper-level longitudinal motion coordination controller, wherein the upper-level longitudinal motion coordination controller is used to obtain the position, speed, and driving torque information of the lead vehicle and calculate the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle;

[0060] S120, constructing a lower-level energy management controller based on model predictive control, which is used to obtain the output power of the fuel cell based on the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle;

[0061] S130, combining the upper-layer longitudinal motion coordination controller and the lower-layer energy management controller to construct a hierarchical collaborative control framework;

[0062] In this embodiment, at the upper layer, networked fuel cell electric vehicles achieve inter-vehicle communication through networking technology and rely on common protocols to form a fleet. The vehicle platoon communication topology adopts a leading vehicle-leading vehicle following topology. Vehicles in the fleet can obtain the position, speed, and driving torque information of the leading vehicle or the leading vehicle through inter-vehicle communication. In order to coordinate the longitudinal motion control of the vehicles and maintain a safe inter-vehicle spacing, the upper-layer controller is designed based on distributed model predictive control (DMPC) to calculate the expected torque and speed required for each vehicle in the fleet to follow the leading vehicle.

[0063] At the same time, at the lower level, each fuel cell vehicle outputs the corresponding expected power according to the expected torque required by the calculation. Taking into account energy saving and power source life, based on model predictive control (MPC), the lower-level energy management strategy optimizes power distribution according to the speed and torque coordinated by the upper-level controller, reduces hydrogen consumption and adjusts the battery's state of charge.

[0064] In summary, the hierarchical control framework for vehicle platoons is designed based on the advanced communication technology (V2V) of connected fuel cell vehicles. Homogeneous fuel cell vehicle platoons are formed. To ensure safe, stable and energy-efficient driving of the entire system, the platoon system is subjected to hierarchical collaborative control. The upper layer is responsible for longitudinal motion coordination to ensure system safety and stability. The lower layer reasonably allocates power based on the output instructions of the upper layer to optimize the system's energy usage.

[0065] Among them, the hierarchical control framework of the platoon system includes the upper layer using a distributed model predictive control algorithm to coordinate the longitudinal movement of each fuel cell vehicle in the platoon and output the expected torque for each vehicle; for the energy management of each fuel cell vehicle in the platoon, the lower layer uses a model predictive control algorithm to continuously optimize the power distribution between power sources based on the received information such as the expected torque.

[0066] S140, equating the lithium-ion battery to a RINT equivalent model to obtain an open-loop voltage and output power of the lithium battery;

[0067] In this embodiment, the lithium-ion battery serves as an auxiliary power source in the fuel cell vehicle. The lithium-ion battery is equated to the RINT equivalent model. Based on the RINT equivalent model, the battery's open-loop voltage and output power are obtained as follows:

[0068] V op =U battery +I battery R battery

[0069] P bat =I battery (V op -I battery R battery )

[0070] In the above formula, U battery is the terminal voltage; R battery is the battery equivalent resistance.

[0071] S150, determining the state of charge of the lithium battery according to the open-loop voltage and output power of the lithium battery;

[0072] In this embodiment, in the RINT equivalent model, the open-loop voltage and equivalent resistance vary with the SoC, so the calculation formula for the battery state of charge is as follows:

[0073]

[0074] In the above formula, C battery Indicates the battery capacity.

[0075] S160, obtaining the output voltage of the fuel cell and combining it with the output power of the lithium battery to determine the total power of the fuel cell electric vehicle;

[0076] In this embodiment, the polarization curve is used to characterize the fuel cell system. N fuel cells are connected in series to provide output power. Due to the loss, the Nernst voltage U n There will be some loss, so the fuel cell output voltage can be shown as follows:

[0077] V fc =N(U n -U act -U oh m -U con )

[0078]

[0079]

[0080] U ohm =I fc R fc

[0081]

[0082] In the above formula, N is the number of fuel cells, V fc is the fuel cell voltage, T fc is the battery stack temperature, and the ideal battery stack temperature is set to 80°C; are the pressures of oxygen and hydrogen, I fc is the current of the battery stack; fc is an empirical parameter, R fc is the internal resistance, φ is a constant, I max is the maximum current.

[0083] The total power provided by the fuel cell and lithium-ion battery is expressed as:

[0084] P fc =(P em -P bat +P auxiliary ) / η DC

[0085] P fc_min ≤P fc ≤P fc_max

[0086] Among them, P auxiliary is the auxiliary load; P fc_min and P fc_max are the minimum power and maximum power of the fuel cell respectively.

[0087] S170: Determine the hydrogen consumption rate of the fuel cell based on the total power of the fuel cell electric vehicle and construct a fuel cell electric vehicle model.

[0088] In this embodiment, the hydrogen consumption rate of the fuel cell can be expressed as:

[0089]

[0090] Among them, η fc is the fuel cell efficiency; C LHV It is the low calorific value of hydrogen.

[0091] Therefore, for fuel cell electric vehicle modeling, in order to achieve platoon vehicle-level safety and system-level energy optimization, it is necessary to model the fuel cell vehicle platoon and fuel cell vehicle power system. At the power system level, the vehicle's power source is modeled to simulate vehicle power transmission. Fuel cell electric vehicle modeling includes setting the fuel cell as the main power source, characterizing the characteristics of the fuel cell with polarization curves, and calculating the instantaneous hydrogen consumption rate of the fuel cell in combination with power; equipping the lithium-ion battery as an auxiliary power source, simplifying the lithium-ion battery into a RINT model, and calculating the output voltage and state of charge (SoC) of the lithium-ion battery.

[0092] S200, performing distributed model predictive control on the fuel cell electric vehicle model based on the upper-layer longitudinal motion coordination controller to obtain the desired torque and speed of the fuel cell electric vehicle;

[0093] First, it should be noted that the upper-level longitudinal motion coordination controller uses shared information obtained through inter-vehicle communication technology, including speed and position information, to calculate the desired safe distance between vehicles. It uses position, speed, and torque as state quantities, adopts distributed model predictive control, and utilizes cloud computing to calculate in real time the desired torque required for the following vehicles in the queue to track the lead vehicle. It also adds state following terms, state smoothing terms, and state constraint terms to the algorithm's objective function to improve the overall following performance of the queue.

[0094] For the upper-level longitudinal motion coordination controller, first, each fuel cell vehicle in the queue obtains the position and speed information of the preceding vehicle and the pilot vehicle through V2V technology; then, the longitudinal dynamic equation is discretized, the predicted value of the state quantity is calculated in the prediction time domain, and the state following term, state smoothing term, and state constraint term are added to the objective function. The expected torque of each vehicle is solved through rolling optimization.

[0095] S210, obtaining information on the resistance that the fuel cell electric vehicle needs to overcome during driving, and constructing a longitudinal dynamic equation of the fuel cell electric vehicle;

[0096] In this embodiment, when a vehicle is traveling on a flat surface, it is subject to resistances such as wind resistance, rolling resistance, and acceleration resistance. The combined force of these resistances constitutes the external force that the vehicle needs to overcome during travel, which is expressed as:

[0097]

[0098] Where m is the vehicle mass, g is the acceleration due to gravity, f is the rolling resistance coefficient, ρ is the air density, and C d is the drag coefficient, A f is the windshield area, δ m is the rotational mass conversion factor, and i represents the i-th vehicle in the queue.

[0099] A single vehicle in a convoy is usually modeled as a node. According to the above formula, the longitudinal dynamic equation of each vehicle can be described as:

[0100]

[0101] Among them, the state variables include position, speed and driving torque, η is the efficiency of the transmission system, r is the vehicle radius, τ fcev is the transmission system response delay time, u i (t) is the control quantity.

[0102] S220, discretizing the longitudinal dynamics equation of the fuel cell electric vehicle to obtain a discretized longitudinal dynamics equation of the fuel cell electric vehicle;

[0103] In this embodiment, it is discretized and the discrete state space expression is derived as follows:

[0104]

[0105]

[0106] Among them, x pi =[s i ,v i ,T i ] T ;y pi =[s i ,v i ] T ; ΔT is the sampling time.

[0107] S230, determining the expected relative distance between fuel cell electric vehicles and establishing a control target for the fuel cell vehicle platoon based on the constant headway strategy;

[0108] In this embodiment, based on the constant headway strategy, the expected relative distance is calculated as follows:

[0109]

[0110] Where h is the headway time, and d0 is the safe relative distance between adjacent vehicles when they are stationary.

[0111] In order to maintain the continuous driving of fuel cell electric vehicles, the control objectives of the fuel cell vehicle platoon are:

[0112]

[0113] In the above formula, s fcev,i represents the position of the i-th vehicle, s fcev,0 Indicates the position of the lead vehicle, d i,0 represents the expected relative distance, v fcev,i represents the speed of the i-th vehicle, v fcev,0 Indicates the speed of the leading vehicle.

[0114] S240, determining a reconstructed state space equation and state matrix in a distributed model predictive control based on the need for the following vehicle to track the speed and position of the leading vehicle;

[0115] In this embodiment, the following vehicle needs to track the speed and position of the leading vehicle. Then, the reconstructed state space equation and state matrix in the distributed model predictive control are given, and their expressions are:

[0116] Y pi (k+1)=Γ p X pi (k)+Θ p U pi (k)+F p W pi (k)

[0117] Y pi (k+1)=[y pi (k+1),y pi (k+2),…,y pi (k+N p )]

[0118] U pi (k)=[u pi (k+1),u pi (k+2),…,u pi (k+N c )]

[0119] W pi (k)=[w pi (k+1),w pi (k+2),…,w pi (k+Np )]

[0120]

[0121]

[0122] Among them, N p is the prediction time domain, N c It is the control time domain.

[0123] S250, designing a cost function for the control target of the fuel cell vehicle platoon based on the reconstructed state space equation and the state matrix and converting the cost function into a quadratic programming form to obtain a cost function in the quadratic programming form;

[0124] In this embodiment, the cost function is designed as follows:

[0125]

[0126] In the above formula, J i represents the cost function, Q1, Q2, R1, R2, H1, H2, G represent the weight matrix, s p 、v p Indicates the position and speed of the vehicle at the last moment, s h 、v h Indicates the position and speed of the preceding vehicle (s i,j -s des ) 2 and (v i,j -v des ) 2 Indicates the state following item, (s i,j -s p ) 2 and (v i,j -v p ) 2 represents the state smoothing term, (s i,j -s h -d h,i ) 2 and (v i,j -v h ) 2 represents the state constraint, u i,j Denotes the control quantity, the fleet can reduce the air resistance and model the air resistance as a disturbance term.

[0127] S260. According to the cost function in the form of quadratic programming, a state following term, a state smoothing term, and a state constraint term are added to determine the error matrix, and the optimal control input is obtained through rolling optimization, where the optimal control input includes the desired torque and speed of the fuel cell electric vehicle.

[0128] In this embodiment, in order to perform calculation and solution, the above formula is converted into a quadratic programming form, and the derivation result is as follows:

[0129]

[0130] Where e is the penalty weight; ∈ is the relaxation factor, Q block is the weight matrix.

[0131] in:

[0132]

[0133] Q track =diag(Q1,Q2,…,Q1,Q2)

[0134] Q smooth =diag(R1,R2,…,R1,R2)

[0135] Q neigh =diag(H1,H2,…,H1,H2)

[0136] The error matrix is ​​given by the following formula, which is expressed as:

[0137] E=Y pi -Y ref =[E track ;E smooth ;E neigh ]

[0138]

[0139] Therefore, the optimal control input can be obtained through rolling optimization Its expression is:

[0140]

[0141] In the above formula, represents the optimal control input, Ξ and Ω represent the inequality matrix, U min Indicates the minimum value of the control quantity, U max Indicates the maximum value of the control amount, J i (k) represents the cost function, U pi (k) represents the control quantity sequence, represents the matrix, U i (k) represents the calculated control quantity, Θ p , Γ p 、F p 、W pi represents the reconstruction matrix, Q block represents the weight matrix, R represents the weight matrix of the control amount, e represents the relaxation factor, xpi (k) represents the state vector, W pi Represents the reconstruction matrix, Y ref Represents the reference value, and k represents the kth moment in the prediction domain.

[0142] S300, based on the lower-layer energy management controller, perform energy management according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery;

[0143] First, it should be noted that the lower-level energy-saving energy management controller calculates the required total desired power based on the desired torque output by the upper-level motion coordination controller. Then, based on the total required vehicle power, combined with the lithium-ion battery state of charge (SoC) and the instantaneous equivalent hydrogen consumption rate, the optimal control output at time k, namely the output power of the fuel cell and lithium-ion battery, is solved based on the model predictive control method.

[0144] The lower-level energy-saving energy management controller first calculates the vehicle's motor speed and total required power based on the desired torque output by the upper-level controller; uses the lithium-ion battery state of charge and the instantaneous equivalent hydrogen consumption rate as state quantities of the control model and discretizes the equation; calculates the predicted value of the state quantity within the prediction time domain, adds system constraints, including the upper and lower limits of the fuel cell output power, the upper and lower limits of the lithium-ion battery output power, the upper and lower limits of the battery state of charge, and the upper and lower limits of the fuel cell output power fluctuation, and performs rolling optimization to solve the optimal output power of the fuel cell.

[0145] S310, determining the rotation speed of the fuel cell electric vehicle and the expected total power of the drive motor according to the expected torque and speed of the fuel cell electric vehicle;

[0146] In this embodiment, the upper-level controller ensures safe and stable driving of the vehicle, and the lower-level energy management strategy comprehensively optimizes SoC and hydrogen consumption, reduces hydrogen consumption, and adjusts SoC.

[0147] First, based on the desired torque and speed of each vehicle calculated by the upper-level longitudinal motion coordination controller, the speed of each vehicle and the desired total power of the drive motor are calculated. The expressions are:

[0148] w m =v i / r

[0149]

[0150] Among them, η m is the motor efficiency.

[0151] S320, combining the state of charge of the lithium battery and the hydrogen consumption rate of the fuel cell and performing discretization processing to construct a discretized state space matrix;

[0152] In this embodiment, the state of charge of the battery and the instantaneous equivalent hydrogen consumption rate are described as follows:

[0153]

[0154] Where S(t) represents the equivalence factor. The equivalence factor is adjusted according to different SoC deviations to achieve adjustment and adaptability. Its formula is as follows:

[0155] S(t)=S0+K s (SoC ref -SoC(t))

[0156] Where, S0=1.9, K s is the scale factor, SoC ref is the reference SoC value.

[0157] Combine the battery's state of charge and instantaneous equivalent hydrogen consumption rate and discretize them. In the discretized state space equation, P fc Set as control input, (P em (t)+P auxiliary ) is set as interference, aiming to calculate the optimal P fc , to distribute power reasonably. The status is as follows:

[0158]

[0159] Among them, H, G, F u 、F w is the state space equation matrix. It is worth mentioning that in order to improve the accuracy of the model, F u 、F w are time-varying. However, they remain constant within the prediction horizon and are updated at the next moment after each solution. The discretized state space equations are shown below:

[0160] Y e (k+1)=Γ e X e (k)+Θ e U e (k)+F e W e (k)

[0161] X e =[x e (k+1),x e (k+2),…x e (k+N p )] T

[0162] U e=[P fc (k),P fc (k+1),…P fc (k+N c -1)] T ;

[0163] W e =[P em (k),P em (k+1),…P em (k+N p -1)] T

[0164] In the above formula, Y e represents the reconstruction vector of the observation vector, X e Represents the reconstruction vector of the state quantity, U e Represents the reconstruction vector of the control quantity, W e Reconstructed vector representing the interference amount.

[0165] S330, applying constraints to the discretized state space to obtain a reconstructed state space matrix;

[0166] In this embodiment, the reconstructed state space matrix is ​​given by the following formula, which is expressed as follows:

[0167]

[0168]

[0169] In the above formula, Γ e 、Θ e 、F e Represents the reconstruction matrix.

[0170] S340, determining the upper and lower limits of the lithium battery state of charge, the upper and lower limits of the lithium battery output power, the upper and lower limits of the fuel cell output power, and the upper and lower limits of the fuel cell output power fluctuation based on the reconstructed state space matrix, and constructing a cost function for the lower-level optimization control;

[0171] S350: Optimize the desired torque and speed of the fuel cell electric vehicle according to the cost function of the lower-level optimization control to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery.

[0172] In this embodiment, MPC has a natural advantage in solving multi-objective optimization problems with system constraints. In this hybrid system, both the fuel cell and the lithium-ion battery have power upper and lower limits. In order to ensure the smooth output of the fuel cell and avoid drastic dynamic changes in the load, ΔP fc Constraints were imposed.

[0173] Therefore, the cost function is:

[0174]

[0175] SoC min ≤SoC(k+i|k)≤SoC max

[0176] P bat_min ≤P bat (k+i|k)≤P bat_max

[0177] P fc_min ≤P fc (k+i|k)≤P fc_max

[0178] ΔP fc_min ≤ΔP fc (k+i|k)≤ΔP fc_max

[0179] In the above formula, represents the optimal control sequence, α1 and α2 represent weight coefficients, and m total Indicates the total amount of hydrogen on board the vehicle, SoC min , SoC max Indicates the upper and lower limits of the battery state of charge, P bat_min 、P bat_max Indicates the upper and lower limits of lithium-ion battery output power, P fc_min 、P fc_max Indicates the upper and lower limits of the fuel cell output power, ΔP fc_min , ΔP fc_max Indicates the upper and lower limits of fuel cell output power fluctuation, SoC ref Indicates reference value, m eq represents the equivalent hydrogen consumption, i represents the i-th vehicle, and k represents the k-th moment in the prediction time domain.

[0180] Four sets of inequalities define the system constraints, including the upper and lower limits of the battery state of charge, the upper and lower limits of the lithium-ion battery output power, the upper and lower limits of the fuel cell output power, and the upper and lower limits of the fuel cell output power fluctuation.

[0181] S400: Combine the optimal output power of the fuel cell and the optimal output power of the lithium battery to achieve tiered collaborative energy management control of fuel cell vehicle platoons.

[0182] In this embodiment, the calculated output power of the fuel cell and lithium-ion battery is transmitted to the battery management system via the CAN bus, and the desired power is output according to the control instruction to provide power for the drive motor and control the longitudinal movement of the fuel cell vehicle.

[0183] In summary, the embodiment of the present invention integrates distributed model predictive control for upper-level longitudinal motion coordination and an energy management strategy based on model predictive control for lower-level power allocation. Simulation results under NEDC conditions show that the upper-level controller effectively ensures safe and stable following behavior between vehicles through V2V communication, while the lower-level strategy significantly improves energy efficiency and reduces hydrogen consumption. Compared with the energy management methods based on the state machine and the minimum equivalent hydrogen consumption, hydrogen consumption is reduced by 19.27% ​​and 7.95%, respectively; SoC control is improved near the target value; fuel cell operation is more stable and efficient, with smaller power fluctuations; and efficiency distribution is more balanced. The proposed hierarchical control strategy is effective and robust in improving the safety, energy economy and collaborative performance of connected fuel cell vehicle fleets.

[0184] Therefore, compared with the prior art, the embodiments of the present invention have the following improvements:

[0185] 1) Considering the real-time nature of fuel cell vehicle platooning and the joint optimization of car-following behavior and energy management, an environmentally friendly hierarchical control strategy for fuel cell electric vehicle platooning based on the DMPC-MPC method is proposed, aiming to ensure car-following performance and minimize energy consumption.

[0186] 2) To enhance the stability and coordination of fuel cell vehicles, a distributed model predictive control strategy based on a fixed headway strategy was developed. This strategy leverages vehicle-to-vehicle communication to exchange real-time information while simultaneously achieving predictive optimization of vehicle-to-vehicle spacing and speed, ensuring safe, adaptive, and energy-efficient car-following performance within system constraints.

[0187] 3) In order to reduce hydrogen energy consumption and extend the service life of the power supply, an MPC-based energy management strategy was developed. Through rolling optimization and constraint processing, SoC adjustment and equivalent hydrogen energy consumption were minimized to achieve reasonable power allocation and energy-saving optimization.

[0188] Reference Figure 2 , a fuel cell vehicle platoon hierarchical collaborative energy management and control system, comprising:

[0189] The first module 201 is used to build a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a networked fuel cell electric vehicle fleet. The hierarchical collaborative control framework includes an upper-layer longitudinal motion coordination controller and a lower-layer energy management controller.

[0190] The second module 202 is used to perform distributed model predictive control on the fuel cell electric vehicle model based on the upper longitudinal motion coordination controller to obtain the desired torque and speed of the fuel cell electric vehicle;

[0191] The third module 203 is used to perform energy management based on the lower-layer energy management controller according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery;

[0192] The fourth module 204 is used to combine the optimal output power of the fuel cell and the optimal output power of the lithium battery to achieve hierarchical collaborative energy management control of the fuel cell vehicle fleet.

[0193] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0194] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A fuel cell vehicle platoon hierarchical collaborative energy management control method, characterized in that: The following steps are involved: Based on a platoon of connected fuel cell electric vehicles, a hierarchical collaborative control framework and a fuel cell electric vehicle model are constructed. The hierarchical collaborative control framework includes an upper-layer longitudinal motion coordination controller and a lower-layer energy management controller. Based on the upper-level longitudinal motion coordination controller, a distributed model predictive control is performed on the fuel cell electric vehicle model to obtain the desired torque and speed of the fuel cell electric vehicle; Based on the lower-level energy management controller, energy management is performed according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery; By combining the optimal output power of the fuel cell with the optimal output power of the lithium battery, hierarchical collaborative energy management control of fuel cell vehicle platoons can be achieved.

2. A fuel cell vehicle platoon stratified collaborative energy management control method according to claim 1, characterized in that: The step of constructing a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a networked fuel cell electric vehicle platoon specifically includes: Based on distributed model predictive control, an upper-level longitudinal motion coordination controller is constructed. The upper-level longitudinal motion coordination controller is used to obtain the position, speed, and driving torque information of the lead vehicle and calculate the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle. Based on model predictive control, a lower-level energy management controller is constructed to obtain the output power of the fuel cell based on the expected torque and speed required for each vehicle in the platoon to follow the lead vehicle; Combining the upper-level longitudinal motion coordination controller with the lower-level energy management controller, a hierarchical collaborative control framework is constructed; Equivalently treat the lithium-ion battery as a RINT equivalent model to obtain the open-loop voltage and output power of the lithium battery; Determine the state of charge of the lithium battery based on the open-loop voltage and output power of the lithium battery; Obtain the output voltage of the fuel cell and combine it with the output power of the lithium battery to determine the total power of the fuel cell electric vehicle; According to the total power of the fuel cell electric vehicle, the hydrogen consumption rate of the fuel cell is determined and a fuel cell electric vehicle model is constructed.

3. A fuel cell vehicle platoon stratified collaborative energy management control method according to claim 2, characterized in that: The step of performing distributed model predictive control on the fuel cell electric vehicle model based on the upper-layer longitudinal motion coordination controller to obtain the desired torque and speed of the fuel cell electric vehicle specifically includes: Obtain information on the resistance that fuel cell electric vehicles need to overcome during driving and construct the longitudinal dynamic equation of fuel cell electric vehicles; Discretizing the longitudinal dynamic equation of the fuel cell electric vehicle to obtain the discretized longitudinal dynamic equation of the fuel cell electric vehicle; Based on the constant headway strategy, the desired relative distance between fuel cell electric vehicles is determined and the control target of the fuel cell vehicle platoon is constructed; Based on the need for the following vehicle to track the speed and position of the leading vehicle, the reconstructed state space equations and state matrix in the distributed model predictive control are determined; Designing a cost function for the control objective of the fuel cell vehicle platoon based on the reconstructed state space equation and the state matrix and converting it into a quadratic programming form to obtain a cost function in a quadratic programming form; According to the cost function in the form of quadratic programming, a state following term, a state smoothing term and a state constraint term are added to determine the error matrix, and an optimal control input is obtained through rolling optimization. The optimal control input includes the desired torque and speed of the fuel cell electric vehicle.

4. A fuel cell vehicle platoon stratified collaborative energy management control method according to claim 3, characterized in that: The cost function expression of the control objective of the fuel cell vehicle platoon is specifically as follows: In the above formula, J i represents the cost function, Q1, Q2, R1, R2, H1, H2, G represent the weight matrix, s p 、v p Indicates the position and speed of the vehicle at the last moment, s h 、v h Indicates the position and speed of the preceding vehicle (s i,j -s des ) 2 and (v i,j -v des ) 2 Indicates the state following item, (s i,j -s p ) 2 and (v i,j -v p ) 2 represents the state smoothing term, (s i,j -s h -d h,i ) 2 and (v i,j -v h ) 2 represents the state constraint, u i,j Indicates the control quantity.

5. A fuel cell vehicle platoon hierarchical collaborative energy management control method according to claim 4, characterized in that: The expression of the optimal control input is specifically as follows: In the above formula, represents the optimal control input, Ξ and Ω represent the inequality matrix, U min Indicates the minimum value of the control quantity, U max Indicates the maximum value of the control amount, J i (k) represents the cost function, U pi (k) represents the control quantity sequence, represents the matrix, U i (k) represents the calculated control quantity, Θ p , Γ p 、F p 、W pi represents the reconstruction matrix, Q block represents the weight matrix, R represents the weight matrix of the control amount, e represents the relaxation factor, x pi (k) represents the state vector, W pi Represents the reconstruction matrix, Y ref Represents the reference value, and k represents the kth moment in the prediction domain.

6. A fuel cell vehicle platoon hierarchical collaborative energy management control method according to claim 5, characterized in that: The step of performing energy management based on the lower-layer energy management controller according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery specifically includes: Determine the rotation speed of the fuel cell electric vehicle and the expected total power of the drive motor according to the expected torque and speed of the fuel cell electric vehicle; The state of charge of the lithium battery and the hydrogen consumption rate of the fuel cell are combined and discretized to construct a discretized state space matrix; Apply constraints to the discretized state space to obtain the reconstructed state space matrix; Based on the reconstructed state space matrix, the upper and lower limits of the lithium battery state of charge, the upper and lower limits of the lithium battery output power, the upper and lower limits of the fuel cell output power, and the upper and lower limits of the fuel cell output power fluctuation are determined to construct the cost function of the lower-level optimization control; The desired torque and speed of the fuel cell electric vehicle are optimized according to the cost function of the lower-level optimization control to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery.

7. A fuel cell vehicle platoon stratified collaborative energy management control method according to claim 6, characterized in that: The expression of the cost function of the lower-level optimization control is specifically as follows: SoC min ≤SoC(k+i∣k)≤SoC max P bat_min ≤P bat (k+i∣k)≤P bat_max P fc_min ≤P fc (k+i∣k)≤P fc_max ΔP fc_min ≤ΔP fc (k+i∣k)≤ΔP fc_max In the above formula, represents the optimal control sequence, α1 and α2 represent weight coefficients, and m total Indicates the total amount of hydrogen on board the vehicle, SoC min , SoC max Indicates the upper and lower limits of the battery state of charge, P bat_min 、P bat_max Indicates the upper and lower limits of lithium-ion battery output power, P fc_min 、P fc_max Indicates the upper and lower limits of the fuel cell output power, ΔP fc_min , ΔP fc_max Indicates the upper and lower limits of fuel cell output power fluctuation, SoC ref Indicates reference value, m eq represents the equivalent hydrogen consumption, i represents the i-th vehicle, and k represents the k-th moment in the prediction time domain.

8. A fuel cell vehicle platoon hierarchical collaborative energy management control system, characterized in that: Includes the following modules: The first module is used to build a hierarchical collaborative control framework and a fuel cell electric vehicle model based on a platoon of connected fuel cell electric vehicles. The hierarchical collaborative control framework includes an upper-level longitudinal motion coordination controller and a lower-level energy management controller. The second module is used to perform distributed model predictive control on the fuel cell electric vehicle model based on the upper-level longitudinal motion coordination controller to obtain the expected torque and speed of the fuel cell electric vehicle; The third module is used to perform energy management based on the lower-level energy management controller according to the desired torque and speed of the fuel cell electric vehicle to obtain the optimal output power of the fuel cell and the optimal output power of the lithium battery; The fourth module is used to combine the optimal output power of the fuel cell with the optimal output power of the lithium battery to achieve hierarchical collaborative energy management control of fuel cell vehicle queues.

Images