CN112925209B - Fuel cell automobile model-interference double-prediction control energy management method and system - Google Patents

Abstract

The invention relates to a fuel cell vehicle model-disturbance double predictive control energy management method and system to minimize the equivalent hydrogen consumption of a fuel cell hybrid power system. The fuel cell vehicle hybrid power system consists of a vehicle speed sensor, a vehicle speed predictor, an energy management controller, a DC/DC converter, a fuel cell engine, and a power battery. Correct future speed. Model-disturbance dual predictive control energy management combines the future power demand of the vehicle, and distributes the power of the fuel cell engine and the power battery through the energy management controller. The system prediction model predicts the parameters including the state of charge of the power battery, and the disturbance prediction uses the predicted vehicle speed Calculate the future load disturbance and input the future load disturbance into the system prediction model, which enhances the accuracy of the system prediction model in the traditional model predictive control, thereby improving the optimal control action of the rolling optimization output.

Description

Fuel cell vehicle model-interference dual predictive control energy management method and system

Technical Field

The invention relates to a fuel cell hybrid power system, and in particular to a fuel cell vehicle model-disturbance dual predictive control energy management method and system.

Background Art

With the continuous development of global science and technology and economy, energy consumption is gradually increasing, and the energy crisis of environmental pollution is becoming increasingly serious. New energy power generation devices that meet the concept of sustainable development have become a research hotspot in the energy field. Fuel cell engines have attracted widespread attention in the automotive field for their advantages of low pollution and high energy conversion efficiency. However, when the output power of the fuel cell engine changes greatly, the membrane electrode assembly of the fuel cell stack is prone to degradation, resulting in a short service life of the fuel cell engine. In order to solve the application problems of fuel cell engines, the structure of fuel cell hybrid power systems is generally adopted. Adding power batteries with good dynamic response capabilities to the power system makes up for the lack of response capabilities of the fuel cell engine. Therefore, how to adjust the power output of the fuel cell engine and the power battery under different levels of power demand to achieve efficient and stable operation of the power system has become an important issue that needs to be solved urgently.

Among many energy management methods, the model predictive control energy management strategy can effectively handle multivariable and constrained problems, has strong robustness and stability, and is widely used in the control and management of strong nonlinear systems such as fuel cells. However, the system response prediction in traditional model predictive control is calculated in the prediction time domain based on the state variable values and disturbance values collected in real time. When the external disturbance changes, the system response prediction of traditional model predictive control is not completely accurate, resulting in the calculated control action being only an approximate optimal solution, which will affect the economy and durability of the fuel cell hybrid system. Therefore, traditional model predictive control has the potential and necessity for improvement when applied to fuel cell vehicle energy management.

Summary of the invention

The purpose of the present invention is to provide a fuel cell vehicle model-disturbance dual predictive control energy management method and system, which allocates the power requirements of the fuel cell engine and the power battery in real time based on the minimum equivalent hydrogen consumption, so that the power system can work stably and efficiently.

To achieve the above object, the technical solution of the present invention is: a fuel cell vehicle model-disturbance dual predictive control energy management method, comprising the following steps:

Step S1, using historical vehicle speed information to predict future vehicle speed, calculating the prediction error between historical vehicle speed information and future vehicle speed information, and establishing a vehicle speed error correction model to correct the predicted future vehicle speed information;

Step S2, inputting the future vehicle speed information into the energy management controller, and calculating the future vehicle power demand in combination with the vehicle dynamics model;

Step S3: Combined with the future vehicle power demand information, important system parameters including the state of charge of the power battery are predicted based on the power system linear prediction model, an equivalent hydrogen consumption objective function is established, and the optimal solution of the equivalent hydrogen consumption objective function is calculated through the optimization algorithm to obtain the optimal power distribution of the fuel cell engine and the power battery.

In one embodiment of the present invention, in step S1, the vehicle speed prediction method used to predict the future vehicle speed using the historical vehicle speed information is a third-order exponential smoothing method, the input is the historical vehicle speed information, and the output is the predicted future vehicle speed information.

In an embodiment of the present invention, in step S1 , the vehicle speed error correction model is a Markov model.

In one embodiment of the present invention, in step S3, the optimization algorithm for calculating the optimal solution of the equivalent hydrogen consumption objective function is an effective set algorithm.

In one embodiment of the present invention, the model-interference dual prediction refers to state quantity prediction and interference prediction respectively. By establishing a system prediction model, important system parameters including the power battery state of charge are predicted. Interference prediction refers to predicting the vehicle speed and calculating the future load disturbance information in combination with the vehicle dynamics model.

The present invention also provides a fuel cell vehicle model-interference dual predictive control energy management system, comprising: a vehicle speed sensor, a vehicle speed predictor and an energy management controller, a DC/DC converter, a fuel cell engine, and a power battery;

The fuel cell engine is connected in series with the DC/DC converter, and then connected in parallel with the power battery on the DC bus;

The vehicle speed predictor inputs historical vehicle speed information and outputs predicted future vehicle speed information;

The energy management controller inputs predicted future vehicle speed information and uses model-disturbance dual prediction control to output energy distribution between the fuel cell engine and the power battery based on minimum equivalent hydrogen consumption;

The model-disturbance dual predictive control improves the prediction accuracy of the model predictive control by predicting and correcting future vehicle speed information.

In one embodiment of the present invention, the vehicle speed predictor predicts future vehicle speed information based on an exponential smoothing method-Markov correction model.

In one embodiment of the present invention, the exponential smoothing method-Markov correction model calculates the future vehicle speed through historical vehicle speed information combined with the third-order exponential smoothing method, and calculates the prediction error between the historical vehicle speed information and the future vehicle speed information, establishes a Markov model vehicle speed error correction model, and corrects the vehicle speed information predicted based on the exponential smoothing method.

In one embodiment of the present invention, the energy management controller uses a model-disturbance dual predictive control strategy to perform energy management on the calculated future vehicle power demand, and obtains the optimal power allocation of the fuel cell engine and the power battery based on the minimum equivalent hydrogen consumption.

In one embodiment of the present invention, the model-disturbance dual prediction control strategy combines future power demand information, predicts important system parameters including the state of charge of the power battery based on the linear prediction model of the power system, establishes an equivalent hydrogen consumption objective function, calculates the optimal solution of the objective function through the effective set algorithm, and obtains the optimal power allocation between the fuel cell engine and the power battery.

Furthermore, the vehicle speed predictor and energy management controller are embedded in a single chip microcomputer used in the vehicle controller.

Furthermore, the energy management controller is a single chip microcomputer for the whole vehicle controller, the input signal includes real-time vehicle speed information, and the output signal includes the optimal power distribution of the fuel cell engine and the power battery.

The present invention also provides a vehicle using the method or system as described above.

Compared with the prior art, the present invention has the following beneficial effects:

The present invention manages the power demand of the fuel cell hybrid power system and allocates the output power of the fuel cell engine and the power battery based on the future power demand information of the whole vehicle, thereby ensuring that the equivalent hydrogen consumption of the whole vehicle is minimized during driving and maintaining stable and efficient operation of the power system.

The future vehicle power demand information is calculated based on the exponential smoothing method-Markov correction model and the vehicle dynamics model using the historical vehicle speed information collected in real time as a parameter, and is used as the predicted interference input of the energy management strategy.

The energy management strategy based on model-disturbance dual predictive control is used to distribute the power of the fuel cell engine and the power battery. The model-disturbance dual predictive control method is a control method that can effectively handle multi-variable and constrained problems, and has strong anti-disturbance ability, strong robustness and stability, can achieve advance control management, and can guarantee the economy and durability of the fuel cell hybrid system to a limited extent.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the overall structure of a fuel cell hybrid power system based on an MPC5604B microcontroller of the present invention;

FIG2 is a schematic diagram of the principle of the fuel cell hybrid power system controller of the present invention;

FIG3 is a schematic diagram of vehicle speed prediction of a fuel cell hybrid power system according to the present invention;

FIG4 is a schematic diagram of power distribution of a fuel cell hybrid power system according to the present invention;

FIG5 is a schematic diagram of the charge state of a power battery of a fuel cell hybrid power system of the present invention;

FIG. 6 is a schematic diagram of the cumulative equivalent hydrogen consumption of the fuel cell hybrid power system of the present invention.

Marking Description:

1-Vehicle speed sensor; 2-Vehicle speed predictor; 3-Energy management controller; 4-MPC5604B microcontroller; 5-Power battery; 6-DC/DC converter; 7-Fuel cell engine.

DETAILED DESCRIPTION

The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings.

As shown in FIG1 , the fuel cell vehicle model-disturbance dual predictive control energy management system proposed in the present invention includes a vehicle speed sensor 1, a vehicle speed predictor 2, an energy management controller 3, an MPC5604B microcontroller 4, a power battery 5, a DC/DC converter 6 and a fuel cell engine 7. The overall structural schematic diagram of the fuel cell hybrid power system based on the MPC5604B microcontroller is shown in FIG1 .

The DC/DC converter converts the voltage at the output of the fuel cell engine and connects it to the DC bus in parallel with the power battery;

The vehicle speed predictor collects vehicle speed information in real time and outputs predicted future vehicle speed information;

The energy management controller calculates the energy distribution between the fuel cell engine and the power battery based on the predicted future vehicle speed information, and transmits the power demand to the fuel cell engine and the power battery.

In this embodiment, a fuel cell vehicle power system energy management system based on the model-disturbance dual predictive control of the MPC5604B microcontroller is designed, please refer to Figure 1, including a vehicle speed sensor, a vehicle speed predictor, an energy management controller, a DC/DC converter, a fuel cell engine and a power battery connected in sequence. The whole vehicle device adopts the MPC5604B microcontroller. The MPC5604B in the control system downloads the code generated in the third-party CodeWarrior compilation environment to the MPC5604B microcontroller by using the standard JTAG simulation debugging interface PC[0].

In this embodiment, MPC5604B is connected to the vehicle speed sensor through GPIO ports PA[0]-PA[2] to read the real-time vehicle speed. After the internal optimization management algorithm, the control logic is output through eMIOS ports PB[11] and PB[12] and connected to the DC/DC converter to control the output power of the fuel cell engine. The power distribution signal of the power battery is output through eMIOS ports PB[13] and PB[14].

In this embodiment, the energy management controller based on MPC5604B takes the measured data of the vehicle speed sensor as input, and outputs a control signal through a model-disturbance dual predictive control algorithm, so that the fuel cell engine and the power battery can output at the optimal power, ensuring the system's goal of low hydrogen consumption.

The present invention is directed to the above-mentioned fuel cell hybrid power system, and distributes the output power of the fuel cell engine and the power battery under different power level requirements of the power system, realizes energy management control based on minimum equivalent hydrogen consumption, and maintains efficient and stable operation of the power system.

The present invention relates to a vehicle speed predictor and an energy management controller, and a schematic diagram of the power system controller is shown in FIG2 .

The vehicle speed predictor collects the vehicle speed information in real time through sensors, and calculates the future vehicle speed information based on the exponential smoothing-Markov correction model. The exponential smoothing-Markov correction model used is built offline through historical data.

Specifically, considering the characteristics of vehicle speed change, a third-order exponential smoothing model is constructed, as shown in formula (1):

Where S _t is the smoothing value at time t, α is the smoothing factor, and x is the vehicle speed sequence value at time _t .

Build a vehicle speed prediction model based on exponential smoothing method, as shown in formula (2):

in

According to the historical vehicle speed information, the vehicle speed prediction sequence based on the exponential smoothing method can be obtained. By comparing with the historical vehicle speed information, the prediction error of the exponential smoothing prediction model is calculated. The vehicle speed prediction error sequence can be regarded as a discrete Markov chain, and a Markov correction model is established based on it. The transfer probability matrix of the vehicle speed prediction error is shown in formula (3):

in

Where _Nij is the number of times the speed prediction error is transferred from state i to state j. The Markov correction model can predict and correct the speed prediction error of the exponential smoothing method.

The vehicle speed predictor uses the historical vehicle speed information collected by the sensor to calculate the future vehicle speed information based on the exponential smoothing method-Markov correction model. The obtained vehicle speed prediction correction result is shown in Figure 3.

The energy management controller combines the future vehicle speed information output by the vehicle speed predictor to calculate the vehicle power demand forecast sequence, which is used as the disturbance quantity of the model predictive control strategy to participate in the prediction of the power system state variables, and calculates the optimal power allocation strategy based on the minimum equivalent hydrogen consumption objective function.

Specifically, combined with the future vehicle speed information, the vehicle power demand prediction sequence is calculated based on the vehicle dynamics model, as shown in formula (4):

Where η _tran is the vehicle transmission system efficiency, _CD is the air resistance coefficient, A is the vehicle frontal area, a is the acceleration, δ is the rotational mass conversion coefficient, _mv is the vehicle mass, _Cf is the rolling resistance coefficient, and v is the vehicle speed.

The calculated vehicle power demand sequence is used as the disturbance of the model predictive control strategy and input into the system prediction model to calculate the predicted value of the power system state variable. The linear incremental state space equation of the system is shown in formula (5):

ΔX(k+1)=A·ΔX(k)+B·ΔU(k)+D·Δd(k) (5)

Where X is the system state variable, which is set to the state of charge of the power battery; U is the system control variable, which is set to the output power of the fuel cell engine; d is the system disturbance, which is set to the power demand of the vehicle; A, B and D are the system state space coefficient matrices.

The system prediction model of the model-disturbance dual prediction control strategy based on the predicted disturbance quantity is shown in formula (6):

X ^p (k+1|k)＝A ^p ΔX(k)+B ^p ΔU(k)+D ^p Δd(k)+X ^s (k) (6)

Wherein, A ^P , B ^P and D ^P are the prediction model coefficient matrices calculated from the state space coefficient matrix. ^{X s} (k) is the system state value collected in real time at time k. Feedback correction is performed based on the actual state value of the system, and the system response increment at future moments is calculated through the model-disturbance dual prediction model, which can better deal with the impact caused by the uncertainty factors of system nonlinearity and model mismatch. The model predictive control strategy establishes the objective function based on the minimum equivalent hydrogen consumption, as shown in formula (7):

in

为混合动力系统的等效氢气消耗质量流量，

为燃料电池发动机消耗的氢气质量流量，

为动力电池等效氢气消耗质量流量，κ为动力电池荷电状态平衡系数，p为模型预测控制预测时域，P_fc为燃料电池发动机输出功率，η_fc为燃料电池发动机效率，LHV_H2为氢气低热值，P_bat为动力电池输出功率，γ为动力电池充放电效率系数。In the formula,

is the equivalent hydrogen consumption mass flow rate of the hybrid system,

is the mass flow rate of hydrogen consumed by the fuel cell engine,

is the equivalent hydrogen consumption mass flow of the power battery, κ is the charge state balance coefficient of the power battery, p is the prediction time domain of the model predictive control, P _fc is the fuel cell engine output power, η _fc is the fuel cell engine efficiency, LHV _H2 is the lower heating value of hydrogen, P _bat is the power battery output power, and γ is the power battery charge and discharge efficiency coefficient.

Model-interference dual predictive control adopts a rolling finite time domain optimization strategy. At the kth moment, the system prediction model based on the predicted interference predicts the future changes in the system state quantity within the set prediction time domain, updates the relevant parameters of the objective function, and solves the optimal power allocation result at the kth moment through the effective set method and applies it to the actual system. At the next moment, the prediction time domain rolls forward, and the optimal power allocation is recalculated based on the system state quantity and interference quantity at the k+1th moment obtained by feedback. Model-interference dual predictive control realizes the rolling optimization process by continuously solving the optimal power allocation at each moment.

The energy management controller outputs the optimal power allocation strategy to the fuel cell engine and the power battery to achieve efficient and stable operation of the vehicle power system. The schematic diagrams of the fuel cell engine output power and the power battery charge state based on the actual working conditions of the vehicle are shown in Figures 4 and 5. In addition, a dynamic programming energy management strategy is designed to minimize hydrogen consumption. It is used as an optimization benchmark to compare the energy management strategies of the two model predictive control. Figure 6 shows the cumulative equivalent hydrogen consumption of the hybrid system under this working condition. The management and control results of the energy management strategy based on model-disturbance dual predictive control and the traditional model predictive control energy management strategy are compared, reflecting the optimization effect and application potential of the energy management strategy based on model-disturbance dual predictive control.

The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions do not exceed the scope of the technical solution of the present invention, belong to the protection scope of the present invention.

Claims (8)

1. A fuel cell vehicle model-disturbance dual predictive control energy management method, characterized in that it includes the following steps: Step S1, using historical vehicle speed information to predict future vehicle speed, calculating the prediction error between historical vehicle speed information and future vehicle speed information, and establishing a vehicle speed error correction model to correct the predicted future vehicle speed information; the vehicle speed prediction method used to predict the future vehicle speed is the third-order exponential smoothing method, the input is the historical vehicle speed information, and the output is the predicted future vehicle speed information: Considering the characteristics of vehicle speed change, a third-order exponential smoothing model is constructed, as shown in formula (1):

In the formula, _St is the smoothing value at time t, α is the smoothing factor, and x is the vehicle speed sequence value at time _xt ; Build a vehicle speed prediction model based on exponential smoothing method, as shown in formula (2):

in

According to the historical vehicle speed information, the vehicle speed prediction sequence based on the exponential smoothing method is obtained; by comparing with the historical vehicle speed information, the prediction error of the exponential smoothing prediction model is calculated; the vehicle speed prediction error sequence is regarded as a discrete Markov chain, and the Markov correction model is established based on it; the speed prediction error transition probability matrix is shown in formula (3):

in

Where _Nij is the number of times the speed prediction error is transferred from state i to state j. The Markov correction model can predict and correct the speed prediction error of the exponential smoothing method. Step S2: input the future vehicle speed information into the energy management controller, and calculate the future vehicle power demand in combination with the vehicle dynamics model: Combined with the future vehicle speed information, the vehicle power demand prediction sequence is calculated based on the vehicle dynamics model, as shown in formula (4):

Where η _tran is the vehicle transmission system efficiency, _CD is the air resistance coefficient, A is the vehicle frontal area, a is the acceleration, δ is the rotation mass conversion coefficient, _mv is the vehicle mass, _Cf is the rolling resistance coefficient, and v is the vehicle speed; Step S3: According to the future vehicle power demand information, the system parameters including the state of charge of the power battery are predicted based on the power system linear prediction model, and an equivalent hydrogen consumption objective function is established. The optimal solution of the equivalent hydrogen consumption objective function is calculated by the optimization algorithm to obtain the optimal power distribution between the fuel cell engine and the power battery: The calculated vehicle power demand sequence is used as the disturbance of the model predictive control strategy and input into the system linear prediction model to calculate the predicted value of the power system state variable; the system linear increment state space equation is shown in formula (5): ΔX(k+1)=A·ΔX(k)+B·ΔU(k)+D·Δd(k)(5) Where X is the system state variable, which is set to the state of charge of the power battery; U is the system control variable, which is set to the output power of the fuel cell engine; d is the system disturbance, which is set to the power demand of the vehicle; A, B and D are the system state space coefficient matrices; The system linear prediction model of the model-disturbance dual prediction control strategy based on the predicted disturbance is shown in formula (6): X ^p (k+1|k)＝A ^p ΔX(k)+B ^p ΔU(k)+D ^p Δd(k)+X ^s (k)(6) Wherein, A ^P , B ^P and D ^P are the system linear prediction model coefficient matrices calculated from the state space coefficient matrix; X ^s (k) is the system state value acquired in real time at time k; feedback correction is performed based on the actual system state value, and the system response increment at future moments is calculated through the model-disturbance dual prediction model. The model prediction control strategy establishes the objective function based on the minimum equivalent hydrogen consumption, as shown in formula (7):

in

In the formula,

is the equivalent hydrogen consumption mass flow rate of the hybrid system,

is the mass flow rate of hydrogen consumed by the fuel cell engine,

is the equivalent hydrogen consumption mass flow of the power battery, κ is the charge state balance coefficient of the power battery, p is the prediction time domain of the model predictive control, P _fc is the fuel cell engine output power, η _fc is the fuel cell engine efficiency, LHV _H2 is the lower heating value of hydrogen, P _bat is the power battery output power, and γ is the power battery charge and discharge efficiency coefficient. 2. The fuel cell vehicle model-disturbance dual predictive control energy management method according to claim 1 is characterized in that, in step S3, the optimization algorithm for calculating the optimal solution of the equivalent hydrogen consumption objective function is an effective set algorithm. 3. According to claim 1, the fuel cell vehicle model-interference dual predictive control energy management method is characterized in that the method realizes fuel cell vehicle power system energy management through model-interference dual prediction, wherein the model-interference dual prediction refers to state quantity prediction and interference prediction respectively, and important system parameters including the power battery state of charge are predicted by establishing a power system linear prediction model, and interference prediction refers to predicting the future vehicle speed and calculating the future load disturbance information in combination with the whole vehicle dynamics model. 4. A fuel cell vehicle model-disturbance dual predictive control energy management system, characterized in that the management system includes: a vehicle speed sensor, a vehicle speed predictor, an energy management controller, a DC/DC converter, a fuel cell engine, and a power battery; The vehicle speed sensor is used to detect current vehicle speed information; The fuel cell engine is connected in series with the DC/DC converter, and then connected in parallel with the power battery on the DC bus; The vehicle speed predictor inputs historical vehicle speed information and outputs predicted future vehicle speed information; The energy management controller inputs predicted future vehicle speed information and uses model-disturbance dual prediction control to output energy distribution between the fuel cell engine and the power battery based on minimum equivalent hydrogen consumption; The model-disturbance dual predictive control is to obtain future load disturbance information in advance by predicting and correcting future vehicle speed information, thereby improving the prediction accuracy of model predictive control; The vehicle speed predictor inputs historical vehicle speed information and outputs predicted future vehicle speed information in the following manner: Taking into account the characteristics of vehicle speed change, a third-order exponential smoothing model is constructed, as shown in formula (1):

In the formula, _St is the smoothing value at time t, α is the smoothing factor, and x is the vehicle speed sequence value at time _xt ; Build a vehicle speed prediction model based on exponential smoothing method, as shown in formula (2):

in

According to the historical vehicle speed information, the vehicle speed prediction sequence based on the exponential smoothing method is obtained; by comparing with the historical vehicle speed information, the prediction error of the exponential smoothing prediction model is calculated; the vehicle speed prediction error sequence is regarded as a discrete Markov chain, and the Markov correction model is established based on it; the speed prediction error transition probability matrix is shown in formula (3):

in

Where _Nij is the number of times the speed prediction error is transferred from state i to state j. The Markov correction model can predict and correct the speed prediction error of the exponential smoothing method. The energy management controller inputs the predicted future vehicle speed information and uses the model-disturbance dual prediction control to output the energy distribution between the fuel cell engine and the power battery based on the minimum equivalent hydrogen consumption in the following manner: 1) Combined with the future vehicle speed information, the vehicle power demand prediction sequence is calculated based on the vehicle dynamics model, as shown in formula (4):

Where η _tran is the vehicle transmission system efficiency, _CD is the air resistance coefficient, A is the vehicle frontal area, a is the acceleration, δ is the rotation mass conversion coefficient, _mv is the vehicle mass, _Cf is the rolling resistance coefficient, and v is the vehicle speed; 2) According to the future vehicle power demand information, the system parameters including the state of charge of the power battery are predicted based on the power system linear prediction model, and the equivalent hydrogen consumption objective function is established. The optimal solution of the equivalent hydrogen consumption objective function is calculated through the optimization algorithm to obtain the optimal power distribution of the fuel cell engine and the power battery: The calculated vehicle power demand sequence is used as the disturbance of the model predictive control strategy and input into the system linear prediction model to calculate the predicted value of the power system state variable; the system linear increment state space equation is shown in formula (5): ΔX(k+1)=A·ΔX(k)+B·ΔU(k)+D·Δd(k)(5) Where X is the system state variable, which is set to the state of charge of the power battery; U is the system control variable, which is set to the output power of the fuel cell engine; d is the system disturbance, which is set to the power demand of the vehicle; A, B and D are the system state space coefficient matrices; The system linear prediction model of the model-disturbance dual prediction control strategy based on the predicted disturbance is shown in formula (6): X ^p (k+1|k)＝A ^p ΔX(k)+B ^p ΔU(k)+D ^p Δd(k)+X ^s (k)(6) Wherein, A ^P , B ^P and D ^P are the system linear prediction model coefficient matrices calculated from the state space coefficient matrix; X ^s (k) is the system state value acquired in real time at time k; feedback correction is performed based on the actual system state value, and the system response increment at future moments is calculated through the model-disturbance dual prediction model. The model prediction control strategy establishes the objective function based on the minimum equivalent hydrogen consumption, as shown in formula (7):

in

In the formula,

is the equivalent hydrogen consumption mass flow rate of the hybrid system,

is the mass flow rate of hydrogen consumed by the fuel cell engine,

is the equivalent hydrogen consumption mass flow of the power battery, κ is the charge state balance coefficient of the power battery, p is the prediction time domain of the model predictive control, P _fc is the fuel cell engine output power, η _fc is the fuel cell engine efficiency, LHV _H2 is the lower heating value of hydrogen, P _bat is the power battery output power, and γ is the power battery charge and discharge efficiency coefficient. 5. The fuel cell vehicle model-disturbance dual predictive control energy management system according to claim 4 is characterized in that the vehicle speed predictor predicts future vehicle speed information based on an exponential smoothing method-Markov correction model. 6. According to claim 5, the fuel cell vehicle model-interference dual predictive control energy management system is characterized in that the exponential smoothing method-Markov correction model is based on historical vehicle speed information, combined with the third-order exponential smoothing method to calculate the future vehicle speed, and calculate the prediction error between the historical vehicle speed information and the future vehicle speed information, and accordingly establishes a Markov model vehicle speed error correction model to correct the vehicle speed information predicted based on the exponential smoothing method. 7. The fuel cell vehicle model-disturbance dual predictive control energy management system according to claim 4 is characterized in that the energy management controller adopts a model-disturbance dual predictive control strategy, combines future power demand information, predicts important system parameters including the state of charge of the power battery based on the power system linear prediction model, establishes an equivalent hydrogen consumption objective function, calculates the optimal solution of the objective function through the effective set algorithm, and obtains the optimal power allocation between the fuel cell engine and the power battery. 8. A vehicle, characterized by using the method according to any one of claims 1 to 3 or the system according to any one of claims 4 to 7.

Images