Fuel cell hybrid electric vehicle power system component size optimization method considering carbon emission

Abstract

The invention provides a fuel cell hybrid electric vehicle power system component size optimization method considering carbon emission, and aims at the size configuration of a fuel cell and a power cell in a hybrid power system to design a related comprehensive optimization target of cell cost and equivalent carbon dioxide emission. Modeling of a fuel cell, a power cell, a vehicle instantaneous power demand and an energy management strategy is combined, and a multi-population grey wolf algorithm is used to solve a multi-target constraint optimization problem. The method provided by the invention can effectively balance the contradiction among the power performance, carbon emission and production cost, and can more accurately adjust the size configuration of the fuel cell and the power cell component through the optimization algorithm, thereby reducing equivalent carbon dioxide emission and production cost to the greatest extent while meeting the performance requirements of the power system, and improving the reliability of the power system. And large-scale popularization and application of the fuel cell hybrid electric vehicle in the future market are facilitated.

Description

Technical Field

[0001] The invention belongs to the technical field of fuel cell hybrid electric vehicle power system design, and in particular relates to a method for optimizing the size of fuel cell hybrid electric vehicle power system components taking carbon emissions into consideration. Background Art

[0002] At present, the traditional fuel cell hybrid system design method usually focuses on the optimization of power system performance, but does not take carbon emissions and production costs into consideration. This single optimization goal has brought limitations in practical applications, making the economic and environmental benefits of fuel cell hybrid vehicles in practical applications unsatisfactory. Although fuel cell hybrid vehicles use hydrogen, a clean energy source, they cannot achieve true zero emissions and zero pollution. Therefore, how to further reduce the production cost and carbon emissions of the fuel cell system while ensuring the vehicle's power performance is still a technical problem that needs to be solved in this field. In the design process, a more comprehensive optimization method needs to be developed to ensure the excellent performance of the vehicle's power performance and effectively reduce production costs and carbon emissions, so as to promote the widespread application of fuel cell hybrid vehicles in the market. Summary of the invention

[0003] In view of this, in view of the technical problems existing in the art, the present invention provides a method for optimizing the size of power system components of a fuel cell hybrid electric vehicle considering carbon emissions, which specifically includes the following steps:

[0004] Step 1: Establish the following models for the fuel cell hybrid system: the fuel cell model and the power battery model are used to describe the fuel cell hydrogen consumption and the power battery state of charge (SOC), as well as the vehicle instantaneous power demand model;

[0005] Step 2: Establish the energy management strategy of the vehicle based on the idea of ​​model predictive control, taking SOC as the state quantity and the instantaneous output power of the fuel cell and the power battery as the decision quantity;

[0006] Step 3: Considering the use and replacement costs of fuel cells and power batteries, establish the component cost function of the fuel cell hybrid system; considering the carbon dioxide emissions related to the size of fuel cells and power batteries, establish the equivalent carbon dioxide emissions function of the fuel cell hybrid system; with the goal of minimizing the combination of these two functions, establish a multi-objective constrained optimization problem for solving the optimal size configuration of fuel cells and power batteries;

[0007] Step 4: Using the models established in step 1, the energy management strategy established in step 2, and the vehicle cycle operating data, the multi-objective constrained optimization problem established in step 3 is solved by executing a multi-population grey wolf algorithm, and finally the optimal size configuration result of the fuel cell and the power battery is obtained.

[0008] Furthermore, the fuel cell model established in step 1 specifically adopts the following form:

[0009]

[0010] in, is the hydrogen consumption of the fuel cell; LHV is the lower heating value of hydrogen; η fc is the efficiency of the fuel cell, which can be expressed as the following fitting form:

[0011] η fc =0.2716; if LR<0.05

[0012] η fc =0.9033LR 5 -2.996LR 4 +3.6503LR 3 -2.0704LR 2

[0013] +0.4623LR+0.3747; if LR≥0.05

[0014] in, is the load ratio, Indicates the rated power of the fuel cell;

[0015] The power battery model is specifically in the following form:

[0016]

[0017] Among them, SOC(t) is the instantaneous state of charge, t is time, Δt is the sampling time, η batt is the power battery charge / discharge efficiency, P batt (t) is the instantaneous output power of the power battery, is the rated capacity of the power battery;

[0018] The vehicle instantaneous power demand model is specifically in the following form:

[0019]

[0020] Among them, P load is the power demand of the fuel cell hybrid vehicle; η tran is the transmission efficiency; m is the vehicle mass; v(t) is the vehicle speed; is the vehicle acceleration; C d is the air resistance coefficient; ρ is the air density; A is the frontal area of ​​the vehicle; θ is the road slope; g is the acceleration of gravity; f is the rolling resistance coefficient;

[0021] The output power of the fuel cell and the power battery satisfies the following relationship:

[0022] P batt (t) = P load (t)-P fc (t)

[0023] Among them, P batt (t) is the instantaneous output power of the power battery, P batt (t)>0 means the power battery is in discharge state, P batt (t)<0 means the power battery is in charging state; P fc (t) is the instantaneous output power of the fuel cell.

[0024] Furthermore, the energy management strategy established in step 2 specifically takes the following form:

[0025]

[0026] Among them, SOC max Indicates the maximum value of SOC; SOC min Indicates the minimum value of SOC; SOC initial Indicates the starting SOC; SOC final Indicates termination of SOC; P batt,max Indicates the maximum output power of the power battery; P batt,min Indicates the minimum output power of the power battery; P fc,max Indicates the maximum output power of the fuel cell; P fc,min represents the minimum output power of the fuel cell; N is the prediction time domain; ω1 and ω2 are the weight coefficients in the objective function of the energy management strategy.

[0027] Furthermore, the fuel cell hybrid power system component cost function established in step 3 specifically adopts the following form:

[0028] g(X)=Cost fc +Cost batt +Cost fc-rep +Cost batt-rep

[0029] Where g(X) is the component cost of the hybrid system; That is, the size configuration of the fuel cell and power battery to be solved, Indicates the rated capacity of the power battery. Indicates the rated power of the fuel cell; Cost fc Cost of fuel cell batt Cost of power battery fc-rep and Cost batt-rep They are the replacement costs of the fuel cell and the power battery respectively; the cost of each part is determined by the following formula:

[0030]

[0031] Among them, c fc is the cost of the fuel cell per unit rated power; c batt is the cost of the power battery per unit rated capacity; k batt is the fixed cost of the power battery; FV fc and FV batt They represent the cost values ​​associated with the future replacement of fuel cells and power batteries, respectively; j represents the interest rate, which refers to the discount rate used to calculate the present value of future costs; M is the annual life of the fuel cell; L is the annual life of the power battery; M rep is the number of times the fuel cell is replaced; L rep It is the number of times the power battery is replaced.

[0032] Furthermore, the equivalent carbon dioxide emission function established in step 3 specifically adopts the following polynomial form:

[0033]

[0034] Where f(X) is the equivalent carbon dioxide emissions; a1~a6 are fitting parameters.

[0035] Furthermore, in step 3, the multi-objective constrained optimization problem is established. First, the component cost function g(X) and the equivalent carbon dioxide emission f(X) are normalized to the interval [0,1], and then and Finally, we get the following multi-objective constrained optimization problem:

[0036]

[0037] in, represents the average required power; m1 and m2 represent the coefficients for weighing different optimization indicators.

[0038] Furthermore, in step 4, when executing the multi-group gray wolf algorithm, the wolf king state is specifically expressed by the following formula:

[0039] K α =min(α1,α2,…,α l )

[0040] Among them, Kα Indicates the wolf king; α j (j=1,…,l) represents the wolf with the best fitness in the jth subpopulation; l∈(1,s) represents the division of the entire gray wolf population into l subpopulations, which is a positive integer, and s represents the number of gray wolf individuals in the entire population;

[0041] The position of the gray wolf is updated using the following rules guided by the wolf king in reverse:

[0042]

[0043] in, is the speed of the gray wolf with the lowest fitness in the jth population, which determines the direction and amplitude of the gray wolf's movement in the search space. represents the specific position of the gray wolf in the current search iteration; w is the learning coefficient, which is used to adjust the gray wolf's reliance on past experience when updating its position; c1 and c2 are two acceleration coefficients, which are used to control the speed at which the gray wolf moves to the target position; r1 and r2 are two random numbers, which take values ​​in the range [0,1];

[0044] The following random recombination threshold is set to trigger random recombination of the subpopulation with the lowest fitness:

[0045] Q=λE

[0046] Among them, Q represents the random recombination threshold, which is used to determine when to trigger the random recombination operation; λ∈(0,1) represents the learning parameter, which is used to control the learning intensity of historical information in the algorithm; E is the total number of iterations.

[0047] The method of the present invention can utilize MATLAB / Simulink software to execute various modeling, solving and simulation processes.

[0048] The above-mentioned fuel cell hybrid electric vehicle power system component size optimization method considering carbon emissions provided by the present invention designs the relevant battery cost and equivalent carbon dioxide emission comprehensive optimization targets for the size configuration of the fuel cell and power battery in the hybrid system, combines the modeling of the fuel cell, power battery, vehicle instantaneous power demand and energy management strategy, and uses a multi-group grey wolf algorithm to solve the multi-objective constraint optimization problem. The method of the present invention can effectively balance the contradictions between power performance, carbon emissions and production costs. The size configuration of the fuel cell and power battery components can be adjusted more accurately through the optimization algorithm, so as to minimize the equivalent carbon dioxide emissions and production costs while meeting the performance requirements of the power system, and is also conducive to the large-scale promotion and application of fuel cell hybrid electric vehicles in the future market. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1is a graph showing the relationship between the size of hybrid powertrain components and equivalent CO2 emissions;

[0050] Figure 2 It is a flowchart of the multi-population grey wolf algorithm;

[0051] Figure 3 It is the characteristic curve of SC03 cycle condition;

[0052] Figure 4 is the configuration result of the optimal component size of the hybrid system, where Figure 4 (a) is the evolution curve of component size, Figure 4 (b) is the evolution curve of the fitness function;

[0053] Figure 5 It is the simulation result of the energy management strategy. DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] The fuel cell hybrid vehicle power system component size optimization method considering carbon emissions provided by the present invention can be built through the software Matlab / Simulink, the software version is Matlab R2022a, and the solver is ode3. The simulation step size is a fixed step size, and the step size is selected as 0.1s.

[0056] The method specifically comprises the following steps:

[0057] Step 1: Establish the following models for the fuel cell hybrid system: the fuel cell model and the power battery model are used to describe the fuel cell hydrogen consumption and the power battery state of charge (SOC), as well as the vehicle instantaneous power demand model;

[0058] Step 2: Establish the energy management strategy of the vehicle based on the idea of ​​model predictive control, taking SOC as the state quantity and the instantaneous output power of the fuel cell and the power battery as the decision quantity;

[0059] Step 3: Considering the use and replacement costs of fuel cells and power batteries, establish the component cost function of the fuel cell hybrid system; considering the carbon dioxide emissions related to the size of fuel cells and power batteries, establish the equivalent carbon dioxide emissions function of the fuel cell hybrid system; with the goal of minimizing the combination of these two functions, establish a multi-objective constrained optimization problem for solving the optimal size configuration of fuel cells and power batteries;

[0060] Step 4: Using the models established in step 1, the energy management strategy established in step 2, and the vehicle cycle operating data, the multi-objective constrained optimization problem established in step 3 is solved by executing a multi-population grey wolf algorithm, and finally the optimal size configuration result of the fuel cell and the power battery is obtained.

[0061] In a preferred embodiment of the present invention, the fuel cell model established in step 1 specifically adopts the following form:

[0062]

[0063] in, is the hydrogen consumption of the fuel cell; LHV is the lower heating value of hydrogen; η fc is the efficiency of the fuel cell, which can be expressed as the following fitting form:

[0064] η fc =0.2716; if LR<0.05

[0065] η fc =0.9033LR 5 -2.996LR 4 +3.6503LR 3 -2.0704LR 2

[0066] +0.4623LR+0.3747; if LR≥0.05

[0067] in, is the load ratio, Indicates the rated power of the fuel cell;

[0068] The power battery model is specifically in the following form:

[0069]

[0070] Among them, SOC(t) is the instantaneous state of charge, t is time, Δt is the sampling time, η batt is the power battery charge / discharge efficiency, P batt (t) is the instantaneous output power of the power battery, is the rated capacity of the power battery;

[0071] The vehicle instantaneous power demand model is specifically in the following form:

[0072]

[0073] Among them, P load is the power demand of the fuel cell hybrid vehicle; η tranis the transmission efficiency; m is the vehicle mass; v(t) is the vehicle speed; is the vehicle acceleration; C d is the air resistance coefficient; ρ is the air density; A is the frontal area of ​​the vehicle; θ is the road slope; g is the acceleration of gravity; f is the rolling resistance coefficient;

[0074] The output power of the fuel cell and the power battery satisfies the following relationship:

[0075] P batt (t) = P load (t)-P fc (t)

[0076] Among them, P batt (t) is the instantaneous output power of the power battery, P batt (t)>0 means the power battery is in discharge state, P batt (t)<0 means the power battery is in charging state; P fc (t) is the instantaneous output power of the fuel cell.

[0077] The goal of the energy management strategy is to reasonably allocate the instantaneous output power of each power source in the hybrid system to ensure that the vehicle can meet the immediate power demand during driving. This strategy not only needs to balance the energy output between the power sources, but also needs to consider the overall efficiency of the system and the long-term energy utilization optimization. To this end, in a preferred embodiment of the present invention, the energy management strategy established in step 2 specifically adopts the following form:

[0078]

[0079] Among them, SOC max Indicates the maximum value of SOC; SOC min Indicates the minimum value of SOC; SOC initial Indicates the starting SOC; SOC final Indicates termination of SOC; P batt,max Indicates the maximum output power of the power battery; P batt,min Indicates the minimum output power of the power battery; P fc,max Indicates the maximum output power of the fuel cell; P fc,min represents the minimum output power of the fuel cell; N is the prediction time domain; ω1 and ω2 are the weight coefficients in the objective function of the energy management strategy.

[0080] In a preferred embodiment of the present invention, the fuel cell hybrid power system component cost function established in step 3 specifically adopts the following form:

[0081] g(X)=Cost fc +Cost batt +Costfc-rep +Cost batt-rep

[0082] Where g(X) is the component cost of the hybrid system; That is, the size configuration of the fuel cell and power battery to be solved, Indicates the rated capacity of the power battery. Indicates the rated power of the fuel cell; Cost fc Cost of fuel cell batt Cost of power battery fc-rep and Cost batt-rep They are the replacement costs of fuel cells and power batteries. Since fuel cells and power batteries may need to be replaced during the life cycle of the vehicle, this part of the cost is also an important part of the economic evaluation of the entire system. The replacement cost depends on the service life of the components, the speed of technological updates, and market price fluctuations. The cost of each part is determined by the following formula:

[0083]

[0084] Among them, c fc =47.67$ / kW reflects the cost of the fuel cell per unit rated power; c batt =651.2$ / kWh reflects the cost of the power battery per unit rated capacity; k batt =680$ is the fixed cost of the power battery; FV fc and FV batt They represent the cost values ​​associated with the future replacement of fuel cells and power batteries respectively; j = 7% represents the interest rate, which refers to the discount rate used to calculate the present value of future costs; M = 4 is the annual life of the fuel cell; L = 10 is the annual life of the power battery; M rep =3 is the number of times the fuel cell is replaced; L rep =1 is the number of times the power battery is replaced.

[0085] In a preferred embodiment of the present invention, the equivalent carbon dioxide emission function established in step 3 specifically adopts the following polynomial form:

[0086]

[0087] Where, f(X) is the equivalent carbon dioxide emissions;

[0088] a1=-2.83×10 4 ,a2=1.57×10 4 , a3=3287, a4=-77.86, a5=37.87, a6=-15.24 are all fitting parameters. The fitting curve is as follows Figure 1 shown.

[0089] In a preferred embodiment of the present invention, in step 3, the multi-objective constrained optimization problem is established. First, the component cost function g(X) and the equivalent carbon dioxide emission f(X) are normalized to the interval [0,1] to obtain and Finally, we get the following multi-objective constrained optimization problem:

[0090]

[0091] in, represents the average required power; m1 and m2 represent the coefficients for weighing different optimization indicators.

[0092] In a preferred embodiment of the present invention, when executing the multi-group gray wolf algorithm in step 4, the following aspects are considered to enhance the global search capability and local optimization capability compared with the traditional gray wolf algorithm:

[0093] (1) Increase population diversity

[0094] In order to overcome the problem that the basic gray wolf optimization algorithm is prone to fall into the local optimal solution during the search process and further enhance the diversity of the population, the present invention improves the algorithm. Specifically, in the initialization stage of the algorithm, the gray wolf population is divided into l sub-populations. This division helps to form multiple different search paths in the search space, thereby increasing the possibility of the algorithm jumping out of the local optimum. Each sub-population still adjusts its position according to the prey search process of the basic gray wolf algorithm, that is, by simulating hunting behavior, the position of each gray wolf is continuously updated. However, after each round of position update, the sub-populations need to share their search results. This sharing mechanism can enable all sub-populations to use the search information of other sub-populations, thereby enhancing the synergy and information exchange of the global search. After all sub-populations complete the position update and share the results, a wolf with the best fitness is selected from the leading wolves of each sub-population. This wolf will be set as the wolf king of the entire population. As the most advantageous individual in the population, the position of the wolf king represents the optimal solution found by the current algorithm. Therefore, the wolf king state is specifically represented by the following formula:

[0095] K α =min(α1,α2,…,α l )

[0096] Among them, K α Indicates the wolf king; α j (j=1,…,l) represents the wolf with the best fitness in the jth subpopulation; l∈(1,s) represents the division of the entire gray wolf population into l subpopulations, which is a positive integer, and s represents the number of gray wolf individuals in the entire population;

[0097] (2) Wolf King Reverse Guidance

[0098] In order to improve the convergence speed of the basic gray wolf algorithm, the present invention adjusts the gray wolf with the lowest fitness in each subpopulation and introduces the reverse guidance strategy of the wolf king. This strategy is inspired by the particle swarm optimization algorithm, in which the movement of particles is not only affected by their own historical best position, but also guided by the global best position. Specifically, in the multi-population gray wolf algorithm, the wolf king is defined as the global optimal solution of the entire population, that is, the optimal position found by the current algorithm. The optimal gray wolf in each subpopulation represents the local optimal solution, reflecting the best performance of the subpopulation in its search area. In order to prevent the subpopulation from falling into the local optimal solution and improve the overall optimization effect, the present invention pays special attention to the gray wolf with the lowest fitness in each subpopulation, and updates its position through the reverse guidance strategy of the wolf king. The specific update rule is to use the position of the wolf king as a benchmark, and reversely adjust the positions of these gray wolves to make them closer to the global optimal solution. The position of the gray wolf is updated using the following wolf king reverse guidance rules:

[0099]

[0100] in, is the speed of the gray wolf with the lowest fitness in the jth population, which determines the direction and amplitude of the gray wolf's movement in the search space. represents the specific position of the gray wolf in the current search iteration; w is the learning coefficient, which is used to adjust the gray wolf's reliance on past experience when updating its position; c1 and c2 are two acceleration coefficients, which are used to control the speed at which the gray wolf moves to the target position; r1 and r2 are two random numbers, which take values ​​in the range [0,1];

[0101] (3) Random recombination threshold

[0102] In order to effectively prevent the basic grey wolf algorithm from falling into the local optimum, the present invention introduces a random recombination threshold as a strategy, and sets the following random recombination threshold to trigger the random recombination of the sub-population with the lowest fitness:

[0103] Q=λE

[0104] Among them, Q represents the random recombination threshold, which is used to determine when to trigger the random recombination operation; λ∈(0,1) represents the learning parameter, which is used to control the learning intensity of historical information in the algorithm; E is the total number of iterations. In each iteration, the fitness value of the wolf king selected in the current iteration is monitored. A marker variable q is set to record the change of the fitness value. If it is detected that the fitness value of the wolf king in the current iteration has changed compared with the previous iteration, the marker variable will be reset. This indicates that the current iteration has made progress and the position of the wolf king has improved during the optimization process. If the fitness value of the wolf king has not changed in a certain iteration, it means that the current optimization progress has stagnated. At this time, the marker variable q is increased and checked whether it exceeds the random recombination threshold Q. When the marker variable exceeds the threshold, that is, q>Q, it means that the optimization progress in the current iteration cycle is insufficient and may be trapped in the local optimum. At this time, the sub-population with the lowest fitness will be randomly reorganized, that is, the position of some gray wolves in these sub-populations will be reinitialized or their speed will be adjusted. This operation is intended to introduce new search paths to help the algorithm jump out of the local optimal solution and enhance the global search capability.

[0105] The above three performance enhancement methods effectively improve the global search ability and local optimization ability of the basic gray wolf optimization algorithm. Specifically, these methods optimize the performance of the algorithm in complex search space by introducing strategies such as multi-population mechanism, wolf king's reverse guidance and random recombination threshold, helping the algorithm to better explore the global optimal solution and avoid falling into the local optimal solution. The detailed flow chart of the designed multi-population gray wolf algorithm is shown in the figure. Figure 2 shown.

[0106] The instantaneous power demand of a vehicle fluctuates with the speed. This fluctuation is particularly significant in urban driving conditions, mainly in low-speed driving, frequent start-stop operations, and various acceleration scenarios. In order to accurately simulate this complex driving environment, the example of the present invention uses the SC03 cycle in the simulation, which includes low-speed driving, frequent start-stop operations, and various acceleration and deceleration scenarios commonly seen in actual driving. These factors all affect the power demand of the vehicle. Figure 3 As shown in the figure, the characteristics of the SC03 cycle can provide real power demand fluctuations. In this simulation framework, the goal of the present invention is to use the developed multi-population gray wolf optimization algorithm to solve the multi-objective constrained optimization problem. Through this optimization algorithm, the optimal size configuration of each component in the hybrid system can be systematically analyzed and determined, including the optimal rated power of the fuel cell and the optimal rated capacity of the power battery.

[0107] Figure 4(a) is a diagram showing the evolution of the optimal component size of the hybrid system, which reflects how the component size gradually approaches the optimal value as the algorithm iterates during the optimization process. The figure details the changing trends of the component sizes at different iteration stages, providing an intuitive understanding of the optimization process. Figure 4 (b) shows the evolution curve of the fitness function, which shows how the fitness value changes with the number of iterations. By observing the changes in the fitness function, the convergence speed and effect of the algorithm can be evaluated. The optimal component size is finally determined to be In this configuration, the component cost of the hybrid system is $43,556; the equivalent carbon dioxide emissions over the entire life cycle are 624,460 kg. Under the selected component size, the system's economy and environmental performance are optimally balanced.

[0108] Figure 5 The upper part shows the power distribution curve determined by the energy management strategy under optimal component sizing. This curve depicts how the system distributes power to meet the dynamic needs of the vehicle under different driving conditions. Figure 5 The lower part shows the change curve of the power battery SOC, showing the change of SOC during the entire driving cycle. Figure 5 It can be seen that the configuration of the fuel cell and the power battery can effectively meet the power demand of the vehicle during driving, and the SOC is always maintained within the predetermined range, which shows that the system performs well in terms of energy management and utilization efficiency.

[0109] It should be understood that the size of the serial numbers of the steps in the embodiment of the present invention does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0110] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the size of power system components of a fuel cell hybrid electric vehicle considering carbon emissions, characterized in that: The specific steps include: Step 1: Establish the following models for the fuel cell hybrid system: the fuel cell model and the power battery model are used to describe the fuel cell hydrogen consumption and the power battery charge state, respectively, as well as the vehicle instantaneous power demand model; Step 2: Establish the energy management strategy of the vehicle based on the idea of ​​model predictive control, taking the state of charge as the state quantity and the instantaneous output power of the fuel cell and the power battery as the decision quantity; Step 3: Considering the use and replacement costs of fuel cells and power batteries, establish the component cost function of the fuel cell hybrid system; considering the carbon dioxide emissions related to the size of fuel cells and power batteries, establish the equivalent carbon dioxide emissions function of the fuel cell hybrid system; with the goal of minimizing the combination of these two functions, establish a multi-objective constrained optimization problem for solving the optimal size configuration of fuel cells and power batteries; Step 4: Using the models established in step 1, the energy management strategy established in step 2, and the vehicle cycle operating data, the multi-objective constrained optimization problem established in step 3 is solved by executing a multi-population grey wolf algorithm, and finally the optimal size configuration result of the fuel cell and the power battery is obtained.

2. The method according to claim 1, characterized in that: The fuel cell model established in step 1 specifically takes the following form: in, is the hydrogen consumption of the fuel cell; LHV is the lower heating value of hydrogen; η fc is the efficiency of the fuel cell, which can be expressed as the following fitting form: η fc =0.2716;if LR<0.05 n fc =0.9033LR 5 -2.996LR 4 +3.6503LR 3 -2.0704LR 2 +0.4623LR+0.3747;if LR≥0.05 in, is the load ratio, Indicates the rated power of the fuel cell; The power battery model is specifically in the following form: Among them, SOC(t) is the instantaneous state of charge, t is time, Δt is the sampling time, η batt is the power battery charge / discharge efficiency, P batt (t) is the instantaneous output power of the power battery, is the rated capacity of the power battery; The vehicle instantaneous power demand model is specifically in the following form: Among them, P load is the power demand of the fuel cell hybrid vehicle; η tran is the transmission efficiency; m is the vehicle mass; v(t) is the vehicle speed; is the vehicle acceleration; C d is the air resistance coefficient; ρ is the air density; A is the frontal area of ​​the vehicle; θ is the road slope; g is the acceleration of gravity; f is the rolling resistance coefficient; The output power of the fuel cell and the power battery satisfies the following relationship: P batt (t)=P load (t)-P fc (t) Among them, P batt (t) is the instantaneous output power of the power battery, P batt (t)>0 means the power battery is in discharge state, P batt (t)<0 means the power battery is in charging state; P fc (t) is the instantaneous output power of the fuel cell.

3. The method according to claim 2, characterized in that: The energy management strategy established in step 2 takes the following form: SOC min ≤SOC(t)≤SOC max P batt,min ≤P batt (t)≤P batt,max P fc,min ≤P fc (t)≤P fc,max P fc (t)=P load (t)-P batt (t) Among them, SOC max Indicates the maximum value of SOC; SOC min Indicates the minimum value of SOC; SOC initial Indicates the starting SOC; SOC final Indicates termination of SOC; P batt,max Indicates the maximum output power of the power battery; P batt,min Indicates the minimum output power of the power battery; P fc,max Indicates the maximum output power of the fuel cell; P fc,min represents the minimum output power of the fuel cell; N is the prediction time domain; ω1 and ω2 are the weight coefficients in the objective function of the energy management strategy.

4. The method according to claim 3, characterized in that: The fuel cell hybrid system component cost function established in step 3 is specifically in the following form: g(X)=Cost fc +Cost batt +Cost fc-rep +Cost batt-rep Where g(X) is the component cost of the hybrid system; That is, the size configuration of the fuel cell and power battery to be solved, Indicates the rated capacity of the power battery. Indicates the rated power of the fuel cell; Cost fc Cost of fuel cell batt Cost of power battery fc-rep and Cost batt-rep They are the replacement costs of the fuel cell and the power battery respectively; the cost of each part is determined by the following formula: Among them, c fc is the cost of the fuel cell per unit rated power; c batt is the cost of the power battery per unit rated capacity; k batt is the fixed cost of the power battery; FV fc and FV batt They represent the cost values ​​associated with the future replacement of fuel cells and power batteries, respectively; j represents the interest rate, which refers to the discount rate used to calculate the present value of future costs; M is the annual life of the fuel cell; L is the annual life of the power battery; M rep is the number of times the fuel cell is replaced; L rep It is the number of times the power battery is replaced.

5. The method according to claim 4, characterized in that: The equivalent carbon dioxide emission function established in step 3 specifically adopts the following polynomial form: Where f(X) is the equivalent carbon dioxide emissions; a1~a6 are fitting parameters.

6. The method according to claim 5, characterized in that: In step 3, the multi-objective constrained optimization problem is established. First, the component cost function g(X) and the equivalent carbon dioxide emission f(X) are normalized to the interval [0,1], and then and Finally, we get the following multi-objective constrained optimization problem: s.t.P load (t)=P fc (t)+P batt (t) SOC min ≤SOC(t)≤SOC max in, represents the average required power; m1 and m2 represent the coefficients for weighing different optimization indicators.

7. The method according to claim 1, characterized in that: Step 4: When executing the multi-group gray wolf algorithm, the wolf king state is specifically expressed by the following formula: K α =min(α1,α2,…,α l ) Among them, K α Indicates the wolf king; α j (j=1,…,l) represents the wolf with the best fitness in the jth subpopulation; l∈(1,s) represents the division of the entire gray wolf population into l subpopulations, which is a positive integer, and s represents the number of gray wolf individuals in the entire population; The position of the gray wolf is updated using the following rules guided by the wolf king in reverse: in, is the speed of the gray wolf with the lowest fitness in the jth population, which determines the direction and amplitude of the gray wolf's movement in the search space. represents the specific position of the gray wolf in the current search iteration; w is the learning coefficient, which is used to adjust the gray wolf's reliance on past experience when updating its position; c1 and c2 are two acceleration coefficients, which are used to control the speed at which the gray wolf moves to the target position; r1 and r2 are two random numbers, which take values ​​in the range [0,1]; The following random recombination threshold is set to trigger random recombination of the subpopulation with the lowest fitness: Q=λE Among them, Q represents the random recombination threshold, which is used to determine when to trigger the random recombination operation; λ∈(0,1) represents the learning parameter, which is used to control the learning intensity of historical information in the algorithm; E is the total number of iterations.

Images