Fuel cell vehicle energy management method, device, system, and storage medium

Abstract

The present invention discloses a fuel cell vehicle energy management method, device, system, and storage medium, comprising: Step S1: Using a long short-term memory neural network to predict short-term vehicle speed based on actual urban driving data; Step S2: Based on the short-term vehicle speed, a prediction domain power demand sequence is obtained using a vehicle longitudinal dynamics model; Step S3: Based on the prediction domain power demand sequence, a prediction domain optimization problem is constructed with fuel cell hydrogen consumption and life decay as optimization objectives; Step S4: Using an equilibrium optimizer (EO) algorithm to solve the prediction domain optimization problem and obtain an optimal control sequence. The technical solution of the present invention can improve the economic efficiency and real-time performance of the fuel cell vehicle energy management method, and has strong adaptability to actual urban operating conditions.

Description

Technical Field

[0001] The present invention belongs to the field of transportation technology, and in particular relates to a fuel cell vehicle energy management method and device, system, and storage medium. Background Art

[0002] Fuel cell vehicles (FCVs), with their zero emissions, long driving range, and fast refueling, have become a key development direction for new energy vehicles. However, in real urban conditions, FCVs have high operating costs, primarily due to hydrogen consumption and powertrain degradation. These costs are dependent on the performance of the FCV's energy management methods.

[0003] Currently, most fuel cell vehicle energy management methods are rule-based. These rules rely on the designer's engineering experience, making them difficult to adapt to the complex and changing urban traffic environment and exhibiting poor adaptability to operating conditions. As a type of optimization-based control method, energy management strategies incorporating the MPC framework offer robust operating conditions and are widely used in fuel cell vehicle energy management technology.

[0004] Traditional MPC energy management methods do not fully consider the fuel cell life attenuation cost. In addition, the optimization performance of the MPC solver affects the economy and real-time performance of the MPC energy management method. Traditional MPC solvers are mostly based on original or simplified dynamic programming algorithms, which have a large amount of solution computation and have potential for further improvement. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a fuel cell vehicle energy management method and device, system, and storage medium.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A fuel cell vehicle energy management method, comprising:

[0008] Step S1: predicting short-term vehicle speed using a long short-term memory neural network based on actual urban driving data;

[0009] Step S2: Obtain a prediction domain power demand sequence based on the short-term vehicle speed through the vehicle longitudinal dynamics model;

[0010] Step S3: Based on the prediction domain power demand sequence, the prediction domain optimization problem is constructed with the fuel cell hydrogen consumption and life attenuation as optimization targets;

[0011] Step S4: Use the balanced optimizer EO algorithm to solve the prediction domain optimization problem and obtain the optimal control sequence.

[0012] The present invention also provides a fuel cell vehicle energy management device, comprising:

[0013] The first processing module is used to predict short-term vehicle speed using a long short-term memory neural network based on actual urban driving data;

[0014] The second processing module is used to obtain a prediction domain power demand sequence based on the short-term vehicle speed through a vehicle longitudinal dynamics model;

[0015] The third processing module is used to construct a prediction domain optimization problem based on the prediction domain power demand sequence and taking the fuel cell hydrogen consumption and life attenuation as optimization targets;

[0016] The fourth processing module is used to solve the prediction domain optimization problem using the equilibrium optimizer EO algorithm to obtain the optimal control sequence.

[0017] The present invention also provides a fuel cell vehicle energy management system, characterized in that it includes: a memory and a processor, wherein the memory stores a computer program run by the processor, and the computer program executes the fuel cell vehicle energy management method when run by the processor.

[0018] The present invention also provides a storage medium on which a computer program is stored. When the computer program is run, the fuel cell vehicle energy management method is executed.

[0019] The present invention adopts an EO algorithm based on a balanced optimizer that is oriented towards urban operating conditions and takes fuel cell life into consideration to achieve energy management of fuel cell vehicles. It takes into account the impact of power changes on fuel cell life and slows down the life decay of the fuel cell. It can further search for the optimal control sequence under the conditions of a smaller population size and number of iterations, and the control method has good economy and real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0021] Figure 1 This is a flow chart of the energy management method for a fuel cell vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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 making creative efforts are within the scope of protection of the present invention.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Example 1:

[0025] like Figure 1 As shown, an embodiment of the present invention provides a fuel cell vehicle energy management method, comprising:

[0026] Step S1: predicting short-term vehicle speed using a long short-term memory neural network based on actual urban driving data;

[0027] Step S2: Obtain a prediction domain power demand sequence based on the short-term vehicle speed;

[0028] Step S3: Based on the prediction domain power demand sequence, the prediction domain optimization problem is constructed with the fuel cell hydrogen consumption and life attenuation as optimization targets;

[0029] Step S4: Use the balanced optimizer EO algorithm to solve the prediction domain optimization problem and obtain the optimal control sequence.

[0030] As an implementation method of the embodiment of the present invention, in step S1, the input variables of the vehicle speed prediction model are the vehicle speed at the current moment and the historical moment. The output variable of the speed prediction model is the prediction domain speed ;The vehicle speed prediction model is based on long short-term memory neural network;

[0031] As an implementation method of the present invention, in step S2, the future vehicle speed is predicted based on the historical vehicle speed, and the power demand sequence of the prediction domain power system is calculated according to equations (1), (2), (3), (4) and (5).

[0032] The longitudinal dynamics of the vehicle are as follows:

[0033] (1)

[0034] in, F t Indicates traction, m Indicates the vehicle mass, g represents the acceleration due to gravity, f represents the rolling resistance coefficient, αIndicates the road slope angle, C D represents the aerodynamic drag coefficient, A represents the frontal area of ​​the vehicle, r represents the air density, v Indicates instantaneous speed;

[0035] Wheel speed oh w and wheel torque T w Calculated from formula (2):

[0036] (2)

[0037] in, r Indicates the wheel radius;

[0038] Motor speed oh m and motor torque T m Calculated from formula (3):

[0039] (3)

[0040] in or fd Indicates the transmission efficiency of the transmission device, R fd Indicates the gear ratio;

[0041] Motor power P m and required power P dem Calculated by formula (4) and formula (5) respectively:

[0042] (4)

[0043] (5)

[0044] in, or m represents the motor efficiency, or DC / AC express DC / AC Converter efficiency.

[0045] As an implementation method of the present invention, in step S3, the optimization targets are selected as hydrogen consumption cost, fuel cell life degradation cost and prediction domain battery SOC Fluctuation, corresponding to equations (6)-(8) respectively;

[0046] (6)

[0047] (7)

[0048] (8)

[0049] in, represents the price of hydrogen, or FCS (.) represents the fuel cell system efficiency as a function of the fuel cell system output power, C FC Indicates the unit power price of the fuel cell stack, P FC_rated Indicates the stack power, α FC_slope represents the fuel cell life decay rate caused by power changes, U cell_limit Indicates the voltage attenuation threshold corresponding to the end of the fuel cell monomer service life;

[0050] The prediction domain cost function is expressed as formula (9)

[0051] (9)

[0052] The power sequence of the fuel cell system in the prediction domain is selected as the decision variable, and the decision variable of the prediction domain optimization problem is formula (10):

[0053] (10)

[0054] in, D is the dimension of decision variables;

[0055] The cost function of the optimization problem is formula (11):

[0056] (11)

[0057] in i 1 , i 2 is a constant;

[0058] The optimization problem constraints are as follows:

[0059] (12)

[0060] in, P FC_min , P FC_max Respectively represent the minimum and maximum output power of the fuel cell system, ΔP FC_min , D PFC_max They represent the power fluctuation limit of the fuel cell system, SOC min , SOC max Respectively represent batteries SOC Lower and upper limits;

[0061] As an implementation of an embodiment of the present invention, the equilibrium optimizer EO algorithm in step S4 is a meta-initial algorithm that evaluates the dynamic state and the equilibrium state based on the control volume mass balance model.

[0062] The initial population of the algorithm is randomly generated according to the population size and the number of time steps in the prediction domain, which is formula (13):

[0063] (13)

[0064] in, Represents an individual i The corresponding initial fuel cell system power sequence, C min and C max Respectively represent the minimum and maximum values ​​of the fuel cell system power at each time step in the prediction domain, r i represents a random number vector whose elements are in the range [0,1]. n Indicates the number of individuals in the population;

[0065] The power sequence of the randomly distributed fuel cell system is expressed as formula (14):

[0066] (14)

[0067] The fitness value of each fuel cell system power sequence is formula (15):

[0068] (15)

[0069] Balancing tank The four best fuel cell system power sequences among all current fuel cell system power sequences and the arithmetic average of the above four fuel cell system power sequences are given by formula (16):

[0070] (16)

[0071] in The arithmetic mean of the power sequences of the four fuel cell systems is represented by equation (17):

[0072] (17)

[0073] Exponential Vector It is used to balance local search and global search, which is formula (18):

[0074] (18)

[0075] in, a 1 represents the global search weight coefficient, sign(.) represents the symbolic function, , represents a random number vector whose elements are in the range [0,1]. t It represents the function related to the number of iterations of the prediction domain optimization problem, which is formula (19):

[0076] (19)

[0077] in, a 2 is a constant, Iter Indicates the current iteration number of the prediction domain optimization problem, Max_iter Represents the maximum number of iterations for the prediction domain optimization problem, generating a rate vector Formula (20):

[0078] (20)

[0079] in represents the generation rate control parameter, which is formula (21):

[0080] (twenty one)

[0081] The power sequence update model of the fuel cell system is formula (22):

[0082] (twenty two)

[0083] Example 2:

[0084] An embodiment of the present invention further provides a fuel cell vehicle energy management device, comprising:

[0085] The first processing module is used to predict short-term vehicle speed using a long short-term memory neural network based on actual urban driving data;

[0086] The second processing module is used to obtain a prediction domain power demand sequence based on the short-term vehicle speed through a vehicle longitudinal dynamics model;

[0087] The third processing module is used to construct a prediction domain optimization problem based on the prediction domain power demand sequence and taking the fuel cell hydrogen consumption and life attenuation as optimization targets;

[0088] The fourth processing module is used to solve the prediction domain optimization problem using the equilibrium optimizer EO algorithm to obtain the optimal control sequence.

[0089] Example 3:

[0090] The present invention also provides a fuel cell vehicle energy management system, characterized in that it includes: a memory and a processor, wherein the memory stores a computer program run by the processor, and the computer program executes the fuel cell vehicle energy management method when run by the processor.

[0091] Example 4:

[0092] The present invention also provides a storage medium on which a computer program is stored. When the computer program is run, the fuel cell vehicle energy management method is executed.

[0093] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A fuel cell vehicle energy management method, characterized in that: include: Step S1: predicting short-term vehicle speed using a long short-term memory neural network based on actual urban driving data; Step S2: Obtain a prediction domain power demand sequence based on the short-term vehicle speed through the vehicle longitudinal dynamics model; Step S3: Based on the prediction domain power demand sequence, the prediction domain optimization problem is constructed with the fuel cell hydrogen consumption and life attenuation as optimization targets; Step S4: using the equilibrium optimizer EO algorithm to solve the prediction domain optimization problem and obtain the optimal control sequence; In step S3, the prediction domain cost function with fuel cell hydrogen consumption and life attenuation as optimization targets is F Expressed as formula (9) F = C H2 + C FC_deg + L SOC (9) C H2 is the hydrogen consumption cost, C FC_deg is the fuel cell life degradation cost, L SOC For the prediction domain battery SOC fluctuation; The power sequence of the fuel cell system in the prediction domain is selected as the decision variable, and the decision variable of the prediction domain optimization problem is formula (10): （10） in, D is the dimension of decision variables; The cost function of the optimization problem is formula (11): （11） in, θ 1 , θ 2 is a constant; The optimization problem constraints are as follows: （12） in, P FC_min , P FC_max Respectively represent the minimum and maximum output power of the fuel cell system, ΔP FC_min , Δ P FC_max They represent the power fluctuation limit of the fuel cell system, SOC min , SOC max Respectively represent batteries SOC Lower and upper limits; In step S4, the equilibrium optimizer EO algorithm evaluates the dynamic state and equilibrium state based on the control volume mass balance model, specifically: The initial population is randomly generated according to the population size and the number of time steps in the prediction domain, which is formula (13): （13） in, Represents an individual i The corresponding initial fuel cell system power sequence, C min and C max Respectively represent the minimum and maximum values ​​of the fuel cell system power at each time step in the prediction domain, r i represents a random number vector whose elements are in the range [0,1]. n Indicates the number of individuals contained in the population; The randomly distributed fuel cell system power sequence C is expressed as formula (14): （14） Fitness value of each fuel cell system power sequence fit Formula (15): （15） Balancing tank The four best fuel cell system power sequences among all current fuel cell system power sequences and the arithmetic average of the above four fuel cell system power sequences are given by formula (16): （16） in, The arithmetic mean of the power sequences of the four fuel cell systems is represented by equation (17): （17） Exponential Vector It is used to balance local search and global search, which is formula (18): （18） in, a 1 represents the global search weight coefficient, sign(.) represents the symbolic function, , represents a random number vector whose elements are in the range [0,1]. t It represents the function related to the number of iterations of the prediction domain optimization problem, which is formula (19): （19） in, a 2 is a constant, Iter Indicates the current iteration number of the prediction domain optimization problem, Max_iter Represents the maximum number of iterations for the prediction domain optimization problem, generating a rate vector Formula (20): （20） in, represents the generation rate control parameter, which is formula (21): （21） The power sequence update model of the fuel cell system is formula (22): （22）。 2. A fuel cell vehicle energy management device for implementing the fuel cell vehicle energy management method according to claim 1, characterized in that: include: The first processing module is used to predict short-term vehicle speed using a long short-term memory neural network based on actual urban driving data; The second processing module is used to obtain a prediction domain power demand sequence based on the short-term vehicle speed through a vehicle longitudinal dynamics model; The third processing module is used to construct a prediction domain optimization problem based on the prediction domain power demand sequence and taking the fuel cell hydrogen consumption and life attenuation as optimization targets; The fourth processing module is used to solve the prediction domain optimization problem using the equilibrium optimizer EO algorithm to obtain the optimal control sequence.

3. A fuel cell vehicle energy management system, characterized in that: include: A memory and a processor, wherein the memory stores a computer program executed by the processor, and when the computer program is executed by the processor, the fuel cell vehicle energy management method as claimed in claim 1 is executed.

4. A storage medium, characterized in that The storage medium stores a computer program, which executes the fuel cell vehicle energy management method according to claim 1 when running.

Images