A method for generating a flying car energy and thermal management coupling system strategy

Abstract

The present invention provides a method for generating a flying car energy and thermal management coupling system strategy. The method comprises: S101, modeling the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor, and cooling subsystem of a hydrogen-fueled hybrid flying car; S102, establishing a hydrogen-fueled flying car energy and thermal management system coupling model based on the models of the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor, and cooling subsystem; and S103, solving the hydrogen-fueled flying car energy and thermal management system coupling model using a PPO algorithm. By coupling the energy and thermal management systems, the present invention dynamically adjusts the operating parameters of the hydrogen fuel cell to adapt to real-time environmental changes and operational requirements, thereby improving energy utilization efficiency. Furthermore, by optimizing heat dissipation and heat dissipation, the system's service life is extended.

Description

Technical Field

[0001] The present invention relates to the fields of thermal management and energy management technology of hydrogen fuel hybrid vehicles, and in particular to a method for generating a strategy for a flying vehicle energy and thermal management coupling system. Background Art

[0002] Previous research on flying cars has shown that during flight, the power systems of flying cars, such as electric motors, fuel cells and battery packs, will generate a lot of heat, causing the system to overheat and seriously affecting the energy distribution of the flying car, thereby reducing the flight endurance and overall efficiency of the flying car, causing system stability failures and even causing safety issues. Summary of the Invention

[0003] In view of this, an object of the present invention is to provide a method for generating a flying car energy and thermal management coupling system strategy to solve or at least partially solve the above-mentioned problems existing in the prior art.

[0004] To achieve the above-mentioned object of the invention, the present invention provides a method for generating a flying car energy and thermal management coupling system strategy, the method comprising the following steps:

[0005] S101. Model the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor and cooling subsystem of the hydrogen fuel hybrid flying car;

[0006] S102. Establishing a coupled model of the energy and thermal management system of a hydrogen fueled flying vehicle based on the models of the fuel cell subsystem, the lithium battery subsystem, the fuel cell hydrogen compressor, and the cooling subsystem. The coupled model of the energy and thermal management system of the hydrogen fueled flying vehicle includes:

[0007] P fc,output =η fc (T fc )·P fc,input

[0008] P battery，actual =P battery ·η battery (T battery )

[0009] P battery，actual =P require -P fc,output

[0010]

[0011] Q loss =P fc,input ·(1-η fc (T fc ))

[0012] Q fc,external =∈ fc ·σ·A fc ·(T fc -T ambient )

[0013] Q internal =P battery ·(1-T battery (T battery ))

[0014] Q bat,external =∈ battery ·σ·A battery ·(T battery -T ambient )

[0015]

[0016] Among them, P fc,output is the power output of the fuel cell, P battery，actual is the actual power of the lithium battery, P compressor is the compressor power, Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, Q internal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment, T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, η fc (T fc ) is the efficiency at temperature T, η fc is the fuel cell efficiency, P fc,inout is the fuel cell input power, P battery is the theoretical charging power, η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P require is the required power, P fc,output is the power output of the fuel cell, is the mass flow rate, R is the gas constant, T1 is the initial temperature, γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure respectively, P fc,input is the fuel cell input power, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant, A fc is the radiation area of the fuel cell stack, ∈ battery is the emissivity of the lithium battery stack, A batteryis the radiation area of the lithium battery stack, T battery_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the last moment, Q bat_total is the total heat actually generated by the lithium battery, Q fc_total is the total heat actually generated by the fuel cell, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems, m bat and m fc are the weight of lithium battery and fuel cell respectively, c p,bat and c p,fc are the specific heat capacity of lithium batteries and fuel cells respectively, T0 is the standard temperature at sea level, L is the temperature lapse rate, h is the altitude, For ambient temperature changes;

[0017] S103. Use the PPO algorithm to solve the energy and thermal management system coupling model of the hydrogen fuel flying car. The specific steps are as follows:

[0018] S31, initialization: Initialize the strategy network and value network and their corresponding parameters;

[0019] S32, Data Collection: Run the current policy in the environment to generate a batch of trajectory data. Each trajectory contains a series of states, actions, rewards, and next states. Collect the advantage estimate and return value of each state. The reward function of this coupled model is as follows:

[0020] Reward total

[0021] =K reward Reward energy +(1-K reward )Reward yhermal

[0022] Reward energy =P fc,output +λ·(P battery，actual +P compressor )+μ(soc-soc target ) 2

[0023]

[0024] Among them, Reward total is the total reward, K reward is the reward function coefficient, ranging from [0,1], Reward energy Reward is the energy management reward function. thermalis the thermal management reward function, λ is the weight factor for converting power into an equivalent factor, SOC is the battery state of charge, between 0 and 1, 0 means no power, 1 means full power; target is the target battery state of charge, and the SOC is kept near the target battery state of charge; μ is the SOC penalty factor, is the lithium battery temperature, fuel cell temperature control factor; T battery,target is the target lithium battery temperature;

[0025] S33. Calculate odds estimate: The odds estimate uses the generalized odds estimate;

[0026] S34, Policy Optimization: Calculate the objective function. The PPO algorithm uses proximal policy optimization pruning to limit the amplitude of policy network updates.

[0027] S35, Value Network Optimization: Update the value network to minimize the mean square error of the value function and define the loss function;

[0028] S36, update and loop: Update the parameters of the policy network and value network, and repeat steps S32 to S35 until the set number of training rounds is reached or the convergence condition is met.

[0029] Furthermore, modeling the fuel cell subsystem includes:

[0030] P fc,output =η fc (T fc )·P fc,input

[0031] Q fc_total =Q loss -Q fc,external +Q comp_total

[0032] Q loss =P fc,input ·(1-η fc (T fc ))

[0033] Q fc,external =∈ fc ·σ·A fc ·(T fc 4 -T ambient 4 )

[0034] η fc (T fc )=η nominal ·g(T fc )

[0035] Among them, P fc,outputis the power output of the fuel cell, Q fc_total The total heat generated for the fuel cell, Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, η fc (T fc ) is the efficiency of the fuel cell at temperature T, η fc is the fuel cell efficiency, P fc,input is the fuel cell input power, Q fc,external is the heat transferred from the fuel cell to the environment, Q comp_total is the total heat of the compressor, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant, A fc is the radiation area of the fuel cell stack, T fc is the fuel cell temperature, T ambient is the ambient temperature, η nominal is the efficiency at standard temperature, g(T fc ) is the function of the effect of temperature on efficiency.

[0036] Furthermore, modeling the lithium battery subsystem includes:

[0037] Q bat_total =Q internal -Q bat,external

[0038] Q internal =P battery ·(1-η battery (T battery ))

[0039] Q bat,external =∈ battery ·σ·A battery ·(T battery 4 -T ambient 4 )

[0040] η battery (T battery )=η battery_ref -K battery ·(T battery -T ref )

[0041] P battery，actual =P battery ·η battery (T battery )

[0042] P battery，actual =P require -P fc,output

[0043] Among them, Q bat_total is the total heat actually generated by the lithium battery, Q internal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment, η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P battery，actual is the actual power of the lithium battery, Q internal is the internally generated heat, P battery is the theoretical charging power, ∈ battery is the lithium battery stack emission rate, σ is the Stefan-Boltzmann constant, A battery is the radiation area of the lithium battery stack, T battery is the lithium battery temperature, T ambient is the ambient temperature, η battery_ref is the efficiency at reference temperature, K battery is the coefficient of temperature on charging and discharging efficiency, T ref is the reference temperature, P require is the required power, P fc,output is the power output of the fuel cell.

[0044] Furthermore, modeling the fuel cell hydrogen compressor includes:

[0045]

[0046] Q comp_total =Q ad +Q mechanical

[0047]

[0048] Q mechanical =P loss

[0049] Among them, P compressor is the compressor power, Q comp_total is the total heat of the compressor, is the mass flow rate, R is the gas constant, T ambient is the ambient temperature, γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure respectively, Q ad The heat of the gas is increased, Q mechanical is the heat generated by mechanical friction and power loss, c p,in is the specific heat capacity of the inflowing fluid, P loss is the power loss.

[0050] Furthermore, the cooling subsystem is a combination of liquid cooling and air cooling. The cooling subsystems of the lithium battery and fuel cell are independent. The liquid cooling and air cooling are forced convection cooling. The forced convection cooling is expressed as follows:

[0051]

[0052] Where Q is the heat transfer, is the mass flow rate, c p is the specific heat capacity, △T is the temperature difference, ρ is the density of the fluid, A is the cross-sectional area of the fluid flow, and V is the flow velocity.

[0053] Furthermore, the cooling capacity of the lithium battery and fuel cell cooling subsystem is expressed as follows:

[0054] Q bat,c =Q bat,fan +Q bat,liquid

[0055] Q fc,c =Q fc,fan +Q fc,liquid

[0056]

[0057]

[0058] Among them, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems respectively, Indicates the temperature change of lithium battery. is the fuel cell temperature change, is the ambient temperature change, T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, Q bat,fan is the cooling capacity of the lithium battery fan, Q bat,liqyid is the cooling capacity of lithium battery fluid, Q fc,fan is the cooling capacity of the fuel cell fan, Q fc,liquid is the cooling capacity of the fuel cell fluid, Q bat_total is the total heat actually generated by the lithium battery, m bat is the weight of lithium battery, c p,bat is the specific heat capacity of lithium battery, Q fc_total Total heat generated by the fuel cell, m fc is the fuel cell weight, c p,fc is the specific heat capacity of the fuel cell, Q ad The heat of the gas is increased, Q bat,external Q is the heat transferred from the lithium battery to the environment. fc,externalis the heat transferred from the fuel cell to the environment, m ambient is the air quality, c p,ambient is the specific heat capacity of air, T battery_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the previous moment, T0 is the standard temperature at sea level, L is the temperature lapse rate, and h is the altitude.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] This paper proposes a method for generating a strategy for a coupled energy and thermal management system for a flying vehicle. Using a coupled energy and thermal management model for a hydrogen fuel cell flying vehicle, the system optimizes overall performance and efficiency. The thermal management system controls the hydrogen fuel cell's operation within an optimal temperature range to improve its energy conversion efficiency and minimize performance degradation caused by overheating. Simultaneously, the energy management system optimizes power distribution between the hydrogen fuel cell and other power sources to maximize energy utilization and meet flight requirements. This strategy generation method, using a PPO algorithm, dynamically adjusts the operating parameters of the hydrogen fuel cell to adapt to real-time environmental changes and operational requirements, improving energy efficiency. It also extends the system's service life by optimizing heat dissipation and heat dissipation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0062] Figure 1 A schematic flow chart of a method for generating a flying car energy and thermal management coupling system strategy according to an embodiment of the present invention;

[0063] Figure 2 This is a structural diagram of the energy and thermal management system of a hydrogen fuel cell hybrid flying car provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The principles and features of the present invention are described below with reference to the accompanying drawings. The enumerated embodiments are only used to explain the present invention and are not used to limit the scope of the present invention.

[0065] This embodiment provides a method for generating a strategy for a flying car energy and thermal management coupling system, the method comprising the following steps:

[0066] like Figure 1As shown, a flow chart of a method for generating a strategy for coupling energy and thermal management of a hydrogen fuel hybrid flying vehicle is provided, including:

[0067] S101. Model the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor, and cooling subsystem of the hydrogen fuel hybrid flying car, specifically including:

[0068] (1) The fuel cell subsystem is modeled as follows:

[0069] P fc,output =η fc (T fc )·P fc,input

[0070] Q fc_total =Q loss -Q fc,external +Q comp_total

[0071] Q loss =P fc,input ·(1-η fc (T fc ))

[0072] Q fc,external =∈ fc ·σ·A fc ·(T fc 4 -T ambient 4 )

[0073] η fc (T fc )=η nominal ·g(T fc )

[0074] Among them, P fc,output is the power output of the fuel cell, Q fc_total The total heat generated for the fuel cell, Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, η fc (T fc ) is the efficiency of the fuel cell at temperature T, η fc is the fuel cell efficiency, P fc,input is the fuel cell input power, Q fc,external is the heat transferred from the fuel cell to the environment, Q comp_total is the total heat of the compressor, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant (5.67×10 -8 w / m 2 ·k4 ), A fc is the radiation area of the fuel cell stack, T fc is the fuel cell temperature, T ambient is the ambient temperature, η nominal is the efficiency at standard temperature, g(T fc ) is the function of the effect of temperature on efficiency;

[0075] (2) The lithium battery subsystem is modeled as follows:

[0076] Q bat_total =Q internal -Q bat,external

[0077] Q internal =P battery ·(1-η battery (T battery ))

[0078] Q bat,external =∈ battery ·σ·A battery ·(T battery 4 -T ambient 4 )

[0079] η battery (T battery )=η batteru_ref -K battery ·(T battery -T ref )

[0080] P battery，actual =P battery ·η battery (T battery )

[0081] P battery，actual =P require -P fc,output

[0082] Among them, Q bat_total is the total heat actually generated by the lithium battery, Q internal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment (W), η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P battery，actual is the actual power of the lithium battery, Q internal is the internally generated heat (W), P battery is the theoretical charging power (W), ∈battery is the lithium battery stack emissivity, σ is the Stefan-Boltzmann constant (5.67×10 -8 w / m 2 ·k 4 ), A battery is the radiation area of the lithium battery stack, T battery is the lithium battery temperature, T ambient is the ambient temperature, η battery_ref is the efficiency at reference temperature, K battery T is the coefficient of temperature on charging and discharging efficiency (usually a negative value), ref is the reference temperature, P require is the required power, P fc,output is the power output of the fuel cell;

[0083] (3) The fuel cell hydrogen compressor is modeled as follows:

[0084]

[0085] Q comp_total =Q ad +Q mechanical

[0086]

[0087] Q mechanical =P loss

[0088] Among them, P compressor is the compressor power (W), Q comp_total is the total heat of the compressor, is the mass flow rate, R is the gas constant, T ambient is the ambient temperature, γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure (Pa), respectively, Q ad The heat of the gas is increased, Q mechanical is the heat generated by mechanical friction and power loss, c p,in is the specific heat capacity of the inflowing fluid, P loss is the power loss (W);

[0089] (4) Cooling subsystem modeling: The cooling system is a combination of liquid cooling and air cooling. The cooling systems of lithium batteries and fuel cells are independent. Liquid cooling and air cooling are forced convection cooling. The forced convection cooling is expressed as follows:

[0090]

[0091] Where Q is the heat transferred in watts; is the mass flow rate, in kilograms per second, which represents the mass of fluid flowing per unit time; cp is the specific heat capacity, in (J / kg·K); △T is the temperature difference, in (K) or (℃), which represents the temperature difference between the fluid temperature and the surface temperature of the object; ρ is the density of the fluid, A is the cross-sectional area of the fluid flow, and V is the flow velocity;

[0092] The cooling capacity of the lithium battery and fuel cell cooling subsystem is expressed as follows:

[0093] Q bat,c =Q bat,fan +Q bat,liquid

[0094] Q fc,c =Q fc,fan +Q fc,liquid

[0095]

[0096] Among them, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems respectively, Indicates the temperature change of lithium battery. is the fuel cell temperature change, is the ambient temperature change, T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, Q bat,fan is the cooling capacity of the lithium battery fan, Q bat,liquid is the cooling capacity of lithium battery fluid, Q fc,fan is the cooling capacity of the fuel cell fan, Q fc,liquid is the cooling capacity of the fuel cell fluid, Q bat_total is the total heat actually generated by the lithium battery, m bat is the weight of lithium battery, c p,bat is the specific heat capacity of lithium battery, Q fc_total is the total heat actually generated by the fuel cell, m fc is the fuel cell weight, c p,fc is the specific heat capacity of the fuel cell, Q ad The heat of the gas is increased, Q bat,external is the heat transferred from the lithium battery to the environment (W), Q fc,external is the heat transferred from the fuel cell to the environment, m ambient is the air quality, c p,ambient is the specific heat capacity of air, c battrey_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the previous moment, T0 is the standard temperature at sea level (288.15K), L is the temperature lapse rate (0.0065K / m), and h is the altitude.

[0097] S102. Establish a coupled model of the energy and thermal management system of the hydrogen fuel flying vehicle based on the models of the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor, and cooling subsystem, specifically including:

[0098] Establish a coupling model, and the coupling model shares the same ambient temperature T ambient , the energy and thermal management systems are coupled through temperature. Excessive temperature affects the efficiency of the energy management system, resulting in excessive heat loss, which increases the temperature, as shown below:

[0099] P fc,output =η fc (T fc )·P fc,input

[0100] P battery，actual =P battery ·η battery (T battery )

[0101] P battery，actual =P require -P fc,output

[0102]

[0103] Q loss =P fc,input ·(1-η fc (T fc ))

[0104] Q fc,external =∈ fc ·σ·A fc ·(T fc -T ambient )

[0105] Q internal =P battery ·(1-η battery (T battery ))

[0106] Q nat,external =∈ battery ·σ·A nattery ·(T battery -T ambient )

[0107]

[0108]

[0109] Among them, P fc,output is the power output of the fuel cell, Pbattery，actual is the actual power of the lithium battery, P compressor is the compressor power (W), Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, Q internal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment (W), T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, η fc (T fc ) is the efficiency at temperature T, η fc is the fuel cell efficiency, P fc,input is the fuel cell input power, P battery is the theoretical charging power (W), η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P require is the required power, P fc,output is the power output of the fuel cell, is the mass flow rate, R is the gas constant, T1 is the initial temperature (K), γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure (Pa), respectively, P fc,input is the fuel cell input power, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant (5.67×10 -8 w / m 2 ·k 4 ), A fc is the radiation area of the fuel cell stack, ∈ battery is the emissivity of the lithium battery stack, A battery is the radiation area of the lithium battery stack, T battery_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the last moment, Q bat_total is the total heat actually generated by the lithium battery, Q fc_total is the total heat actually generated by the fuel cell, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems, m bat and m fc are the weight of lithium battery and fuel cell respectively, c p,bat and c p,fc are the specific heat capacity of lithium batteries and fuel cells, respectively. T0 is the standard temperature at sea level (288.15K), L is the temperature lapse rate (0.0065K / m), and h is the altitude. For ambient temperature changes;

[0110] The structural diagram of the energy and thermal management system of the flying car is as follows Figure 2 As shown in the figure, the energy management system consists of a fuel cell subsystem, a lithium battery subsystem, a DC / DC converter, an AC / DC converter and a permanent magnet synchronous motor. The hydrogen fuel cell and the battery pack are connected through a high-voltage electrical bus. The hydrogen fuel cell provides power to the high-voltage bus through the DC / DC. The hydrogen fuel cell is connected to the high-voltage bus through a bidirectional DC / DC converter. The propulsion load includes all components from the inverter to the propeller / wheel of the flying car. The AC / DC inverter receives power from the high-voltage bus and inverts the DC power into three-phase AC power to drive the rotor motor or wheel motor. Six pairs of coaxial rotors are used to control the speed of the rotor propeller and generate lift in flight mode. One wheel motor is used for driving in ground mode. The thermal management system consists of a fan system and a water cooling system. The specific process is that the cooling fluid is pumped out of the water tank by a pump and removes the heat of the fuel cell stack, lithium battery pack, compressor and other equipment through the cooling pipe. The fan system dissipates heat from the fluid and equipment. The coupled control system controls the energy and thermal management coupling system.

[0111] S103. Use the PPO algorithm to solve the energy and thermal management system coupling model of the hydrogen fuel flying car. The specific steps are as follows:

[0112] S31. Initialization: Initialize the policy network and value network and their corresponding parameters, and initialize the experience replay buffer (used to save the experience of each round);

[0113] S32, Data Collection: Run the current policy network in the environment to generate a batch of trajectory data. Each trajectory contains a series of states, actions, rewards, and next states. The advantage estimate and return value of each state are collected.

[0114] The status of this coupled model is as follows:

[0115] [T battery ,T fc ,T ambient ,P require ]

[0116] The actions of this coupled model are as follows:

[0117] [V fc,liquid ,V fc,fan ,V battery,liquid ,V battery,fan ,P battery，actual ]

[0118] Among them, Vfc,liquid is the fuel cell coolant flow rate, V fc,fan is the fuel cell fan speed, V battery,liquid is the flow rate of lithium battery coolant, V battery,fan is the wind speed of the lithium battery fan;

[0119] The reward function of this coupled model is as follows:

[0120] Reward total

[0121] =K reward Reward energy +(1-K reward )Reward thermal

[0122] Reward energy =P fc,output +λ·(P battery，actual +P compressor )+μ(soc-soc target ) 2

[0123]

[0124] Among them, Reward total is the total reward, K reward is the reward function coefficient, ranging from [0,1], Reward energy Reward is the energy management reward function. thermal is the thermal management reward function, λ is the weight factor for converting power into an equivalent factor, SOC is the battery state of charge, between 0 and 1, where 0 means no power and 1 means full power;

[0125] soc target is the target battery state of charge, and the SOC is kept near the target battery state of charge; μ is the SOC penalty factor, is the lithium battery temperature, fuel cell temperature control factor; T battery,target is the target lithium battery temperature;

[0126] S33. Calculate the odds estimate: The odds estimate uses the generalized odds estimate (GAE), and the formula is as follows:

[0127]

[0128] δ t =r t +γv(s t+1 )-v(s t )

[0129] Where A(k) is the advantage estimate at time t, which reflects the advantage of taking a certain action over other actions in a certain state; T is the termination time step, l is the current time step, γ is the weight factor, and λ is in the (0,1) interval, indicating a compromise between bias and variance; δ t is the time difference at time t, which represents the value difference between the current state and the next state plus the moment reward; δ t+l is the time difference at time t+1, ν(s t+1 ) is state s t+1 The value estimate of ν(s t ) is state s t The estimated value of r t is the reward at time t.

[0130] S34, Policy Optimization: Calculate the objective function. The PPO algorithm uses proximal policy optimization clipping (ClippedObjective Function) to limit the range of policy network updates. The objective function is expressed as follows:

[0131]

[0132] Among them, J(θ) is the loss function of the PPO algorithm network, θ is the policy network parameter, θ′ is the old policy network parameter, P θ (a t |s t ) is the state of s under the current policy network parameters θ t Select action a when in state t With the probability that when the number of training steps reaches the update threshold, θ′ is updated to θ, and this update continues to iterate; A θ′ (s t ,a t ) is in state s t Use the old policy network parameters θ′ to select action a t The advantage function of , clip is the clipping function, and ε is a small hyperparameter used to limit the amplitude of the update.

[0133] S35. Value Network Optimization: Update the value network to minimize the mean square error (MSE) of the value function. The loss function is defined as:

[0134]

[0135] Among them, L(θ V ) is the loss function, V(s t θ V ) is the value network (with parameter θ) at time t V Represents) for state s tThe value estimate of , that is, the output of the value network, is expressed in state s t The expected value of future cumulative rewards; Expected value calculation symbol, indicating the expected value of all sampled data; E t is the expected value of all sampled data.

[0136] S36, update and loop: update the parameters of the policy and value networks, and repeat steps S32 to S35 until the set number of training rounds is reached or the convergence condition is met.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for generating a flying car energy and thermal management coupling system strategy, characterized in that: include: S101. Model the fuel cell subsystem, lithium battery subsystem, fuel cell hydrogen compressor and cooling subsystem of the hydrogen fuel hybrid flying car; S102. Establishing a coupled model of the energy and thermal management system of a hydrogen fueled flying vehicle based on the models of the fuel cell subsystem, the lithium battery subsystem, the fuel cell hydrogen compressor, and the cooling subsystem. The coupled model of the energy and thermal management system of the hydrogen fueled flying vehicle includes: P fc,output =the fc (T fc )·P fc,input P battery，actual =P battery ·or battery (T battery ) T battery，actual =P require -P fc,output Q loss =P fc,input ·(1-η fc (T fc )) Q fc,external =∈ fc ·σ·A fc ·(T fc -T ambient ) Q internal =P battery ·(1-η battery (T battery )) Q bat,external =∈ battery ·σ·A battery ·(T battery -T ambient ) Among them, P fc,output is the power output of the fuel cell, P battery，actual is the actual power of the lithium battery, P compressor is the compressor power, Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, Q intetnal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment, T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, η fc (T fc ) is the efficiency at temperature T, η fc is the fuel cell efficiency, P fc,input is the fuel cell input power, P battery is the theoretical charging power, η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P require is the required power, P fc,output is the power output of the fuel cell, is the mass flow rate, R is the gas constant, T1 is the initial temperature, γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure respectively, P fc,input is the fuel cell input power, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant, A fc is the radiation area of the fuel cell stack, ∈ battery is the emissivity of the lithium battery stack, A battery is the radiation area of the lithium battery stack, T battery_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the previous moment, Q bat_total is the total heat actually generated by the lithium battery, Q fc_total is the total heat actually generated by the fuel cell, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems, m bat and m fc are the weight of lithium battery and fuel cell respectively, c p,bat and c p,fc are the specific heat capacity of lithium batteries and fuel cells respectively, T0 is the standard temperature at sea level, L is the temperature lapse rate, h is the altitude, The ambient temperature changes; S103. Use the PPO algorithm to solve the energy and thermal management system coupling model of the hydrogen fuel flying car. The specific steps are as follows: S31, initialization: Initialize the strategy network and value network and their corresponding parameters; S32, Data Collection: Run the current policy in the environment to generate a batch of trajectory data. Each trajectory contains a series of states, actions, rewards, and next states. Collect the advantage estimate and return value of each state. The reward function of this coupled model is as follows: Reward total =K rrward ·Reward energy +(1-K yeward )Reward thermal Reward energy =P fc,output +λ·(P battery，actual +P compressor )+μ(soc-soc target ) 2 Among them, Reward total is the total reward, K reward is the reward function coefficient, ranging from [0,1], Reward energy Reward is the energy management reward function. thermal is the thermal management reward function, λ is the weight factor for converting power into an equivalent factor, SOC is the battery state of charge, between 0 and 1, where 0 means no power and 1 means full power; soc target is the target battery state of charge, and the SOC is kept near the target battery state of charge; μ is the SOC penalty factor, is the lithium battery temperature, fuel cell temperature control factor; T battery,target is the target lithium battery temperature; S33. Calculate odds estimate: The odds estimate uses the generalized odds estimate; S34, Policy Optimization: Calculate the objective function. The PPO algorithm uses proximal policy optimization pruning to limit the amplitude of policy network updates. S35, Value Network Optimization: Update the value network to minimize the mean square error of the value function and define the loss function; S36, update and loop: Update the parameters of the policy network and value network, and repeat steps S32 to S35 until the set number of training rounds is reached or the convergence condition is met.

2. The method for generating a flying car energy and thermal management coupling system strategy according to claim 1, characterized in that: Modeling the fuel cell subsystem includes: P fc,output =the fc (T fc )·P fc,input Q fc_total =Q loss -Q fc,external +Q comp_total Q loss =P fc,input ·(1-η fc (T fc )) Q fc,external =∈ fc ·σ·A fc ·(T fc 4 -T ambient 4 ) or fc (T fc )=η nominal ·g(T fc ) Among them, P fc,output is the power output of the fuel cell, Q fc_total The total heat generated for the fuel cell, Q loss is the heat loss, Q fc,external is the heat transferred from the fuel cell to the environment, η fc (T fc ) is the efficiency of the fuel cell at temperature T, η fc is the fuel cell efficiency, P fc,input is the fuel cell input power, Q fc,external is the heat transferred from the fuel cell to the environment, Q comp_total is the total heat of the compressor, ∈ fc is the fuel cell stack emissivity, σ is the Stefan-Boltzmann constant, A fc is the radiation area of the fuel cell stack, T fc is the fuel cell temperature, T ambient is the ambient temperature, η nominal is the efficiency at standard temperature, g(T fc ) is the function of the effect of temperature on efficiency.

3. The method for generating a flying car energy and thermal management coupling system strategy according to claim 1, characterized in that: Modeling the lithium battery subsystem includes: Q bat_total =Q internal -Q bat,external Q internal =P battery ·(1-η battery (T battery )) Q bat,external =∈ battery ·σ·A battery ·(T battery 4 -T ambient 4 ) or battery (T battery )=η battery_ref -K battery ·(T battery -T ref ) P battery，actual =P battery ·or battery (T battery ) P battery，actual =P require -P fc,output Among them, Q bat_total is the total heat actually generated by the lithium battery, Q internal is the heat generated inside the battery, Q bat,external is the heat transferred from the lithium battery to the environment, η battery (T battery ) is the efficiency of the lithium battery at temperature T, η battery is the lithium battery efficiency, P battery，actual is the actual power of the lithium battery, Q internal is the internally generated heat, P battery is the theoretical charging power, ∈ battery is the lithium battery stack emission rate, σ is the Stefan-Boltzmann constant, A battery is the radiation area of the lithium battery stack, T battery is the lithium battery temperature, T ambient is the ambient temperature, η battery_ref is the efficiency at reference temperature, K battery is the coefficient of temperature on charging and discharging efficiency, T ref is the reference temperature, P require is the required power, P fc,output is the power output of the fuel cell.

4. The method for generating a flying car energy and thermal management coupling system strategy according to claim 1, characterized in that: Modeling the fuel cell hydrogen compressor includes: Q comp_total =Q ad +Q mechanical Q mechanical =P loss Among them, P compressor is the compressor power, Q comp_total is the total heat of the compressor, is the mass flow rate, R is the gas constant, T ambient is the ambient temperature, γ is the adiabatic index, P2 and P1 are the compressed pressure and intake pressure respectively, Q ad The heat of the gas is increased, Q mechanical is the heat generated by mechanical friction and power loss, c p,in is the specific heat capacity of the inflowing fluid, P loss is the power loss.

5. The method for generating a flying car energy and thermal management coupling system strategy according to claim 1, characterized in that: Modeling the cooling subsystem includes: The cooling subsystem is a combination of liquid cooling and air cooling. The cooling subsystems of lithium batteries and fuel cells are independent. Liquid cooling and air cooling are forced convection cooling. The forced convection cooling is expressed as follows: Where Q is the heat transfer, is the mass flow rate, c p is the specific heat capacity, △T is the temperature difference, ρ is the density of the fluid, A is the cross-sectional area of the fluid flow, and V is the flow velocity.

6. The method for generating a flying car energy and thermal management coupling system strategy according to claim 5, characterized in that: The cooling capacity of the lithium battery and fuel cell cooling subsystem is expressed as follows: Q bat,c =Q bat,fan +Q bat,liquid Q fc,c =Q fc,fan +Q fc,liquid Among them, Q bat,c and Q fc,c are the cooling capacity of lithium battery and fuel cell subsystems respectively, Indicates the temperature change of lithium battery. is the fuel cell temperature change, is the ambient temperature change, T battery is the lithium battery temperature, T fc is the fuel cell temperature, T ambient is the ambient temperature, Q bat,fan is the cooling capacity of the lithium battery fan, Q bat,liquid is the cooling capacity of lithium battery fluid, Q fc,fan is the cooling capacity of the fuel cell fan, Q fc,liquid is the cooling capacity of the fuel cell fluid, Q bat_total is the total heat actually generated by the lithium battery, m bat is the weight of lithium battery, c p,bat is the specific heat capacity of lithium battery, Q fc_total is the total heat actually generated by the fuel cell, m fc is the fuel cell weight, c p,fc is the specific heat capacity of the fuel cell, Q ad The heat of the gas is increased, Q bat,external Q is the heat transferred from the lithium battery to the environment. fc,external is the heat transferred from the fuel cell to the environment, m ambient is the air quality, c p,ambient is the specific heat capacity of air, T battery_past is the lithium battery temperature at the last moment, T fc_past is the fuel cell temperature at the previous moment, T0 is the standard temperature at sea level, L is the temperature lapse rate, and h is the altitude.

Images