An energy management method and system for an aviation hybrid powertrain

Abstract

The application belongs to the technical field of new energy aircraft power system, and provides an energy management method and system for an aviation hybrid power system, which comprises the following steps: S1, constructing a power flow calculation model of the hybrid power system; S2, constructing a multi-objective optimization reward function based on fuel consumption mass, pollutant emission mass and battery health state; S3, outputting an energy management strategy of the hybrid power system under the TD3 algorithm framework based on the constructed power flow calculation model of the hybrid power system and the multi-objective optimization reward function. The application has the advantages that the reward value corresponding to each action value is solved based on the multi-objective TD3 algorithm, and the action value that makes the multi-objective reward function converge to the maximum is outputted, so that the energy management strategy that can act on each component of the hybrid power system is obtained, the fuel economy and reliability of the hybrid power system are improved, and the endurance of the aircraft is increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy aircraft power systems, and particularly relates to an energy management method and system for an aviation hybrid power system. BACKGROUND

[0002] The pure electric power currently used by electric aircraft can only support a cruising time of dozens of minutes, which is far from meeting the needs of various flight missions with long endurance. To solve the endurance problem, the aviation turbine generator hybrid power system combines the advantages of high energy density and power transmission of aviation turbine engines and electricization, and can greatly improve the aircraft load and endurance time compared to the pure electric propulsion system. It has broad application prospects for equipment and has attracted the attention of global electric aircraft manufacturers.

[0003] Energy management technology is a key technology for realizing the reasonable allocation of power demand of different energy sources of the aviation turbine generator hybrid power system under complex flight conditions and ensuring the stable operation of the power system. It plays an important role in improving energy saving and emission reduction benefits, increasing energy utilization, and prolonging the service life of power batteries. At present, energy management technology has been maturely applied in hybrid electric vehicles, but it is still in its infancy in the aviation field.

[0004] Energy management technology can be divided into rule-based strategy, optimization-based strategy, and learning-based strategy. Chinese patent CN117585170A designs an energy management strategy based on a rule-based method, which has the disadvantage that the rule setting relies too much on artificial experience and cannot guarantee the optimal performance of each component of the hybrid power system. The energy management strategy designed based on rules will produce a state of charge (SOC) boundary for charging and discharging, and the hybrid power system is prone to power allocation result jumping or even oscillation when switching at this boundary, affecting the normal operation of the hybrid power system. Chinese patent CN115983109A trains a power prediction neural network model using a large amount of flight condition data, and then obtains an energy optimal allocation strategy for the hybrid power system under flight range using a dynamic programming optimization strategy. The disadvantage is that deep learning is only applied to demand power prediction, and the dynamic programming energy management strategy as a global optimization strategy needs to know all the flight condition information of the flight range, which has poor adaptability. Moreover, the strategy design only focuses on reducing the consumption of equivalent fuel, without considering the multi-objective optimization of other component performance. SUMMARY

[0005] The purpose of the present invention is to provide an energy management method and system for aviation hybrid power systems. By adopting a TD3-based hybrid power energy management deep reinforcement learning framework, and by constructing a competitive Actor-Critic network architecture, flight condition variables are introduced into the state variables of the deep reinforcement learning network, thereby increasing the adaptability of the energy management strategy in the flight envelope and solving the above-mentioned problems.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] An energy management method for an aviation hybrid power system comprises the following steps:

[0008] S1. Construct a power flow calculation model for the hybrid power system. The power flow calculation model includes a turbine power generation system power flow model and a power battery equivalent circuit model. The turbine power generation system power flow model represents the power-torque conversion relationship between the aircraft turbine engine and the generator in the hybrid power system.

[0009] S2. Construct a multi-objective optimization reward function based on fuel consumption quality, pollutant emission quality, and battery health status;

[0010] S3. Based on the constructed power flow calculation model of the hybrid power system and the multi-objective optimization reward function, the energy management strategy of the hybrid power system is output under the TD3 algorithm architecture.

[0011] Among them, the fuel consumption quality is formed by the real-time prediction model of fuel consumption, which is based on the aircraft turbine engine component-level model and is trained using the lightweight gradient boosting machine method.

[0012] The present invention also provides an energy management system for an aviation hybrid power system, the system comprising a memory and at least one processor, the memory storing a computer program, and the processor executing the computer program to implement the above-mentioned energy management method for an aviation hybrid power system.

[0013] Compared with the prior art, the present invention has at least the following beneficial effects:

[0014] (1) The present invention establishes a hybrid power system power flow calculation model as the environmental model of the TD3 optimization algorithm, sets the power battery state of charge (SOC), required power (Preq), altitude (H) and Mach number (Ma) of the aircraft turbine engine as state variables, and increases the adaptability of the energy management optimization method in the flight envelope.

[0015] (2) A multi-objective reward function fusion mechanism was designed, and a real-time fuel consumption prediction model, a battery state of health (SOH) degradation estimation model, and a pollutant emission quality estimation model suitable for aviation turbine power generation hybrid systems were established. Heterogeneous objectives such as fuel consumption quality, pollutant emission quality, and battery performance were mapped into a learnable reward weight space, achieving multi-objective optimization that takes into account the performance of other components, so that the energy management strategy takes into account both energy conservation and emission reduction performance and reliability;

[0016] (3) The multi-objective optimization problem is iteratively solved using the TD3 algorithm to calculate the reward value corresponding to each action value. The output is the action value that makes the multi-objective reward function converge to the maximum, thereby obtaining energy management power distribution instructions that can be applied to each component of the hybrid power system, thereby improving the fuel economy and reliability of the hybrid power system and increasing the endurance of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 is a schematic diagram of the hybrid power system and related system structures in an embodiment provided by the present invention;

[0019] Figure 2 This is a schematic diagram of the steps of the energy management method for an aviation turbine power generation hybrid system according to an embodiment of the present invention;

[0020] Figure 3 1 is a schematic diagram of a component-level model of an aviation turbine engine in an embodiment provided by the present invention;

[0021] Figure 4 It is a flowchart of the steps of the energy management optimization method based on the TD3 architecture in the embodiment provided by the present invention. DETAILED DESCRIPTION

[0022] The technical solutions between the various embodiments of the present invention can be combined with each other, but they must be based on the fact that ordinary technicians in this field can implement them. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0023] The following are specific embodiments of the present invention, and the technical solutions of the present invention are further described in conjunction with the accompanying drawings, but the present invention is not limited to these embodiments.

[0024] The corresponding hybrid power system and related system structures in this embodiment are as follows: Figure 1 As shown, the energy management controller obtains the required power of the propulsion system through the flight controller, executes the energy management strategy, and sends power distribution instructions to the hybrid power system.

[0025] The hybrid power system includes a turbine engine, a generator, an AC / DC inverter, a battery, and a DC / DC transmitter. The turbine engine drives the generator to convert mechanical energy into electrical energy. The AC / DC inverter converts the AC power generated by the generator into DC power and aggregates it into the DC power grid. The battery uses a bidirectional DC / DC transmitter to achieve "peak shaving and valley filling" of the propulsion system power. When the propulsion system power demand is greater than the efficient working power of the turbine power generation system, the battery provides supplementary power; when the propulsion system power demand is less than the efficient working power of the turbine power generation system, the battery stores the excess power of the turbine power generation system for charging, and the turbine power generation system does not need to frequently control the engine speed changes.

[0026] On this basis, if Figure 2 As shown, this embodiment proposes an energy management method for an aviation hybrid power system, including the following steps:

[0027] S1. Construct a power flow calculation model for the hybrid power system. The power flow calculation model includes a turbine power generation system power flow model and a power battery equivalent circuit model. The turbine power generation system power flow model represents the power-torque conversion relationship between the aircraft turbine engine and the generator in the hybrid power system.

[0028] S2. Construct a multi-objective optimization reward function based on fuel consumption quality, pollutant emission quality, and battery health status;

[0029] S3. Based on the constructed power flow calculation model of the hybrid power system and the multi-objective optimization reward function, the energy management strategy of the hybrid power system is output under the TD3 algorithm architecture.

[0030] In step S1, based on the principle of component power flow, the relationship between power and torque from the aircraft turbine engine to the generator is established. The power flow model of the turbine power generation system is:

[0031] ;

[0032] Where: T eng is the engine shaft torque, P eng is the engine shaft power, T gen is the generator shaft torque, P gen is the generator shaft power, N is the engine and generator speed, η gen is the generator efficiency.

[0033] The equivalent circuit model of the power battery is:

[0034] In discharge mode: ;

[0035] In charging mode: ;

[0036] Battery SOC changes: ;

[0037] Among them, P bat is the power of the battery. When in discharge mode, P bat >0, while in charging mode, P bat <0;P req is the total power required by the hybrid system, P gen is the generator shaft power, η AC / DC is the efficiency of the AC / DC components, η DC / DC is the efficiency of the bidirectional DC / DC component; V oc is the open circuit voltage of the battery, R0 is the internal resistance of the battery, Q ini is the initial capacity of the battery, Q0 is the nominal total capacity of the battery, I b is the battery current, and t is the time.

[0038] This embodiment increases the adaptability of the energy management optimization method in the flight envelope by establishing a hybrid power system power flow calculation model as an environmental model of the TD3 optimization algorithm.

[0039] In step S2, a multi-objective optimization reward function is established, which comprehensively considers fuel consumption, pollutant emissions, and battery health status (SOH). The form of the multi-objective optimization reward function is as follows:

[0040] ;

[0041] in, 、 、 are the weight coefficients of their respective corresponding items, is the fuel consumption mass, ΔSOH is the battery health status loss percentage, The quality of pollutant emissions.

[0042] Furthermore, to calculate the fuel consumption of an aviation turbine power generation system, it first constructs an aviation turbine engine component-level model, and then generates a data set for training a real-time fuel consumption prediction model based on the model. The data set covers the entire flight envelope operating conditions.

[0043] Therefore, if Figure 3 As shown, the input is the altitude H, Mach number Ma, engine speed N and turbine power required by the system Pe , the output is the fuel consumption rate W f The aviation turbine engine component-level model is solved to obtain the fuel consumption quality of the aviation turbine power generation system.

[0044] The solution principle of the aircraft turbine engine component-level model is as follows:

[0045] ① Air intake

[0046] It calculates the static temperature and static pressure of the undisturbed section of the current engine inlet according to the altitude H and Mach number Ma, and further calculates the total temperature, total pressure and airflow velocity of the inlet section, and finally obtains the total temperature T at the inlet outlet. t2 and total pressure.

[0047] ② Compressor

[0048] The function of the compressor is to pressurize the airflow from the intake to the combustion chamber so that the airflow can reach the combustion conditions. The calculation of the thermal parameters of the airflow in the compressor depends on the component characteristic diagram, according to the current engine speed N and the total temperature T t2 The current shaft reduced speed N can be calculated c Then, in the compressor component characteristic diagram, the R line is introduced, based on the initial guess value R Line , the current compressor efficiency can be obtained by using the two-dimensional interpolation method , converted flow W C2 and pressure ratio Further calculations can be performed to obtain the total pressure at the compressor outlet and the actual flow rate W3. Based on the inlet cross-sectional enthalpy h2 and the actual compressor outlet enthalpy h3, the compressor power consumption P is obtained. HPC for: .

[0049] ③Combustion chamber

[0050] The combustion chamber is a component where the high-pressure gas compressed by the high-pressure compressor is mixed with the injected fuel to burn, converting the chemical energy of the fuel into heat energy and kinetic energy. After the fuel is added to the combustion chamber, the flow rate becomes W4=W3+W f , where W4 is the combustion chamber outlet flow rate.

[0051] ④Turbine

[0052] The turbine is a component that converts high-temperature and high-pressure airflow into kinetic energy and generates work. The high-pressure and high-temperature airflow from the combustion chamber hits the turbine, and the resulting expansion work drives the compressor and the load (high-speed generator). Similar to the working principle of the compressor, based on the characteristic parameters of the airflow at the combustion chamber outlet, the combined turbine pressure drop ratio The initial guess value and the turbine characteristic diagram can be used to perform the thermal calculation of the turbine, which will not be repeated here. The final result is the turbine work power P HPT= W5(h5-h4). Where h4 is the turbine inlet cross-sectional enthalpy and h5 is the actual turbine outlet enthalpy.

[0053] ⑤Tail nozzle

[0054] The turbine converts the engine's work into the kinetic energy of the rotor, and the outlet airflow is discharged into the atmosphere through the nozzle and exhaust. Assume that the model fluid is compressible, subsonic, and unsteady. The fluid flows through a flow channel with varying area without performing work or transferring heat. Calculate the flow rate W8 based on the outlet area.

[0055] ⑥Controller

[0056] The controller adopts a closed-loop proportional-integral (PI) controller, and the input is the command speed N cmd and actual speed N, the output is the engine model input fuel consumption rate W f .

[0057]

[0058] Among them, k p is the proportionality coefficient, k i is the integral coefficient, and z is the integral term.

[0059] ⑦Joint working equations and solutions

[0060] The common working equations of the engine component-level model include three flow balance equations and one rotor dynamics equation. The flow balance equations are shown in Table 1.

[0061] Table 1 Specific correspondence of flow balance equation

[0062] Equation Description Equation expression Initial guess parameters Compressor flow balance equation #timg# [R Line ]] Turbine flow balance equation #timg# #timg# Core engine nozzle flow balance equation #timg# <![CDATA[ W 0]]>

[0063] In the table, W C2 is the actual converted flow rate of the compressor, W C2map is the compressor converted flow obtained by interpolation of the compressor characteristic diagram; W C4 is the actual converted flow rate of the turbine, W C4map is the turbine converted flow obtained by interpolation of the turbine characteristic map; W0 is the intake inlet flow.

[0064] The rotor dynamics equations are as follows:

[0065] ;

[0066] Where, J is the moment of inertia of the engine shaft, P HPT is the turbine power, P HPC is the power consumption of the compressor, P e is the power extracted from the engine shaft by the generator.

[0067] The flow rate of each section (including the inlet W0) of an aircraft turbine engine during actual operation is difficult to measure directly. In order to understand the characteristics of engine components through modeling, it is necessary to first assume several initial guesses. There are two ways to calculate the flow rate of each section. One is to directly calculate the actual W from the inlet W0 through the component flow balance. i (i=1, 2, ..., n), another way to get W by initial guessing the value according to the component characteristic diagram imap (i=1, 2, ..., n), for each flight condition, all common equations are iteratively solved by numerical methods. When (i=1, 2, ..., n) is large, the initial guess value is adjusted based on the numerical method. When the errors of each equation converge or the maximum number of iterations is reached, that is, W i Basically equal to W imap When the rotor acceleration of the engine shaft is 0, the model is judged to have converged to a steady state, and the fuel consumption rate at this time is output.

[0068] Then, based on the above-mentioned aircraft turbine engine component-level model, a data set for training the real-time fuel consumption prediction model is generated. In the process of generating the data set, it is based on the altitude H, Mach number Ma, engine speed N and turbine power generation system required power P. e As the input, the thermodynamic parameters of each component are calculated by the above thermodynamic equations, and then the initial guess value is updated according to the result obtained by the joint working equation, and the thermodynamic parameters of each component are recalculated using the updated initial guess value. The error value in the joint working equation is When (i=1, 2, 3, 4) is less than or equal to the preset value, the loop ends, thereby obtaining the predicted fuel consumption rate W under multiple flight conditions. f to form a data set.

[0069] Finally, the Lightweight Gradient Boosting Machine (LightGBM) method was used to accelerate computation using a histogram algorithm. Gradient unilateral sampling and mutually exclusive feature bundling were used to optimize data and feature processing, and a leaf-based decision tree strategy was combined to achieve efficient training. The real-time fuel consumption prediction model obtained from training on the dataset can be expressed as:

[0070] ;

[0071] in, Fuel consumption quality.

[0072] In step S2, a degradation model that characterizes the battery state of health (SOH) is constructed. The discrete transient change expression of SOH is as follows:

[0073] ;

[0074] Where c is the battery discharge rate, ΔSOH is the percentage of battery health loss, N k is the total number of remaining battery cycles; I is the power battery current, ΔC n is the capacity loss percentage, is the pre-exponential factor positively correlated with the battery discharge rate, R is the ideal gas constant, Q t is the cumulative power throughput, T b is the internal temperature of the battery; E b is the battery activation energy and is negatively correlated with the discharge rate c. The more times the power battery switches between operating modes and the greater the transient discharge rate, the faster the battery state of health (SOH) degrades.

[0075] In step S2, the pollutant emission mass calculation principle is as follows: the initial electrical energy stored in the battery is considered equivalent to fuel. To ensure comparability of pollutant emissions from electricity and fuel, this embodiment tracks the emissions and consumption of both energy sources throughout their entire lifecycle. This means that fuel consumption includes not only post-combustion emissions but also emissions throughout its entire lifecycle, from production to consumption. The calculation of initial electricity consumption also takes this into account.

[0076] Therefore, the calculation expression of pollutant emission mass is as follows:

[0077] ;

[0078] Among them, among them, is the pollutant emission quality, is the CO2 emission mass, NO x Emission quality, is the fuel consumption quality, The power consumed by the power battery.

[0079] Furthermore, in step S2, the weight coefficients are obtained by a normalization method so that each target is comparable.

[0080] In this embodiment, by designing a multi-objective reward function fusion mechanism, heterogeneous objectives such as fuel consumption quality, pollutant emission quality, and battery performance are mapped into a learnable reward weight space, achieving multi-objective optimization that takes into account the performance of other components, so that the energy management optimization method takes into account both energy saving and emission reduction performance and reliability.

[0081] In step S3, this embodiment uses the TD3 algorithm framework to design an energy management optimization method for an aviation turbine generator hybrid system. The input state s and action a of the TD3 algorithm framework are defined as:

[0082] ;

[0083] a = P eng ;

[0084] Among them, SOC is the state of charge of the power battery, H and Ma are the altitude and Mach number of the aircraft turbine engine respectively, P req is the total power required by the hybrid system, P eng is the engine shaft power.

[0085] like Figure 4 As shown in Figure 2, the specific calculation process of the hybrid system energy management optimization method based on the TD3 algorithm architecture is as follows:

[0086] 1. Initialize the network

[0087] Initialize the six network parameters of the TD3 algorithm architecture, namely the strategy actor network , Strategy Critic1 Network , Strategy Critic2 Network , target Actor network , target Critic1 network , target Critic2 network , and initialize the buffer D for storing interaction data.

[0088] 2. Data sampling and storage

[0089] According to the current state t , target Actor network generates action , where ξ~N(0,σ) is Gaussian noise, and the noise range is clipped by clip(ξ,−c,c) (c represents the boundary value of the truncated action noise), and action a is performed. t , get reward r t and the next state s t+1 , storage(s t ,a t ,r t ,s t+1 ) to buffer D.

[0090] 3. Strategic Dual Critic Network Update

[0091] Sample batch data from buffer D and calculate the target Q value by taking the minimum value of the dual-objective Critic network:

[0092] ;

[0093] Among them, λ is a discount factor ≤1, and the larger λ is, the more emphasis is placed on long-term interests.

[0094] Minimize the critic's mean square error:

[0095] ;

[0096] At the same time, the stochastic gradient descent method is used to update the parameters of the current value network (θ i ).

[0097] 4. Delayed Strategy Actor Network Update

[0098] Every d=2 steps, by maximizing Update the policy Actor network to avoid premature optimization due to non-convergence of the policy Critic network:

[0099] ;

[0100] Where ▽ is the gradient operator symbol.

[0101] 5. Target network soft update

[0102] Slowly update the target network with a soft update factor coefficient τ≪1 to improve training stability:

[0103] .

[0104] Based on the energy management optimization method of the TD3 algorithm architecture above, repeat steps 2 to 5 until the strategy converges or the maximum number of training rounds is reached, and obtain the energy management strategy that maximizes the reward function of the multi-objective optimization in step 2.

[0105] This embodiment iteratively solves the multi-objective optimization problem through the TD3 algorithm architecture that combines a dual critic network, delayed updates, and target policy smoothing. Based on the multi-objective TD3 algorithm, the reward value corresponding to each action value is calculated, and the output makes the multi-objective reward function converge to the maximum action value, thereby obtaining energy management power allocation instructions that can be applied to various components of the hybrid power system, thereby forming an energy management strategy, thereby improving the fuel economy and reliability of the hybrid power system and increasing the endurance of the aircraft.

[0106] This embodiment also provides an energy management system for an aviation hybrid power system. The system includes a memory and at least one processor. The memory stores a computer program. The processor is used to execute the computer program to implement the energy management method for the aviation hybrid power system as described above.

[0107] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Persons skilled in the art may make various modifications, additions, or substitutions to the described specific embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.

[0108] It should be noted that all directional indications, such as upper, lower, left, right, front, rear, etc., are only used for the purpose of explanation and are not to be construed as indicating or implying relative importance or constituting limitations of the application. The terms "first", "second", "one", "the other", etc., are used only for the purpose of description and are not to be construed as indicating or implying relative importance or constituting limitations of the indicated technical features.

[0109] In addition, the descriptions such as "first", "second", "one", etc. in the present application are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise specifically limited.

[0110] In the present application, unless otherwise specifically defined and limited, the terms "connection", "fixing", etc. should be understood in a broad sense, for example, "fixing" can be fixed connection, or detachable connection, or integral; can be mechanical connection, or electrical connection; can be directly connected, or indirectly connected through intermediate medium; can be the internal communication of two elements or the interaction relationship between two elements, unless otherwise specifically limited. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

Claims

1. An energy management method for an aviation hybrid power system, characterized in that: Including steps: S1. Constructing a power flow calculation model for the hybrid power system, the power flow calculation model including a turbine power generation system power flow model and a power battery equivalent circuit model. The turbine power generation system power flow model represents a power-torque conversion relationship between an aircraft turbine engine and a generator in the hybrid power system; S2. Construct a multi-objective optimization reward function based on fuel consumption quality, pollutant emission quality, and battery health status; S3: Based on the constructed hybrid power system power flow calculation model and multi-objective optimization reward function, output the hybrid power system energy management strategy under the TD3 algorithm architecture; In step S1, the power flow model of the turbine power generation system is: ； Where: T eng is the engine shaft torque, P eng is the engine shaft power, T gen is the generator shaft torque, P gen is the generator shaft power, N is the engine and generator speed, η gen is the generator efficiency; In step S2, the fuel consumption quality is formed by a real-time prediction model of fuel consumption, which is based on a component-level model of an aviation turbine engine and is trained using a lightweight gradient boosting machine method; the input quantities are altitude H, Mach number Ma, engine speed N and turbine power generation system required power P e , the output is the fuel consumption rate W f The aviation turbine engine component level model is solved to obtain the fuel consumption quality of the aviation turbine power generation system; In step S3, the input state s and action a of the TD3 algorithm architecture are defined as: ； a = P eng ; Among them, SOC is the state of charge of the power battery, H and Ma are the altitude and Mach number of the aircraft turbine engine respectively, P req is the total power required by the hybrid system, P eng is the engine shaft power.

2. The energy management method for an aviation hybrid power system according to claim 1, characterized in that: In step S1, the power battery equivalent circuit model is: In discharge mode: ; In charging mode: ; Battery SOC changes: ; Among them, P bat is the power of the battery, P req is the total power required by the hybrid system, P gen is the generator shaft power, η AC / DC is the efficiency of the AC / DC components, η DC / DC is the efficiency of the bidirectional DC / DC component; V oc is the open circuit voltage of the battery, R0 is the internal resistance of the battery, Q ini is the initial capacity of the battery, Q0 is the nominal total capacity of the battery, I b is the battery current, and t is the time.

3. The energy management method for an aviation hybrid power system according to claim 1, characterized in that: The steps to obtain the real-time prediction model of fuel consumption are: Build component-level models of aviation turbine engines; generating a data set for training a real-time fuel consumption prediction model, wherein the data set includes all operating conditions of the flight envelope; The fuel consumption real-time prediction model is obtained by training using a lightweight gradient boosting machine based on the data set.

4. The energy management method for an aviation hybrid power system according to claim 1, characterized in that: In step S2, the reward function of pollutant emission quality is obtained according to the following formula, which is expressed as: ； in, is the pollutant emission quality, is the CO2 emission mass, NO x Emission quality, is the fuel consumption quality, The power consumed by the power battery.

5. The energy management method for an aviation hybrid power system according to claim 1, characterized in that: In step S2, the battery health state is obtained based on the degradation model of the battery health state, which is expressed as follows: ； Where c is the battery discharge rate, ΔSOH is the percentage of battery health loss, N k is the total number of remaining battery cycles; I is the power battery current, ΔC n is the capacity loss percentage, is the pre-exponential factor positively correlated with the battery discharge rate, R is the ideal gas constant, Q t is the cumulative power throughput, T b is the internal temperature of the battery, E b is the battery activation energy.

6. The energy management method for an aviation hybrid power system according to claim 1, characterized in that: The TD3 algorithm architecture includes a dual critic network, adds noise to the target actor network, and delays the update of the actor network to achieve training and obtain the energy management strategy of the hybrid power system.

7. An energy management system for an aviation hybrid power system, characterized in that: The system includes a memory and at least one processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the energy management method for an aviation hybrid power system according to any one of claims 1 to 6.

Images