A method and apparatus for determining a gas supply strategy for a thermal management combined power plant

Abstract

This application provides a method and apparatus for determining the air supply strategy of a thermal management combined power unit, belonging to the field of aviation electromechanical technology. The method provided includes: acquiring the operating state parameters of the thermal management combined power unit; inputting the operating state parameters into a pre-trained prediction model to determine the air supply control parameters; training the prediction model using training samples, wherein the training process of the prediction model is based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints, and air supply flow constraints to construct a target loss function; controlling the distribution of cooling airflow in the cabin air supply circuit according to the cooling turbine outlet split ratio, controlling the distribution of fuel in each fuel branch according to the first fuel branch split ratio, the second fuel branch split ratio, and the third fuel branch split ratio, and controlling the air supply temperature entering the cabin according to the cabin air supply temperature. The method and apparatus provided in this application can solve the air supply control problem under multiple constraint coupling.

Description

Technical Field

[0001] This application relates to the field of aviation electromechanical technology, and in particular to a method and apparatus for determining the air supply strategy of a thermal management type combined power plant. Background Technology

[0002] With the development of more-electric / all-electric aircraft technology, combined power units have gradually integrated multiple functions such as power generation, starting, emergency power, and thermal management, becoming a core component of aircraft electromechanical systems. For example, existing technologies have proposed a thermal management-type combined power unit that can provide a cooling air source for the cabin in environmental control mode, while simultaneously using fuel circulation to remove heat generated by multiple heat sources within the power unit (such as lubricating oil and engine bleed air). However, in the actual operation of this unit, there is a deep coupling between the bleed air flow rate and temperature required for cabin air supply and the thermal management requirements within the power unit.

[0003] For the environmental control systems of such thermal management-type combined power units, existing control strategies mostly employ simple logic control or PID regulation. For example, they directly adjust the cooling turbine outlet flow ratio based on cabin temperature deviation, or switch fuel branch valves based on lubricating oil temperature limits. Existing solutions primarily focus on the structural integration of the unit and do not provide a coordinated control method between air supply and thermal management. Traditional methods typically treat cabin air supply control and power unit thermal management as two relatively independent loops, lacking a comprehensive consideration of their coupling relationship.

[0004] Due to the lack of effective coordinated control methods, existing solutions face numerous problems in practical applications. Under variable operating conditions, such as rapid changes in flight altitude, engine speed, or hydraulic power, it is difficult to simultaneously meet the cooling requirements of cabin air supply flow, air supply temperature, and multiple heat sources within the power plant, easily leading to insufficient or excessive air supply flow. Excessive air supply flow increases engine bleed air load, causing thrust loss and increased fuel consumption, while further deteriorating the thermal balance of the power plant, creating a vicious cycle where more bleed air results in heavier thermal load, requiring more fuel for cooling, and decreasing system efficiency. Furthermore, existing control methods struggle to incorporate engine safety boundaries such as exhaust temperature limits and compressor surge margin maintenance into the environmental control air supply strategy, posing a risk of engine overheating or surge. Therefore, a method for determining the air supply strategy that can simultaneously consider cabin air supply quality and engine safety constraints is urgently needed. Summary of the Invention

[0005] In view of this, this application provides a method and apparatus for determining the air supply strategy of a thermal management type combined power unit, in order to solve the technical problem that air supply control under multiple constraint coupling is difficult to simultaneously meet the engine safety boundary and the cabin air supply requirements.

[0006] Specifically, this application is implemented through the following technical solution:

[0007] The first aspect of this application provides a method for determining the gas supply strategy of a thermal management type combined power unit, the method comprising:

[0008] Obtain operating status parameters of thermal management type combined power unit;

[0009] The operating status parameters are input into a pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature.

[0010] The prediction model is obtained by training through training samples. The training process of the prediction model is based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints and air supply flow constraints to construct a target loss function. The air supply flow constraints are used to suppress the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity.

[0011] The distribution of cooling airflow in the cabin air supply circuit is controlled according to the cooling turbine outlet split ratio, the distribution of fuel in each fuel branch is controlled according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and the air supply temperature entering the cabin is controlled according to the cabin air supply temperature.

[0012] A second aspect of this application provides a gas supply strategy determination device for a thermal management type combined power unit, the device comprising an acquisition module, a determination module, and a processing module;

[0013] The acquisition module is used to acquire the operating status parameters of the thermal management type combined power unit;

[0014] The determining module is used to input the operating status parameters into a pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature.

[0015] The prediction model is obtained by training through training samples. The training process of the prediction model is based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints and air supply flow constraints to construct a target loss function. The air supply flow constraints are used to suppress the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity.

[0016] The processing module is used to control the distribution of cooling airflow in the cabin air supply circuit according to the cooling turbine outlet split ratio, control the distribution of fuel in each fuel branch according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and control the air supply temperature entering the cabin according to the cabin air supply temperature.

[0017] The method and apparatus for determining the air supply strategy of a thermal management combined power plant provided in this application simultaneously introduce exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints, and air supply flow constraints during the training process of the predictive model. This ensures that the model output naturally considers both engine safety boundaries and cabin air supply requirements, achieving coordinated control under multiple constraints. During online operation, operating state parameters are directly input into the predictive model, and the cooling turbine outlet split ratio, three-way fuel distribution ratio, and cabin air supply temperature are quickly output via forward propagation without iterative solutions, thus meeting real-time control requirements under varying operating conditions such as takeoff, climb, cruise, and descent. The air supply flow constraint uses the engine's allowable air supply capacity as a benchmark rather than the cabin demand flow, actively suppressing redundancy in the actual air supply flow relative to engine capacity. This solves the problem of increased thermal load and decreased system efficiency due to excessive bleed air, effectively reducing fuel compensation losses caused by bleed air. Adjusting the corresponding valves according to the output split ratio ensures precise matching of cabin air supply flow with demand, and achieving reasonable allocation of thermal management load through three-way fuel splitting, ultimately achieving coordinated control between engine safety constraints and environmental control air supply requirements. Attached Figure Description

[0018] Figure 1 This is a structural schematic diagram of the thermal management type combined power unit environmental control system provided in this application;

[0019] Figure 2 A flowchart of an embodiment of the method for determining the gas supply strategy of a thermal management type combined power unit provided in this application;

[0020] Figure 3 A schematic diagram of the second embodiment of the gas supply strategy determination device for the thermal management type combined power unit provided in this application;

[0021] Explanation of reference numerals in the attached drawings: 1. Compressor; 2. Combustion chamber; 3. Cooling turbine; 4. Power turbine; 5. First fuel-air heat exchanger; 6. Second fuel-air heat exchanger; 7. Fuel-hydraulic heat exchanger; 8. Regenerator; 9. Cockpit. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0023] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although this application may use the terms first, second, third, etc., to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0025] Example 1

[0026] The following specific embodiments are given to illustrate the technical solution of this application in detail.

[0027] To provide a better explanation, this application will first introduce the environmental control system of the thermal management type combined power unit involved.

[0028] Figure 1 Please refer to the structural schematic diagram of the thermal management type combined power unit environmental control system provided in this application. Figure 1 The thermal management type combined power unit mainly includes a compressor 1, a combustion chamber 2, a cooling turbine 3, a power turbine 4, a first fuel-air heat exchanger 5, a second fuel-air heat exchanger 6, a fuel-hydraulic heat exchanger 7, a regenerator 8, and a cabin 9.

[0029] The air flow path includes: outside air is compressed by compressor 1 and divided into two paths; the first path of air serves as the cabin air supply source, and after being cooled by exchanging heat with fuel in the first fuel-air heat exchanger 5, it mixes with the low-temperature cold air at the outlet of cooling turbine 3 and is delivered to cabin 9; the second path of air is cooled by exchanging heat with fuel in the second fuel-air heat exchanger 6 and then enters the regenerator 8, and after being cooled by exchanging heat in the regenerator 8, it enters the cooling turbine 3 to expand, do work and cool down, forming low-temperature cold air.

[0030] The high-temperature gas generated in combustion chamber 2 drives power turbine 4 and cooling turbine 3 to do work in sequence. The gas after doing work is discharged after heat exchange through regenerator 8.

[0031] The fuel flow path includes: after the fuel is output from the fuel tank, it is divided into three branches. The first fuel branch flows through the first fuel-air heat exchanger 5 to cool the cabin air supply source; the second fuel branch flows through the second fuel-air heat exchanger 6 to cool the compressed air; and the third fuel branch flows through the fuel-hydraulic heat exchanger 7 to remove the heat generated by the hydraulic system. After the heat exchange is completed, the fuel branches merge to form a confluence fuel.

[0032] Figure 2 This is a flowchart of an embodiment of the method for determining the gas supply strategy of the thermal management type combined power unit provided in this application. Please refer to... Figure 2 The method provided in this embodiment may include:

[0033] S201. Obtain the operating status parameters of the thermal management type combined power unit.

[0034] It should be noted that the operating status parameters of the thermal management combined propulsion system can be collected in real time through airborne sensors, namely, flight conditions, engine operating status, and thermal management load. These operating status parameters include flight altitude, flight Mach number, engine bleed air temperature, engine bleed air pressure, fuel tank outlet fuel temperature, fuel flow rate, and hydraulic power.

[0035] Among them, flight altitude and flight Mach number are provided by the air data system and belong to flight condition parameters. The former determines the ambient pressure and temperature, affecting bleed air characteristics and heat exchange efficiency; the latter reflects flight speed, affecting ramjet effect and engine operating point. Engine bleed air temperature and bleed air pressure belong to engine operating condition parameters, which are taken from temperature and pressure sensors at the compressor outlet or bleed air duct, respectively. Bleed air temperature determines the heat load of the first fuel-air heat exchanger 5, while bleed air pressure affects the expansion ratio of the cooling turbine 3 and the upper limit of air supply capacity.

[0036] Fuel tank outlet fuel temperature, fuel flow rate, and hydraulic power are thermal management load parameters. Fuel temperature is measured by a pipeline temperature sensor, fuel flow rate is provided by a flow meter, and hydraulic power is calculated by a hydraulic system power sensor. The lower the fuel temperature, the stronger the heat absorption capacity; the total fuel flow rate is the total heat sink capacity of the entire fuel cooling system, and also the total constraint on the proportional distribution of the three fuel branch flows; the hydraulic power directly reflects the heat that needs to be removed from the third fuel branch, i.e., the heat flowing through the fuel-hydraulic heat exchanger 7.

[0037] In addition to the seven real-time parameters mentioned above, pre-set safety boundary values ​​also need to be read from the airborne controller: engine exhaust temperature limit value. Compressor surge margin lower limit Maximum permissible air supply flow rate of the engine Maximum permissible temperature of fuel Cabin air supply flow rate requirements The cabin air supply temperature range and other boundary values ​​are used in subsequent constraint judgments and loss function calculations. The collected raw data, along with the safety boundary values, are fed into the prediction model.

[0038] S202. Input the operating status parameters into the pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature.

[0039] The prediction model is obtained by training through training samples. The training process of the prediction model is based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints, and air supply flow constraints to construct a target loss function. The air supply flow constraints are used to suppress the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity.

[0040] It should be noted that the prediction model is a BP neural network model, whose input consists of seven normalized operating state parameters collected by S201, and whose output consists of five air supply control parameters: cooling turbine outlet flow split ratio. First fuel branch diversion ratio Second fuel branch diversion ratio Third fuel branch diversion ratio and cabin air supply temperature .in, , , The proportions of fuel flow through the first fuel-air heat exchanger 5, the second fuel-air heat exchanger 6, and the fuel-hydraulic heat exchanger 7 to the total fuel flow, respectively, and satisfying the following conditions: + =1.

[0041] The BP neural network model needs to be obtained through pre-training, and the training process includes:

[0042] (1) Establish a thermodynamic simulation model of the thermal management type combined power unit.

[0043] in accordance with Figure 1 The combined power unit structure shown is used to construct component-level mathematical models for the compressor, combustion chamber, power turbine, and cooling turbine, respectively:

[0044] Compressor model: ;

[0045] Combustion chamber model: ;

[0046] Powered Turbine Model: ;

[0047] Cooling turbine model: ;

[0048] Then, by integrating the above sub-models based on the flow balance, pressure balance, and energy balance relationships of the combined power unit, the overall machine simulation model is obtained. Specific balance relationships include:

[0049] Flow balance: The combustion chamber outlet flow rate equals the power turbine outlet flow rate, that is... ;

[0050] Pressure balance: Ambient pressure equals the cooling turbine outlet pressure multiplied by the pipeline pressure loss coefficient, i.e. *P_loss.

[0051] Energy balance: Electricity generation equals the work done by the power turbine plus the work done by the cooling turbine and the work done by the decompression turbine, i.e. .

[0052] in, These are environmental pressure and environmental temperature, respectively. The compressor pressure ratio. The shaft speed is 1. These are the compressor's outlet flow rate, outlet temperature, outlet pressure, and power consumption, respectively. These are the inlet pressure, inlet temperature, inlet flow rate, and fuel quantity of the combustion chamber. These are the outlet flow rate, outlet temperature, and outlet pressure of the combustion chamber, respectively. These are the inlet pressure, inlet temperature, and expansion ratio of the power turbine, respectively. For the power turbine outlet flow rate, outlet temperature, outlet pressure, and output power; These are the inlet pressure, inlet temperature, and expansion ratio of the cooling turbine, respectively. The parameters are: turbine outlet flow rate, outlet temperature, outlet pressure, and output power; P_loss represents the pressure loss coefficient along the pipeline. This indicates the amount of electricity generated.

[0053] The above modeling methods can be found in existing technologies, and will not be repeated here.

[0054] (2) Based on the thermodynamic simulation model, training samples covering different flight conditions, engine conditions and thermal management conditions are generated.

[0055] Using the established thermodynamic simulation model, training samples were generated by running it under various typical operating conditions. The specific coverage included: flight conditions encompassing takeoff, climb, cruise, descent, and go-around phases, with five different combinations of altitude and Mach number for each phase; engine conditions covering 80%–100% of the converted engine speed; and thermal management conditions where hydraulic power varied from 0 to 100 kW, and cabin flow demand varied from 100 kg / h to 1000 kg / h. Each sample set included seven input parameters (flight altitude, Mach number, engine bleed air temperature, engine bleed air pressure, fuel tank outlet fuel temperature, total fuel flow, and hydraulic power) and five corresponding output parameters (cooling turbine outlet flow split ratio). Three-way fuel branch diversion ratio , , and cabin air supply temperature Hundreds of samples were generated, with 70% used as the training set, 20% as the validation set, and 10% as the test set. All input parameters were normalized before being fed into the neural network to eliminate the influence of different units, mapping the data to the [0, 1] interval. The normalization formula is: .

[0056] (3) The BP neural network model is trained using the training samples.

[0057] Construct a BP neural network structure: The input layer has 7 nodes (corresponding to 7 input parameters), and the hidden layers consist of two layers. The number of nodes in each layer is determined based on empirical formulas and experiments, typically ranging from 12 to 32 nodes. The activation function is either ReLU or Tanh. For example, the first layer has 16 nodes with ReLU; the second layer has 12 nodes with Tanh; and the output layer has 5 nodes with Purelin. During training, gradient descent with momentum or the Learning Model (LM) algorithm is used. The initial learning rate is set to 0.05, decaying every 100 training iterations. Training stops when the validation set loss does not decrease for 20 consecutive iterations or when the maximum number of iterations (e.g., 1500) is reached.

[0058] During training, the network continuously adjusts its weights and biases through backpropagation to minimize the objective loss function. The objective loss function L is a weighted sum of the mean squared error of the prediction error and four constraint penalty terms, i.e. .in, The mean squared error of the output represents the deviation between the predicted value and the true value. This is a penalty for exceeding the exhaust temperature limit, used to indicate that the exhaust temperature exceeds the limit value. The degree of exceeding the limit, This serves as a safety buffer zone to prevent the system from frequently triggering penalties at critical boundaries. This is a penalty term for insufficient compressor surge margin, used to characterize the actual margin. Below the preset lower limit The degree of inadequacy; This is a redundancy penalty term for air supply flow, used to suppress the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity; This is a fuel temperature overheating penalty term, used to characterize fuel temperatures exceeding permissible values. The degree of overheating. The above. , , , These are preset weighting coefficients, which can be adjusted according to the actual project situation; for example, all can be set to 1.

[0059] After training, the BP neural network model is deployed as a prediction model in the airborne controller. During online operation, the seven normalized operating state parameters collected in S201 are input into the prediction model, and the model outputs the results directly after forward propagation. , , , , Five gas supply control parameters.

[0060] S203. Control the distribution of cooling airflow in the cabin air supply circuit according to the cooling turbine outlet split ratio, control the distribution of fuel in each fuel branch according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and control the air supply temperature entering the cabin according to the cabin air supply temperature.

[0061] After determining the cooling turbine outlet flow split ratio First fuel branch diversion ratio Second fuel branch diversion ratio Third fuel branch diversion ratio and cabin air supply temperature Subsequently, based on these control parameters, the valves in the environmental control system of the thermal management type combined power unit are adjusted to achieve coordinated control of gas supply distribution and thermal management.

[0062] Specifically, according to Adjusting the flow divider valve at the cooling turbine outlet controls the distribution ratio of cooling airflow between the cabin air supply branch and the bypass branch. The airflow from the cabin air supply branch mixes with the hot air cooled by the first fuel-air heat exchanger 5 before being delivered to the cabin 9, while the airflow from the bypass branch is directly discharged into the atmosphere or recirculated. Adjust the flow control valve on the first fuel branch to control the fuel flow entering the first fuel-air heat exchanger 5. This portion of the fuel is used to cool the bleed air, which serves as the cabin air source. Adjust the flow control valve on the second fuel branch to control the fuel flow entering the second fuel-air heat exchanger 6. This portion of the fuel is used to cool the compressed air at the compressor outlet; according to Adjust the flow control valve on the third fuel branch to control the fuel flow into the fuel-hydraulic heat exchanger 7. This portion of fuel is used to remove heat generated by the hydraulic system. After heat exchange, the three fuel lines merge and return to the engine or fuel tank. Simultaneously, according to... Adjust the opening of the hot and cold air mixing valve to ensure that the temperature of the air finally delivered into the cabin 9 reaches the set value.

[0063] After implementing the above controls, it is also necessary to obtain the actual gas supply flow rate in real time. Exhaust temperature Compressor surge margin and fuel temperature This is to monitor whether the system is within a safe operating range. These measured values ​​are obtained through airborne sensors: the air supply flow rate is measured by a flow sensor on the cabin air supply line, the exhaust temperature is measured by a temperature sensor at the power turbine outlet, the compressor surge margin is calculated from a table based on the ratio of the compressor outlet pressure to the inlet pressure and the current speed, and the fuel temperature is measured by a temperature sensor on the fuel return line.

[0064] (1) When at least one of the actual air supply flow rate, exhaust temperature, compressor surge margin and fuel temperature does not meet the corresponding operating requirements, switch to the control mode based on the preset physical model.

[0065] It should be noted that the corresponding operating requirements refer to the pre-set safety boundaries, including: the exhaust temperature does not exceed the limit value. (e.g., 700℃), compressor surge margin not lower than the lower limit (e.g., 5%), fuel temperature does not exceed the maximum allowable temperature. (e.g., 93℃), actual air supply flow rate and cabin air demand flow rate The deviation between the parameters does not exceed a preset threshold (e.g., ±5%). If any of the above four measured parameters does not meet the corresponding safety boundary, it is determined that the current neural network output may cause engine safety risks or a decrease in air supply quality, and the neural network control mode is automatically exited, switching to the control mode based on the preset physical model.

[0066] It should be noted that the switching condition in this step is based on the measured value after control execution. In addition, if the output parameters themselves significantly exceed the safety boundary (e.g., cabin air supply temperature exceeds the allowable range of 10℃ to 30℃) after the neural network outputs control parameters but before control execution, switching can also be triggered directly, and the neural network output will no longer be executed. This embodiment uses post-execution measured monitoring as an example for explanation; the judgment logic before execution can be implemented in the same way, and will not be repeated here.

[0067] (2) Based on the preset physical model, the cooling turbine outlet flow ratio, the first fuel branch flow ratio, the second fuel branch flow ratio, the third fuel branch flow ratio and the cabin air supply temperature are re-determined, and air supply control is performed according to the re-determined air supply control parameters.

[0068] Under the control mode based on the preset physical model, the BP neural network model is no longer relied upon. Instead, a pre-stored thermodynamic model is invoked, namely the component-level mathematical model and equilibrium relationship established in S202, to recalculate the air supply control parameters. This physical model adopts an iterative solution method, with exhaust temperature and compressor surge margin as the primary constraints, prioritizing the engine's safety boundary, and then adjusting the cooling turbine split ratio within the remaining feasible region. Three-way fuel diversion ratio , , and cabin air supply temperature This ensures that the actual gas supply flow rate is as close as possible to the cabin's required flow rate. The redefined control parameters are then controlled by adjusting the corresponding valves to maintain the gas supply until the measured parameters return to a safe range, or until ground maintenance personnel intervene to check after the flight mission is completed.

[0069] Furthermore, considering the impact of different flight phases and environmental conditions on the air supply strategy, this embodiment also provides an adaptive weight adjustment mechanism based on flight phase and environmental conditions. Specifically, before inputting the operational status parameters into the prediction model, the current flight phase (takeoff, climb, cruise, descent, or go-around) and special environments (high temperature, high altitude, or extreme cold) are automatically identified based on parameters such as flight altitude, Mach number, engine speed change rate, and ambient temperature and pressure. For example, during the takeoff phase, the engine heat load is high, but the requirements for cabin air supply quality are relatively relaxed; in this case, the weight of the air supply flow redundancy penalty term can be appropriately reduced. Prioritize ensuring exhaust temperature and surge margin; during cruise, when the engine is at its steady-state operating point and the air supply demand is stable, the pressure can be appropriately increased. The focus should be on suppressing bleed air waste and reducing fuel compensation losses; during the descent phase, bleed air pressure changes drastically, and the surge margin penalty weight should be temporarily increased. To prevent compressor stall, in emergency situations such as go-around, the air supply flow redundancy constraint can be temporarily abandoned, and exhaust temperature and surge margin can be given the highest priority. Meanwhile, in high-temperature environments, fuel cooling capacity decreases, necessitating an increase in fuel temperature penalty weighting. And appropriately lower the upper limit of the cabin air supply temperature setpoint; in high-altitude environments, the compressor surge margin is reduced, requiring an increase in... The maximum air supply flow rate is limited. In extremely cold environments, fuel viscosity increases and heat exchange efficiency decreases, necessitating a reduction in fuel flow constraints and an increase in exhaust temperature penalty weights to prevent thermal shock. These adaptive adjustments are executed in real-time during each control cycle, dynamically modifying the weighting coefficients in the loss function. ~ In addition, boundary parameters such as the allowable range of cabin air supply temperature and the upper limit of air supply flow enable the prediction model to output the optimal air supply control parameters under different scenarios, thereby achieving proactive adaptation to complex flight environments.

[0070] The following example, using a specific flight condition, illustrates the gas supply strategy determination method of this embodiment.

[0071] Flight settings: Altitude 6000m, Mach number 0.65. Engine settings: Bleed air temperature 161℃, bleed air pressure 200kPa. Thermal management settings: Fuel tank outlet temperature 45℃, total fuel flow rate 16000L / h, hydraulic power 35kW. Preset safety boundaries include: engine exhaust temperature limit. =700℃, lower limit of compressor surge margin =5%, maximum permissible air supply flow rate of the engine =3000 kg / h, maximum allowable fuel temperature =93℃, cabin air supply flow requirement =500kg / h, cabin air supply temperature range is 10℃~30℃.

[0072] The above seven operating parameters (flight altitude, Mach number, bleed air temperature, bleed air pressure, fuel outlet temperature, total fuel flow, and hydraulic power) are normalized and then input into the trained BP neural network model. After forward propagation, the model outputs the following control parameters: cooling turbine outlet split ratio. =0.78, the first fuel branch diversion ratio =0.3, the diversion ratio of the second fuel branch =0.55, the third fuel branch diversion ratio =0.15, cabin air supply temperature =22.5℃.

[0073] After adjusting the corresponding valves according to the above control parameters, the measured system response was: actual gas supply flow rate. =512kg / h, compared with the demand value In comparison, the redundancy is only 2.4%; exhaust temperature =367℃, far below the limit of 700℃; compressor surge margin =14.77%, higher than the lower limit by 5%; fuel return temperature =91.5℃, which is lower than the maximum allowable temperature of 93℃. All safety constraints are met, and gas supply flow redundancy is effectively suppressed.

[0074] In contrast, if an unconstrained BP neural network is used (i.e., the loss function does not include penalties for exhaust temperature, surge margin, air supply flow rate, and fuel temperature), under the same input conditions, the model output will be... =0.75, =0.4, =0.4, =0.2, At 20℃, the actual air supply flow rate reached 615 kg / h after execution, with an air supply redundancy of 23%, significantly increasing the bleed air load and engine fuel compensation losses. Therefore, this application, by constructing a target loss function with multiple constraint penalty terms to train a neural network, can significantly suppress bleed air waste and achieve precise matching between air supply flow rate and demand, while meeting engine safety boundaries.

[0075] The air supply strategy determination method for a thermal management-type combined power unit provided in this embodiment trains a BP neural network by using exhaust temperature, compressor surge margin, fuel temperature, and air supply flow redundancy as multiple constraints to construct a target loss function, thus achieving deep coupling and synergy between air supply control and thermal management. In actual flight, this method only needs to collect seven operating state parameters to quickly output the cooling turbine split ratio, the three-way fuel distribution ratio, and the cabin air supply temperature, eliminating the need for complex online iterative calculations and significantly improving response speed under varying operating conditions. Since the air supply flow constraint is based on the engine's allowable air supply capacity rather than simply the cabin demand flow, it can actively suppress bleed air redundancy, effectively avoiding the vicious cycle of more bleed air leading to heavier thermal load and decreased system efficiency, and significantly reducing fuel compensation losses. Simultaneously, exhaust temperature constraints, surge margin constraints, and fuel temperature constraints jointly ensure the safety of engine hot-end components, aerodynamic stability, and the fuel system, preventing the problem of sacrificing engine safety boundaries due to excessive pursuit of air supply quality, as seen in traditional methods. Building upon this foundation, and incorporating an adaptive weight adjustment mechanism based on flight phases and environmental conditions, this application becomes applicable to various complex scenarios across the entire flight envelope, including takeoff, climb, cruise, descent, go-around, and conditions such as high temperature, high altitude, and extreme cold. Furthermore, the safety backoff mechanism seamlessly switches to physical model control based on thermodynamic equations when the neural network output is abnormal or measured parameters exceed limits, prioritizing engine safety before adjusting cockpit air supply, significantly improving the system's robustness and fault tolerance.

[0076] Example 2

[0077] Corresponding to the aforementioned embodiment of the method for determining the gas supply strategy of a thermal management type combined power unit, this application also provides an embodiment of an apparatus for determining the gas supply strategy of a thermal management type combined power unit.

[0078] Figure 3 This is a schematic diagram of Embodiment 2 of the gas supply strategy determination device for the thermal management type combined power unit provided in this application. Please refer to... Figure 3 The apparatus provided in this embodiment includes an acquisition module 310, a determination module 320, and a processing module 330;

[0079] The acquisition module 310 is used to acquire the operating status parameters of the thermal management type combined power unit;

[0080] The determining module 320 is used to input the operating status parameters into a pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature.

[0081] The prediction model is obtained by training through training samples. The training process of the prediction model is based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints and air supply flow constraints to construct a target loss function. The air supply flow constraints are used to suppress the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity.

[0082] The processing module 330 is used to control the distribution of cooling airflow in the cabin air supply circuit according to the cooling turbine outlet split ratio, control the distribution of fuel in each fuel branch according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and control the air supply temperature entering the cabin according to the cabin air supply temperature.

[0083] The apparatus of this embodiment can be used to perform... Figure 2 The steps of the method embodiment shown are similar in principle and process, and will not be repeated here.

[0084] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0085] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0086] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for determining the gas supply strategy of a thermal management type combined power unit, characterized in that, The method for determining the gas supply strategy of the thermal management type combined power unit includes: Obtain operating status parameters of thermal management type combined power unit; The operating status parameters are input into a pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature. The prediction model is trained using training samples. The training process constructs a target loss function based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints, and air supply flow constraints. The air supply flow constraint is used to suppress redundancy in the actual air supply flow relative to the engine's allowable air supply capacity. Specifically… ,in, This is a penalty for redundant gas supply flow. For the sample size, For the first The actual gas supply flow rate of each sample For the first The permissible air supply capacity of the engine corresponding to each sample; The distribution of cooling airflow in the cabin air supply circuit is controlled according to the cooling turbine outlet split ratio, the distribution of fuel in each fuel branch is controlled according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and the air supply temperature entering the cabin is controlled according to the cabin air supply temperature.

2. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The operating parameters include flight altitude, flight Mach number, engine bleed air temperature, engine bleed air pressure, fuel tank outlet fuel temperature, fuel flow rate, and hydraulic power.

3. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The first fuel branch diversion ratio, the second fuel branch diversion ratio, and the third fuel branch diversion ratio correspond to the proportions of fuel flow through the first fuel-air heat exchanger, the second fuel-air heat exchanger, and the fuel-hydraulic heat exchanger to the total fuel flow, respectively. The sum of the first fuel branch diversion ratio, the second fuel branch diversion ratio, and the third fuel branch diversion ratio is 1.

4. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The prediction model is a BP neural network model; the training process of the BP neural network model includes: Establish a thermodynamic simulation model of the thermal management type combined power unit; Training samples covering different flight conditions, engine conditions, and thermal management conditions are generated based on the aforementioned thermodynamic simulation model. The BP neural network model is trained using the training samples.

5. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The target loss function includes a loss term corresponding to the prediction error, a penalty term corresponding to the exhaust temperature constraint, a penalty term corresponding to the compressor surge margin constraint, a penalty term corresponding to the air supply flow constraint, and a penalty term corresponding to the fuel temperature constraint. The prediction model is trained based on the target loss function.

6. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 5, characterized in that, The penalty term corresponding to the exhaust temperature constraint is used to characterize the degree to which the exhaust temperature exceeds the exhaust temperature limit value. The penalty term corresponding to the compressor surge margin constraint is used to characterize the degree of inadequacy of the compressor surge margin being lower than the preset margin lower limit; The penalty term corresponding to the air supply flow constraint is used to characterize the redundancy of the actual air supply flow relative to the engine's allowable air supply capacity. The penalty term corresponding to the fuel temperature constraint is used to characterize the degree of overheating when the fuel temperature exceeds the fuel temperature limit value.

7. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 5, characterized in that, The target loss function is obtained by weighting and summing the loss term corresponding to the prediction error, the penalty term corresponding to the exhaust temperature constraint, the penalty term corresponding to the compressor surge margin constraint, the penalty term corresponding to the air supply flow constraint, and the penalty term corresponding to the fuel temperature constraint according to preset weights.

8. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The distribution ratio of cooling airflow between the cabin air supply branch and the bypass branch is controlled according to the cooling turbine outlet split ratio. The fuel flow rate entering the first fuel-air heat exchanger is controlled according to the first fuel branch diversion ratio. The fuel flow rate entering the second fuel-air heat exchanger is controlled according to the second fuel branch diversion ratio. The flow rate of fuel entering the fuel hydraulic heat exchanger is controlled according to the third fuel branch diversion ratio.

9. The method for determining the gas supply strategy of a thermal management type combined power unit according to claim 1, characterized in that, The method further includes: After executing the gas supply control, the actual gas supply flow rate, exhaust temperature, compressor surge margin, and fuel temperature are obtained. When at least one of the actual air supply flow rate, exhaust temperature, compressor surge margin and fuel temperature fails to meet the corresponding operating requirements, the control mode based on the preset physical model is switched. Based on the preset physical model, the cooling turbine outlet flow split ratio, the first fuel branch flow split ratio, the second fuel branch flow split ratio, the third fuel branch flow split ratio, and the cabin air supply temperature are redefined, and air supply control is executed according to the redefined air supply control parameters.

10. A device for determining the gas supply strategy of a thermal management type combined power unit, characterized in that, The device includes an acquisition module, a determination module, and a processing module; The acquisition module is used to acquire the operating status parameters of the thermal management type combined power unit; The determining module is used to input the operating status parameters into a pre-trained prediction model to determine the air supply control parameters, which include the cooling turbine outlet split ratio, the first fuel branch split ratio, the second fuel branch split ratio, the third fuel branch split ratio, and the cabin air supply temperature. The prediction model is trained using training samples. The training process constructs a target loss function based on exhaust temperature constraints, compressor surge margin constraints, fuel temperature constraints, and air supply flow constraints. The air supply flow constraint is used to suppress redundancy in the actual air supply flow relative to the engine's allowable air supply capacity. Specifically… ,in, This is a penalty for redundant gas supply flow. For the sample size, For the first The actual gas supply flow rate of each sample For the first The permissible air supply capacity of the engine corresponding to each sample; The processing module is used to control the distribution of cooling airflow in the cabin air supply circuit according to the cooling turbine outlet split ratio, control the distribution of fuel in each fuel branch according to the first fuel branch split ratio, the second fuel branch split ratio and the third fuel branch split ratio, and control the air supply temperature entering the cabin according to the cabin air supply temperature.

Images