An airborne power generation system can be thermally coupled to a multi-task cooperative testing method

Abstract

The application discloses a kind of airborne power generation system energy thermal coupling multi-task collaborative test method, belong to airborne energy system real-time simulation and test technical field;The test method includes: the airborne power generation system energy thermal coupling model of pre-constructed is configured in FPGA real-time simulator;By host computer, preset flight mission profile and working condition parameter are loaded;Motor electrothermal coupling submodel is with first simulation step length real-time solution electromagnetic transient process and loss power of motor, fuel thermal management subsystem submodel is with second simulation step length real-time solution fuel / oil thermal inertia dynamic process, two submodels are realized by cross clock domain synchronization two-way data exchange;Real-time acquisition FPGA feedback signal and generation collaborative control instruction, drive energy thermal coupling model response;Collect electrothermal dynamic response data, compare with preset safety threshold, judge the thermal safety boundary and the effectiveness of control strategy of system under the condition of limited cold source and multi-physical domain coupling, generate test evaluation result.

Description

Technical Field

[0001] This invention belongs to the field of real-time simulation and testing technology of airborne energy systems, specifically relating to a method for multi-task collaborative testing of airborne power generation systems with energy and heat coupling. Background Technology

[0002] With the deepening development of the More Electric Aircraft (MEA) and All Electric Aircraft (AEA) concepts, the capacity requirements for airborne electrical systems are growing exponentially. To meet the stringent weight and volume requirements of the aerospace industry, next-generation airborne generators are constantly evolving towards higher power density and higher speed. However, with the increase in power density, the electromagnetic loss density inside the generator also increases dramatically, leading to increasingly prominent heat dissipation and thermal management issues.

[0003] In airborne environments, thermal management systems face the stringent constraint of limited cooling sources. Aircraft typically utilize fuel as the final heat sink, transferring heat generated by the generator to the fuel system via lubricating oil circulation, which then carries the heat to the engine for combustion or dissipates it through the fuel tank surface. Therefore, a complex, strongly nonlinear coupling relationship exists between the airborne power generation system and the fuel thermal management system: the generator's output power determines the heat generated, while fuel consumption, temperature, and flow rate directly determine the system's heat dissipation capacity. Throughout flight missions such as takeoff, cruise, maneuvering, and landing, this heat source-heat sink balance is constantly undergoing dramatic dynamic changes. To ensure the safe operation of the aircraft across its entire flight envelope and under various extreme conditions, comprehensive and high-fidelity testing and verification of the airborne power generation system and its control strategies are particularly urgent.

[0004] Currently, the testing and verification of airborne power generation systems and their thermal management systems mainly rely on the following two technical methods, but both have their own limitations: (1) Physical iron bird bench testing and independent subsystem testing. This is currently the main verification method, which usually involves placing the generator on a ground-mounted bench and using an industrial constant-temperature oil source to simulate the lubricating oil cooling system. This method artificially severs the dynamic connection between the generator and the fuel system. In the test, the cooling medium is usually set to a constant temperature or a preset curve, which cannot realistically simulate the closed-loop heat accumulation effect caused by the decrease in heat sink capacity due to fuel consumption during flight, which in turn leads to an increase in return oil temperature. In addition, building a physical full-system bench that includes a real aircraft engine and complete fuel pipelines is extremely costly and poses a risk of fuel leakage and fire, making it difficult to conduct a large number of fault condition tests in the early stages of development.

[0005] (2) Multiphysics domain joint analysis based on offline simulation. High-precision electromagnetic-thermal-fluid coupling analysis is performed using commercial software (such as Ansys Maxwell combined with AMESim). Although this method has high accuracy, its computational efficiency is extremely low. Due to the huge difference in time scale between electromagnetic transients and fluid thermal inertia, completing a typical flight mission profile that lasts several hours often requires several days or even weeks of computation time, which is completely unable to meet the real-time requirements. Therefore, it cannot be used to verify the real-time control logic of the flight control computer or power generation controller, and it is difficult to verify the response performance of the control strategy under dynamic thermal boundaries.

[0006] To balance safety, efficiency, and real-time performance in testing, Hardware-in-the-Loop (HIL) technology has gradually become the mainstream verification method in the industry. However, existing HIL testing for airborne power generation systems mainly focuses on simulations in a single electrical domain, which has the following shortcomings: First, the model boundaries are idealized. Existing HIL platforms typically only run the electrical model of the motor, simplifying the thermal management system to constant boundary conditions (such as constant cooling oil temperature). Second, physical coupling is lacking. This "electrothermal separation" testing method cannot simulate the closed-loop process in real flight where fuel consumption leads to a decrease in heat sink capacity, which in turn causes a rise in generator temperature. The controller cannot detect the drift of motor parameters caused by temperature changes, resulting in a significant deviation between the test results and actual flight conditions.

[0007] The current HIL technology struggles to achieve high-fidelity real-time simulations of multi-physics domain coupling involving "electricity, heat, and fluidity" primarily due to two major technical bottlenecks: First, the challenge of modeling across time scales. The time constant of electromagnetic transients in motors is on the order of microseconds, requiring extremely small simulation steps, while the thermal inertia time constant of fuel / lubricating oil fluid networks is on the order of seconds, resulting in complex calculations and strong nonlinearity. Second, the real-time computational bottleneck. Traditional CPU-based real-time simulators, limited by serial instruction architectures, struggle to simultaneously solve high-frequency electromagnetic equations and complex fluid network equations within microsecond-level steps. Forcing such operation often leads to computational overload or model divergence.

[0008] To address the aforementioned technical challenges, there is an urgent need for a testing method that can overcome the difficulties of real-time calculation across time scales, deeply integrate high-precision motor electrothermal models with fuel thermal management models, and conduct multi-task profile verification with the participation of real controller hardware. Summary of the Invention

[0009] The technical problem to be solved: To avoid the shortcomings of existing technologies, this invention provides a method for thermally coupled multi-task collaborative testing of airborne power generation systems. By constructing a heterogeneous parallel solution architecture based on FPGA, it achieves hard real-time bidirectional coupling calculation of the electromagnetic transients of the motor at the nanosecond level and the thermal inertia of the fuel / lubricating oil fluid at the microsecond level. Full-task profile verification is performed with the participation of real controller hardware, thereby providing a low-cost, high-fidelity, and risk-free system-level collaborative testing method for airborne power generation systems.

[0010] The technical solution of this invention is: a method for thermal coupling multi-task collaborative testing of an airborne power generation system, the specific steps of which include: Step 1, Test Preparation: Configure the pre-built airborne power generation system energy-thermal coupling model in the FPGA real-time simulator. The energy-thermal coupling model includes a motor electrothermal coupling sub-model and a fuel thermal management sub-model, and there is a two-way closed-loop coupling relationship between the two: electrothermal generation → thermal parameter change → parameter conversion to electrothermal generation. Connect the FPGA real-time simulator with the real controller, host computer, output adapter module and oscilloscope to form a hardware-in-the-loop test platform. Step 2, Test Loading: Load the preset flight mission profile and operating parameters into the FPGA real-time simulator through the host computer. The flight mission profile includes typical flight phases such as takeoff, cruise, maneuvering and landing. The operating parameters include time-varying speed commands, load power requirements and dynamic fault injection commands. Step 3, Test Run: Start the test. The energy-thermal coupling model in the FPGA real-time simulator runs in real time under the drive of operating parameters. The motor electrothermal coupling sub-model calculates the electromagnetic transient process and power loss of the motor in real time with a first simulation step size. The fuel thermal management sub-model calculates the fuel / lubricating oil thermal inertia dynamic process in real time with a second simulation step size. The two sub-models achieve bidirectional coupling data exchange through cross-clock domain synchronization. The real controller collects the feedback signal output by the FPGA real-time simulator in real time, calculates and outputs cooperative control commands according to the preset control strategy, and feeds back the cooperative control commands to the FPGA real-time simulator to drive the energy-thermal coupling model to respond. Step 4, Test Data Acquisition: Real-time acquisition and recording of the electrothermal dynamic response data output by the energy-thermal coupling model during the test operation. The electrothermal dynamic response data includes dynamic curves of multiple parameters such as motor winding temperature, permanent magnet temperature, phase current, bus voltage, lubricating oil outlet temperature, and fuel tank temperature over time. Step 5, Test and Evaluation: Compare the collected electrothermal dynamic response data with the preset safety threshold to determine the effectiveness of the thermal safety boundary and control strategy of the system under the conditions of limited cold source and multi-physical domain coupling, and generate test and evaluation results.

[0011] A further technical solution of the present invention is: the motor electrothermal coupling sub-model includes an electrical model, a loss model, a thermal network model, and an electrical parameter correction module, wherein: The electrical model calculates the electromagnetic transient process of the motor based on the current electrical parameters and outputs current, frequency, and flux linkage information to the loss model. The loss model calculates copper loss based on the current output by the electrical model and the stator resistance updated in real time, calculates iron loss based on frequency and magnetic flux density, and inputs the calculated total loss power as a heat source into the thermal network model. The thermal network model calculates the real-time temperature of each key node inside the motor using the thermal equivalent circuit method based on the total power loss and preset heat capacity and thermal resistance parameters, and outputs the real-time temperature to the electrical parameter correction module. The electrical parameter correction module, based on the received real-time temperature, calls the preset temperature-parameter mapping relationship to correct the stator resistance and permanent magnet flux linkage online, and feeds back the corrected electrical parameters to the electrical model to replace the original parameters in the calculation of the next moment, thereby realizing information closed loop and dynamic collaboration.

[0012] A further technical solution of the present invention is: the fuel thermal management subsystem submodel includes a fuel tank model, a regulating pump model, a fuel distribution valve model, and a heat exchanger model, wherein, The fuel tank model dynamically updates the temperature and remaining mass of fuel in the tank based on the return oil heat injection and fuel consumption, simulating the process of heat sink capacity gradually decreasing with fuel consumption under limited cold source conditions. The regulating pump model adjusts the fuel mass flow rate in the fuel line according to the speed control command, providing a variable fluid boundary; The heat exchanger model includes at least a fuel oil / lubricating oil heat exchanger, which is used to receive the lubricating oil heat source from the electric motor electrothermal coupling sub-model, and to perform countercurrent heat exchange using the temperature difference between fuel oil and lubricating oil, and to calculate the outlet temperature after the lubricating oil cools down and the temperature after the fuel oil heats up. The fuel distribution valve model dynamically distributes the fuel flowing through the heat exchanger model into two paths according to the engine operating condition command and temperature feedback signal: one path supplies the engine combustion chamber, and the other path returns to the fuel tank after being cooled by the fuel cooler, forming a return bypass to regulate the overall temperature of the fuel tank.

[0013] A further technical solution of the present invention is: the bidirectional closed-loop coupling relationship of electrothermal generation → thermal parameter change → parameter conversion to electrothermal generation: First-stage heat transfer: The sum of stator copper loss and iron loss calculated in real time by the motor electrothermal coupling sub-model is used as the heat flow input and transferred to the lubricating oil circuit to calculate the lubricating oil temperature rise; Second-stage heat transfer: The heated lubricating oil is introduced into the fuel / lubricating oil heat exchanger model to exchange heat with the low-temperature fuel, and the lubricating oil outlet temperature and fuel temperature rise are calculated. Parameter feedback: The cooled lubricating oil outlet temperature and real-time flow rate are fed back to the motor electrothermal coupling sub-model as its cooling boundary conditions to participate in the calculation of the internal temperature rise rate of the motor. Parameter correction: Based on the real-time monitoring of the temperature of key motor nodes, the temperature change parameters, including stator resistance and permanent magnet flux linkage, are corrected online, and the updated parameters are fed back to the motor solver.

[0014] A further technical solution of the present invention is: the online correction of the temperature change parameter includes: The correction formula for stator resistance is:

[0015] In the formula, This is the real-time resistance value of the stator winding at the current temperature. The resistance value of the stator winding at the initial temperature. To represent the temperature coefficient of resistance of copper, This represents the current temperature of the stator winding. This is the initial temperature of the stator winding; The change in the flux linkage amplitude of a permanent magnet is reflected by the remanence of the permanent magnet, and its relationship with temperature is as follows:

[0016] In the formula, This represents the real-time remanence value of the permanent magnet at the current temperature. This is the initial remanence value. It is the reversible temperature coefficient of remanence in a permanent magnet. This represents the current temperature of the permanent magnet. This is the initial temperature of the permanent magnet; The real-time flux linkage and real-time remanence of permanent magnets are linearly positively correlated, with the following relationship:

[0017] In the formula, It is the flux linkage amplitude of the permanent magnet at the current temperature. These are fixed coefficients related to the stator and rotor structure, number of winding turns, and pole arc coefficient of the motor.

[0018] A method for constructing an FPGA solver for a thermally coupled model of an airborne power generation system, comprising the following specific steps: Step 1: Build a multi-physics domain real-time model library: Establish a component-level model library that includes an electric motor electrothermal coupling sub-model and a fuel thermal management subsystem sub-model; The motor electrothermal coupling sub-model has a built-in mapping relationship between motor electromagnetic parameters and temperature, as well as a mapping relationship between electromagnetic loss and electrical state, and a reserved parameter update interface for dynamically updating electrical parameters during simulation to reflect thermal decay characteristics, while also being able to output real-time power loss. The fuel thermal management subsystem sub-model includes at least a fuel tank model, a pump and valve model, and a heat exchanger model. The fuel tank model is used to describe the dynamic temperature rise process of fuel as the return oil heat is injected. The pump and valve model is used to simulate the dynamic response characteristics of fuel and lubricating oil flow rate adjusted by speed or control command. The heat exchanger model is used to calculate the heat transfer efficiency between lubricating oil and fuel and between fuel and other branches in real time. Step 2: Establish an energy-thermal coupling interaction mechanism: Define a two-stage heat transfer path and closed-loop feedback logic between the generator heat source-lubricating oil-fuel mixture sub-model and the fuel thermal management sub-model. Step 3: Construct an electrothermal coupled FPGA solver for the motor: The electric motor electrothermal coupling sub-model and its energy-thermal coupling interaction mechanism are mapped to the first clock domain of the FPGA, and the first clock domain is configured to run at the first simulation step size. The FPGA pipeline architecture is used to realize the real-time calculation of the electrical model, loss model, thermal network model and electrical parameter correction module in the electric motor electrothermal coupling sub-model, and output the real-time power loss and motor state parameters of the motor. Step 4: Construct the FPGA solver for the fuel thermal management system: The fuel thermal management subsystem sub-model and its energy-thermal coupling interaction mechanism are mapped to the second clock domain of the FPGA, and the second clock domain is configured to run at a second simulation step size, which is larger than the first simulation step size. First, the state space of the oil tank model, pump and valve model and heat exchanger model are reconstructed to clarify the state variables, input variables and output variables of each model. The Euler method is used to transform the continuous domain differential equations of each model into discrete time domain state update equations and output equations; By utilizing the parallel pipeline technology of FPGA, the computational logic of each model is mapped to hardware logic modules that are executed in parallel. Based on the physical topology of the fuel thermal management system, the input-output connection relationships between each model are defined to form the data flow network of the overall solver; Step 5: Design a cross-clock domain synchronization module and integrate the solver: A cross-clock domain synchronization module is set up between the FPGA solver for the electric motor electrothermal coupling and the FPGA solver for the fuel thermal management system to realize deterministic data exchange between the two clock domains; The electric motor electrothermal coupling FPGA solver, the fuel thermal management system FPGA solver, and the cross-clock domain synchronization module are integrated on the same FPGA chip to form a unified energy-thermal coupling model FPGA solver. This solver can realize simultaneous, parallel, hard real-time bidirectional coupling calculation of electromagnetic transient processes and fluid thermal inertial processes during operation.

[0019] A further technical solution of the present invention is: the two-stage heat transfer path and closed-loop feedback logic of the generator heat source-lubricating oil-fuel specifically include: The total power loss calculated in real time by the electric-thermal coupling sub-model of the motor is transmitted to the lubricating oil circuit as a heat flow input to calculate the lubricating oil temperature rise. The heated lubricating oil is introduced into the fuel / lubricating oil heat exchanger model to exchange heat with the low-temperature fuel; The cooled lubricating oil outlet temperature and real-time flow rate are fed back to the motor electrothermal coupling sub-model as its dynamic cooling boundary conditions. The electric motor electrothermal coupling sub-model calculates the internal temperature rise of the motor based on the boundary conditions, and calls the parameter update interface to dynamically and adaptively correct the stator resistance and permanent magnet flux linkage, forming a system-level closed-loop coupling of electrothermal generation → thermal parameter change → parameter-electrical modification.

[0020] A further technical solution of the present invention is: the cross-clock domain synchronization module transmits the real-time power loss calculated by the motor electrothermal coupling FPGA solver to the fuel thermal management system FPGA solver as the heat source input boundary, and transmits the cooling medium state parameters calculated by the fuel thermal management system FPGA solver to the motor electrothermal coupling FPGA solver as the dynamic cooling boundary condition.

[0021] An airborne power generation system thermal coupling multi-task collaborative test platform for the method, comprising: An FPGA real-time simulator is configured with an onboard power generation system energy-thermal coupling model. The energy-thermal coupling model includes a motor electrothermal coupling sub-model and a fuel thermal management sub-model, and the two have a bidirectional closed-loop coupling relationship of electrothermal generation → thermal parameter change → parameter conversion to electricity. The FPGA real-time simulator is used to calculate the electrothermal dynamic process in real time with a first simulation step size and a second simulation step size, respectively, and output feedback signals. The real controller is connected to the FPGA real-time emulator and is used to collect the feedback signal, run the control strategy, and output cooperative control commands to the FPGA real-time emulator. The host computer is connected to the FPGA real-time simulator and is used to load flight mission profiles and operating parameters, and to receive, display and store the electrothermal dynamic response data collected during the test. The output adapter module is connected between the FPGA real-time emulator and the real controller and is used for signal conditioning and interface conversion. The FPGA real-time simulator, the real controller, and the host computer form a closed-loop test circuit, outputting test evaluation results.

[0022] A further technical solution of the present invention is: it also includes an oscilloscope, which is connected in parallel to the analog output port of the output adapter module, for capturing and displaying high-frequency dynamic waveforms of key nodes in real time, assisting in verifying the quality of the control commands output by the real controller and the response characteristics of the system under transient conditions, and providing independent physical waveforms for the test results.

[0023] Beneficial effects The beneficial effects of this invention are as follows: (1) This invention breaks through the technical bottleneck of real-time coupled simulation across multiple physical domains and significantly improves model fidelity. Addressing the limitations of existing HIL tests, which can only run a single electrical model and ignore the dynamic influence of the thermal management system, this invention constructs a heterogeneous parallel solution architecture based on FPGA. Utilizing the parallel pipeline technology of FPGA, a hard real-time bidirectional coupled solution of the electromagnetic transients of the motor at the nanosecond level and the thermal inertia of the fuel / lubricating oil fluid at the microsecond level is achieved. This architecture not only ensures the high-frequency response accuracy of electromagnetic calculations but also introduces dynamic thermal boundary feedback of the fluid network, enabling the simulation model to update the motor electrical parameters in real time to reflect thermal decay characteristics. This achieves high-fidelity reproduction of the closed-loop physical process of "electricity generating heat → thermal parameter change → parameter conversion to electricity," effectively solving the simulation error problem caused by "electrical-thermal separation" in traditional HIL tests.

[0024] (2) This invention solves the problem of insufficient coverage of the full mission profile under the constraint of limited cold source, and enhances the completeness of system-level verification. Addressing the limitation of traditional physical bench tests, which are typically limited to steady-state testing under constant cooling boundaries due to equipment limitations and safety risks, this invention establishes a full mission profile loading mechanism. It can simulate typical flight phases in real-time simulation, and in particular, by introducing a fuel tank model, it can calculate the impact of fuel mass reduction on the total system heat capacity in real time. This allows the testing process to no longer be limited to a single steady-state point, but to cover the entire long-term, highly dynamic flight mission cycle, dynamically capturing the system-level heat accumulation and thermal runaway risks caused by fuel level drops, high-G maneuvers, etc., thus providing a comprehensive verification method for evaluating the thermal safety boundary of the power generation system under the constraint of limited cold source.

[0025] (3) This invention provides a low-cost, high-safety testing environment with extreme fault injection capabilities, reducing R&D risks. Addressing the extremely high cost of building a physical iron bird test bench containing a real engine and complete fuel lines, and the significant fuel leakage and fire hazards associated with destructive fault testing, this invention utilizes digital twin technology to construct a virtual thermal management object. It supports real-time simulation of extreme conditions such as oil pump failure and empty fuel tanks by modifying internal FPGA variables, even without the involvement of physical fuel and lubrication systems. Testers can repeatedly verify the controller's over-temperature protection strategy and control logic under fault conditions in an absolutely safe environment, without consuming real fuel or shortening motor lifespan, significantly reducing the R&D costs and flight test risks of the airborne power generation system. Attached Figure Description

[0026] Figure 1 This is a general block diagram of a method for thermal coupling multi-task collaborative testing of an airborne power generation system according to an embodiment of the present invention; Figure 2 This is a flowchart of the airborne power generation system thermal coupling multi-task collaborative testing method in an embodiment of the present invention; Figure 3 This is a structural diagram of the electric motor electrothermal coupling model in an embodiment of the present invention; Figure 4 This is a schematic diagram of the fuel thermal management system in an embodiment of the present invention; Figure 5 This is a structural diagram of the FPGA solver for the PMSG electrothermal model in an embodiment of the present invention; Figure 6 This is a structural diagram of the FPGA model of the fuel thermal management system in an embodiment of the present invention; Figure 7 This is a structural diagram of the hardware-in-the-loop test platform in an embodiment of the present invention; Figure 8 This is a waveform diagram of temperature rise test under rated load conditions of the generator in an embodiment of the present invention; Figure 9 This is a waveform diagram of the temperature rise test under the generator's double overload condition in an embodiment of the present invention; Figure 10 This is a test waveform diagram of the generator state change under a pulse load profile with a pulse width of 5s and an interval of 10s in an embodiment of the present invention. Figure 11 This is a test waveform diagram of the generator's state change under a pulse load profile with a pulse width of 5s and an interval of 5s in an embodiment of the present invention. Figure 12 This is a test waveform diagram showing the state change of a generator under a pulse load profile with a pulse width of 5s and an interval of 5s in an embodiment of the present invention, where the generator has a high flow rate for heat dissipation. Detailed Implementation

[0027] The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0028] To address the shortcomings and deficiencies of existing methods, this invention proposes a thermally coupled multi-task collaborative testing method for airborne power generation systems. It aims to overcome the technical challenges of high costs and incomplete operational condition coverage in traditional physical bench testing, low efficiency of offline multi-physics domain simulation computation that cannot be run in real time, and model distortion caused by neglecting dynamic coupling in existing hardware-in-the-loop testing.

[0029] Reference Figure 1 As shown, the test method proposed in this invention is based on the FPGA heterogeneous parallel computing architecture and constructs a closed-loop test system that includes a multi-task condition generation layer, a heterogeneous parallel solution layer, and a parameter adaptation and feedback layer. It aims to realize real-time verification of the airborne power generation system under all operating conditions in a complex dynamic thermal environment.

[0030] The multi-task condition generation layer serves as the loading center for the full-task profile of the test system, responsible for flexibly defining typical flight mission sequences such as takeoff, cruise, maneuvering, and landing on the host computer. This layer generates time-varying speed commands, load power requirements, and dynamic fault injection commands, and sends the condition data to the heterogeneous parallel solution layer for solving. The heterogeneous parallel solution layer is the core computing unit. The motor electrothermal coupling solver operates in a hundred-nanosecond clock domain, responsible for real-time calculation of the motor's electromagnetic transient processes and power losses; the thermal management system solver operates in a microsecond clock domain, responsible for the dynamic calculation of fluid thermal inertia. The two can achieve precise synchronous interaction of electromagnetic power losses, fluid flow rate, and temperature. The parameter adaptation and feedback layer monitors the motor key node temperatures and fluid temperatures calculated by the intermediate layer in real time. Through a pre-set online correction module for heat-sensitive parameters, it corrects temperature-varying parameters such as stator resistance and permanent magnet flux linkage online, and feeds the updated parameters back to the motor solver, forming a closed-loop physical mechanism of "electricity generating heat - heat changing parameters - parameters changing electricity". Meanwhile, this layer can transmit key information monitored in real time to the host computer for monitoring, and provide real-time alarms in case of faults such as overheating.

[0031] The above technical solution will be further explained below with reference to the accompanying drawings and examples: In one embodiment, a high-power-density permanent magnet synchronous generator (PMSG) is used as an example to illustrate the construction of its electrothermal coupling sub-model. It should be understood that the energy-thermal coupling test method proposed in this invention is also applicable to other types of airborne generators such as wound-rotor synchronous generators and switched reluctance generators. Similarly, in the construction of the fuel thermal management system sub-model, this embodiment selects a simplified typical fluid network including a core fuel tank, basic lubricating oil pump valves, and a single-stage heat exchanger as a specific example for illustration. It should be noted that in practical engineering applications, this thermal management system model can be flexibly expanded into a refined model including multi-stage heat exchangers, complex pipeline branches, and multiphase flow states, depending on the specific engine requirements; the test architecture and energy-thermal coupling mechanism of this invention are equally applicable to such models.

[0032] Reference Figure 2 As shown, a method for constructing an FPGA solver for a thermally coupled model of an airborne power generation system is disclosed, including the following steps: Step 1: Construct a multi-physics domain real-time model library for airborne power generation systems 1.1 Electrothermal Coupler Model of Motor In this embodiment, the motor model employs a real-time electrothermal coupling calculation architecture based on FPGA as a specific example. This model pre-establishes the mapping relationship between the motor's electromagnetic parameters (resistance, flux linkage) and temperature, as well as the mapping relationship between electromagnetic losses and electrical states. During simulation, the model receives the cooling medium status (flow rate, temperature) from the thermal management system, calculates the motor's internal temperature in real time, and dynamically updates the electrical parameters, thereby accurately reflecting the motor's thermal attenuation characteristics under different cooling conditions. Notably, this model can output real-time heat flow data as the input boundary for the downstream thermal management system. The structure of the motor electrothermal coupling model used in this embodiment is as follows: Figure 3 As shown, it mainly consists of four functional modules: electrical model, loss model, thermal network model, and electrical parameter correction module.

[0033] The electrical model can be established based on the dq coordinate system modeling method, facilitating subsequent integration with the thermal network model. The loss model primarily considers copper losses and iron losses. Copper losses are determined by the stator winding current and its corresponding resistance at the current temperature, while iron losses are mainly affected by changes in the motor's operating frequency and magnetic flux density. Losses are input to the thermal network model as heat sources, driving the temperature evolution of different nodes within the motor. Furthermore, the introduction of iron losses not only provides a heat source but also feeds back to the electrical model, creating a need for reverse correction of the electrical model's structure and parameters. The thermal network model is used to simulate the steady-state and transient temperature distribution within the motor, implemented using the thermal equivalent circuit method, systematically modeling the heat capacity, thermal resistance, convective heat transfer conditions, and boundary conditions of each region of the motor. The output of the thermal network model directly affects temperature-sensitive electrical parameters and is fed back to the electrical model through the electrical parameter correction module, achieving information closure and dynamic coordination between the electrical and thermal models.

[0034] 1.2 Sub-model of fuel oil thermal management system This embodiment takes a simplified fuel thermal management system as an example, which mainly consists of components such as a fuel tank, pump, fuel distribution valve, and heat exchanger. A schematic diagram of the system structure is shown below. Figure 4 As shown: the fuel tank model describes the dynamic temperature rise process of fuel as heat is injected from the return oil; the pump and valve model simulates the dynamic response characteristics of fuel and lubricating oil flow rates as adjusted by speed or control commands, providing a variable fluid boundary for the system; the heat exchanger model calculates the heat transfer efficiency between lubricating oil and fuel, and between fuel and other branches in real time, reflecting the change in heat dissipation capacity under limited cold source conditions.

[0035] The system precisely controls the fuel flow rate in the pipeline by adjusting the pump speed. After being pressurized by the pump, the fuel in the tank enters the fuel / oil heat exchanger. The fuel / oil heat exchanger uses the temperature difference between the fuel and the oil to perform counter-current heat exchange, achieving oil cooling. Subsequently, the fuel passes through other heat exchangers for heat exchange before being delivered to the fuel distribution valve. This valve dynamically distributes the fuel into two paths based on engine operating condition commands and temperature feedback signals: Main flow path: Directly supplied to the engine combustion chamber to maintain combustion stability; Bypass path: After flowing through the fuel cooler and cooling to a safe temperature, it returns to the fuel tank, achieving overall temperature balance and heat sink storage in the fuel tank.

[0036] Next, we will model and analyze the components of the fuel thermal management system.

[0037] (1) Fuel tank model The mass of fuel in the fuel tank is The fuel temperature inside the fuel tank is The specific heat capacity of fuel is , and This refers to the fuel mass flow rate at the fuel tank inlet and outlet. and These are the corresponding fuel inlet and outlet temperatures. , For the structural heat flux, we have the following mass and energy conservation equations: (1) The above formula can be used to update the fuel quality in the tank at different times. and fuel temperature .

[0038] (2) Pump and valve model This analysis uses an ideal constant-displacement thermohydraulic pump, meaning the pump has no flow or mechanical losses, and the flow rate is determined by the shaft speed and pump displacement. Since the fundamental change involves altering the mass flow rate of fuel, pressure variations and valve opening changes in the circuit are not considered to simplify the model. The pump output fuel volumetric flow rate is... and mass flow It can be represented as: (2) in the formula It is the pump's displacement (m³ / r). Rotational speed (r / min) The density of the fuel is (kg / m³), and the value used here is the typical density of aviation fuel. .

[0039] In this study, a proportional valve is used, and only the flow split ratio is considered. Assuming the aircraft is flying at a certain altitude, the mass flow rate to the valve is... The engine consumes fuel by mass flow rate of 100 liters. , then Figure 4 As shown, the mass flow rate of the return oil is .

[0040] (3) Heat exchanger model The mathematical model of the heat exchanger is analyzed below, using the lumped parameter method. Taking a fuel oil / lubricating oil heat exchanger as an example, the modeling and analysis of the heat exchanger are performed.

[0041] Assumption The mass flow rate of the hot edge (lubricating oil), The mass flow rate of the cold edge (fuel); and The inlet temperatures of the hot and cold sides are respectively. , and These are the outlet temperatures for the hot and cold sides, respectively. The dynamic temperature of the heat exchanger wall is a quantity that changes over time.

[0042] If the fluid does not undergo a phase change during heat exchange, the arithmetic mean of the inlet and outlet temperatures on one side can be approximated as its average temperature, i.e.: (3) The lumped parameter method is used to perform steady-state modeling of the fluids on both sides (fuel oil and lubricating oil) and dynamic modeling of the wall.

[0043] Heat dissipation at the hot edge (heat flow): (4) Cold edge heat absorption (heat flow): (5) Wall energy conservation equation: (6) In the above formula, and The convective heat transfer coefficients for the hot and cold sides are respectively, indicating the heat transfer efficiency between the fluid and the wall. and For heat transfer area, and These are the heat transfer correction factors for the hot and cold sides. For wall surface quality.

[0044] Combine equations (3) and (4), let , eliminate ,get: (7) Combine equations (3) and (5), let , eliminate ,get: (8) Substituting equations (7) and (8) into (6), we obtain the result containing only the heat exchanger input parameters. and state variables The differential equation is then solved to obtain the wall temperature, from which the outlet temperatures of the cold and hot sides of the heat exchanger can be calculated. This completes the modeling of the fuel / lubricating oil heat exchanger; the modeling method for other heat exchangers is the same.

[0045] In this preferred embodiment, the dynamic adaptive correction is implemented using a linear fitting model based on the first-order temperature coefficient, and the specific correction formula is as follows: Based on the temperature difference between the stator winding nodes and the initial temperature, and the initial stator resistance, the stator resistance is corrected and updated. The correction formula for the stator resistance is: (9) In the formula, This is the real-time resistance value of the stator winding at the current temperature. The resistance value of the stator winding at the initial temperature. To represent the temperature coefficient of resistance of copper, This represents the current temperature of the stator winding. This is the initial temperature of the stator winding.

[0046] The change in the flux linkage amplitude of a permanent magnet is reflected by the remanence of the permanent magnet, and its relationship with temperature is as follows: (10) In the formula, This represents the real-time remanence value of the permanent magnet at the current temperature. This is the initial remanence value. It is the reversible temperature coefficient of remanence in a permanent magnet. This represents the current temperature of the permanent magnet. This represents the initial temperature of the permanent magnet.

[0047] Furthermore, the real-time flux linkage and real-time remanence of the permanent magnet are linearly positively correlated, with the following relationship: (11) In the formula, It is the flux linkage amplitude of the permanent magnet at the current temperature. These are fixed coefficients related to the stator and rotor structure, number of winding turns, and pole arc coefficient of the motor.

[0048] Step 2: Establish an energy-thermal coupling interaction mechanism between the generator and the fuel thermal management system. As one of the heat sources in the fuel circuit, the generator system is coupled to the fuel circuit by establishing a heat exchange relationship with the fuel through a heat exchanger. However, the generator does not directly exchange heat with the fuel. First, it exchanges heat with the generator through lubricating oil, and then the lubricating oil transfers the heat to the fuel system.

[0049] Assuming the lubricating oil carries away heat from the heat source (generator housing) as follows: , and These are the temperatures of the lubricating oil before and after it flows through the generator. This refers to the mass flow rate of the lubricating oil. The specific heat capacity of lubricating oil. The heat transfer coefficient, For heat exchange area, Let the temperature of the heat source node be: (12) (13) Combining the two formulas, we get: (14) For simplicity of calculation, we can assume that the temperature of the lubricating oil remains constant before and after flowing through the heat source, thus treating the lubricating oil as a temperature node. This allows us to calculate the heat flow carried away by the motor. The calculation can be simplified to (15) In the formula, The heat transfer coefficient, For heat exchange area, and These are the heat source and the lubricating oil temperature, respectively.

[0050] The heat flow rate that the lubricating oil can carry away from the generator, as well as the temperature rise of the lubricating oil after flowing through the generator nodes, can be calculated. Then, heat exchange is performed with the fuel circuit using the heat exchanger model described earlier.

[0051] As shown in equation (12), when the mass flow rate of the lubricating oil is determined, the amount of heat the lubricating oil can carry away from the generator mainly depends on the temperature of the lubricating oil, or the temperature difference between the lubricating oil and the generator node. When the lubricating oil acts as the generator coolant, its temperature inevitably rises with the accumulation of working time, and the temperature difference between the lubricating oil and the generator node becomes smaller and smaller, thus reducing its heat dissipation capacity. However, fuel can carry away the heat from the lubricating oil, causing its temperature to drop, thereby restoring or maintaining its heat dissipation capacity. This process depends on the heat exchange capacity with the fuel. In engineering, the heat exchange capacity of the fuel can be changed by adjusting the mass flow rate of the fuel. When the system's heat dissipation capacity is insufficient, increasing the fuel flow rate can improve its heat exchange capacity.

[0052] Step 3: Design of FPGA solver for the thermal coupling model of the airborne power generation system In this step, the parallel processing advantage of FPGA is utilized to construct a two-layer parallel simulation architecture that includes a motor electrothermal coupling solver and a fuel thermal management solver.

[0053] 3.1 Calling the FPGA solver for the electrothermal coupling model of the motor Given the high-frequency characteristics of the electromagnetic transient process of the permanent magnet synchronous motor in this specific embodiment, the present invention uses a pre-developed high-fidelity permanent magnet synchronous motor FPGA solver module with electrothermal bidirectional coupling solution function as a specific embodiment, such as... Figure 5 As shown, this module operates in the high-speed clock domain (hundreds of nanoseconds) of the FPGA and is responsible for real-time calculation of the electromagnetic power loss of the motor and the temperature rise of internal key nodes.

[0054] To ensure the real-time performance of the model, the computation time for each model update must be less than the simulation step size. Therefore, LabVIEW's SCTL structure is used for algorithm and logic design. SCTL requires that all algorithms and logic within a loop must be calculated within a specified clock cycle; otherwise, timing conflicts will occur during the editing phase. Therefore, SCTL design must pay attention to the timing characteristics of the code to achieve high solution efficiency. SCTL also optimizes the timing of the code to meet the set loop clock. The PMSG electrothermal coupling model sets the SCTL clock to 5MHz, thus achieving real-time simulation with a step size of 200ns.

[0055] The solver receives PWM signals from an external controller and electrical angular velocities set by the host computer. Simultaneously, the feedback status from the previous moment is read from the internal registers, including... dq shaft current Winding temperature and permanent magnet temperature The online correction module for heat-sensitive parameters is invoked. and Correct current step size stator resistance and current step size flux amplitude .

[0056] Specifically, let's take an electrical model as an example.

[0057] First, and Input the equations into the flux linkage module to calculate the current time step. dq Axial flux .

[0058] Next, the converter model is based on the PWM signal and the target voltage. Calculate the three-phase phase voltages Then, it is converted to 3s / 2r transformation. dq shaft voltage .

[0059] Then , , And the signal derived from the PWM signal dq shaft voltage Substitute these values ​​into the stator voltage equation module, and through discrete difference iteration, calculate the voltage at the next time step. dq shaft current .

[0060] Calculated After a 2r / 3s inverse coordinate transformation, the three-phase currents at the next moment are obtained. This is one of the outputs of the model. At the same time, it will... As feedback, the input to the flux linkage equation module is updated for use in the next simulation cycle.

[0061] Finally, the electromagnetic torque equation module calculates the flux linkage in real time. and current And iron loss from the loss model Calculate the corrected electromagnetic torque at the current moment. This torque value serves as an output for performance evaluation.

[0062] Based on this, the focus is on building the solver logic for the fuel thermal management system that works in conjunction with this high-speed module.

[0063] 3.2 FPGA Solver Design for Fuel Oil Thermal Management System Figure 4 The diagram shows the overall structure of the fuel thermal management system, which mainly includes key components such as the fuel tank, fuel pump, proportional valve, and heat exchanger. Considering the complexity of the actual fuel thermal management system, the numerous coupled variables, and its significant nonlinear characteristics, steps 1 and 2 establish simplified mathematical models for each major component, focusing on the heat dissipation function of the fuel system and its energy-thermal coupling relationship with the generator, and analyze its energy flow characteristics. Based on this, this section constructs an FPGA-based real-time solver for simulating the dynamic behavior of the thermal management system.

[0064] First, solution modules were established for key components such as the fuel tank, fuel pump, proportional valve, and heat exchanger. Each component model was organized according to state variables, input variables, and output variables, and the state-space equations were discretized using the Euler method to obtain the state update equations and output equations in the discrete-time domain.

[0065] In this invention, the discretization process is illustrated using a fuel tank model as an example. First, the differential equations for the conservation of mass and energy of the fuel tank are established, as shown in formula (1).

[0066] Subsequently, the forward Euler method was used to transform the system of differential equations into a system of discrete difference equations suitable for iterative hardware computation.

[0067] (16) As shown in the above formula, within each simulation step, the current time is used... Known state variables (fuel mass) and temperature ) and the current input volume (entry traffic) Export flow Inlet temperature Structural heat flow The next moment can be calculated iteratively through simple operations. The state variable, i.e., the output quantity (fuel mass at the next moment). Fuel temperature at the next moment This method successfully maps the complex process of solving differential equations into efficient hardware logic, which is key to realizing real-time simulation of system-level energy-thermal coupling.

[0068] Similarly, the dynamic models of other key components in the fuel / oil thermal management system, such as pumps, proportional valves, and heat exchangers, also follow a similar processing flow.

[0069] For example, the fuel pump model receives speed commands from the controller and fuel conditions (such as temperature and density) from the fuel tank, and calculates the real-time flow rate and outlet temperature. In this embodiment, an ideal constant displacement hot hydraulic pump model is used, and the outlet temperature is consistent with the inlet temperature.

[0070] By discretizing all component models into state update equations, this invention transforms the entire complex fluid thermal network system into a set of discrete difference equations that can be efficiently executed in parallel on an FPGA, thus laying a solid foundation for real-time simulation of system-level energy-thermal coupling.

[0071] Subsequently, according to Figure 4 The system structure shown systematically organizes the input-output relationships between various components, and further designs the overall FPGA solution architecture of the thermal management system (such as...). Figure 6 (As shown). This system comprises two main circulation branches: one is the fuel circulation, where the fuel pump draws fuel from the fuel tank at a set speed, forming a closed loop. The speed of the fuel pump is the core control variable of the system, determining the mass flow rate in the fuel circulation and thus affecting the heat dissipation capacity of the fuel system. In practical applications, aviation fuel systems are typically used not only for generator lubricating oil cooling but also for cooling other airborne equipment and branch systems. To focus on the energy-thermal coupling relationship between the fuel circulation and the generator studied in this invention, the non-core heat load is simplified in modeling, using a single heat exchange unit to equivalently represent the heat power absorbed by the fuel from the external system. After absorbing heat, the fuel exchanges heat with the lubricating oil in the fuel / lubricating oil heat exchanger, and then exchanges heat with other heat exchange units. It then flows through a proportional valve, which regulates the flow rate distribution to the combustion chamber and the return fuel branch. The fuel in the return fuel branch is cooled by a cooling device and then flows back to the fuel tank, thus completing the circulation.

[0072] The second circulation branch is the lubricating oil circuit. To simplify the modeling, it is assumed that all the heat carried away by the lubricating oil from the generator is transferred to the fuel circuit, and energy exchange between the lubricating oil system and other components is no longer considered. The heat exchange between the generator and the lubricating oil occurs in the lubricating oil heating module (or generator cooling module).

[0073] The motor model calculates the heat dissipation power input to the lubricating oil circuit model in real time, which is used to heat the flowing lubricating oil, and calculates the temperature rise and outlet temperature of the lubricating oil. Simultaneously, the lubricating oil circuit model feeds back its real-time flow rate and inlet temperature as dynamic cooling boundary conditions to the generator's thermal network model. Through this two-way data interaction, a tight coupling is achieved between generator heat generation and fuel / lubricating oil system heat dissipation.

[0074] Through such Figure 6 The FPGA solution structure shown clearly defines the physical interfaces and logical interactions between the components of the thermal management system, which helps improve the modularity and maintainability of the model. At the same time, this structure has good scalability and uniformity, enabling flexible upgrades and expansions of the component models while maintaining the consistency of the system framework.

[0075] In this invention, the fuel / lubricating oil thermal management solver operates in a single simulation step ( The complete iterative calculation sequence within the given timeframe strictly follows the physical flow path and energy transfer relationship of the fluid, as detailed below: Phase 1: External Input and Status Reading 1. Input Acquisition: Obtain the engine fuel consumption rate command for the current step size from the host computer. Oil pump control commands Fuel pump control commands and proportional valve opening command; obtain the generator heat source power at the current moment from the motor solver. .

[0076] 2. Loading historical state: Reading the previous time step from the register ( The calculated states of each key node, including the fuel mass in the fuel tank. With temperature And the temperature of each pipeline node.

[0077] Phase 2: Solution of Lubricating Oil Circuit 3. Lubricating oil flow rate calculation: based on lubricating oil pump control commands. The total mass flow rate of the lubricating oil circuit is calculated by calling the pump model. .

[0078] 4. Temperature rise of lubricating oil on the generator side: (This refers to the total flow rate of lubricating oil.) and generator heat source power Input is sent to the lubricating oil heating module (generator heat exchanger interface). The lubricating oil inlet temperature from the previous moment is used. Calculate the current temperature of the lubricating oil outlet after it flows through the generator. .

[0079] 5. Feedback to the motor model: The calculated lubricating oil flow rate is fed back to the motor model. and its inlet temperature As a cooling boundary condition, it is fed back to the generator electrothermal coupling solver in real time.

[0080] Phase 3: Fuel Circuit Calculation 6. Fuel Flow Calculation: The total outlet flow rate of fuel flowing out of the fuel tank. fuel pump control command The decision can also be made based on the engine's fuel consumption rate. and bypass flow to the cooling module A joint decision.

[0081] 7. Fuel temperature rise on the heat exchanger side: High-temperature lubricating oil ( , ) and low-temperature fuel from the fuel tank ( , Simultaneously input into the fuel / oil heat exchanger model. Calculate the heat exchange to obtain the heated fuel outlet temperature. and the outlet temperature of the cooled lubricating oil (This temperature will become the inlet temperature of the lubricating oil circuit in the next cycle.)

[0082] 8. Fuel flows through other units: the heated fuel ( , By sequentially applying other heat exchange unit models and calculating their outflow temperatures based on their heat exchange characteristics, the final outflow temperature can be determined. .

[0083] Phase 4: Fuel Diversion and Tank Status Update 9. Proportional valve flow splitting: The fuel flow after passing through other heat exchange units is split into two paths according to the proportional valve opening command: one path directly supplies the engine (flow rate...). The other path enters the fuel cooling module (flow rate). ).

[0084] 10. Bypass Cooling and Return Fuel: The bypass fuel entering the fuel cooling module is cooled to reach the final return fuel temperature. .

[0085] 11. Fuel Tank Status Update: Update return fuel flow rate With temperature (From the fuel cooling module), engine consumption, and structural heat flow from other external heat sources. Substitute these values ​​into the discrete difference equation of the fuel tank, and iteratively calculate the fuel tank state at the next moment. and ).

[0086] Phase 5: State Storage 12. Write all state variables for the next time step into the register for use in the next simulation cycle.

[0087] Because the dynamic process of a thermal management system has greater thermal inertia than the electromagnetic transient process of a motor, its response speed is slower, thus eliminating the need for nanosecond-level simulation step sizes. Therefore, the FPGA solver design for the thermal management system has a large time margin. This invention employs a 2μs step size to design an FPGA solver for the thermal management system.

[0088] Step 4: Establish a multi-task collaborative hardware-in-the-loop test platform A multi-task hardware-in-the-loop test platform based on the energy-thermal coupling model of the airborne power generation system, such as... Figure 7 As shown, it consists of a real-time simulator that runs a real-time model, a controller that runs a control strategy, a host computer, an output adapter module, and an oscilloscope. The output adapter module is either an I / O module or a signal conditioning module.

[0089] The host computer serves as the human-computer interaction and management center of the entire testing system. It establishes a two-way communication connection with the real-time simulator through a standard Ethernet interface. It is responsible for sending configuration parameters of the simulation model (such as motor parameters and initial state of the thermal management system), flight mission profile commands (such as speed curves and load sequences), and fault injection commands to the real-time simulator. In addition, the host computer can receive and display operating data from the simulator in real time, such as motor temperature, bus voltage, and lubricating oil flow, and display waveform charts on the front panel for testers to perform online analysis and recording.

[0090] The real-time simulator is the core computing platform of this invention. It integrates a high-performance FPGA chip and runs a multi-task collaborative model of the airborne power generation system with thermal coupling. Utilizing the parallel pipeline architecture of the FPGA, the real-time simulator runs the motor electrothermal coupling solver and the fuel thermal management solver in real time, completing the synchronous solution of electromagnetic transients at the nanosecond level and thermofluid dynamics at the microsecond level. Furthermore, the real-time simulator acquires externally input control commands to drive the internal converter model; simultaneously, it converts the calculated motor current, voltage, temperature, and other state variables into digital quantities and outputs them to the output converter module.

[0091] The output conversion module serves as a physical signal conditioning and conversion interface, connecting to the real-time simulator via a high-speed internal bus. It is responsible for the mutual conversion between digital and analog signals. The output conversion module converts digital quantities calculated by the real-time simulator, such as three-phase current, bus voltage, and motor node temperature, into standard analog voltage signals, which are then output to the controller and oscilloscope. Simultaneously, it receives cooperative control commands from the controller, performs level conversion and isolation processing, and then transmits them to the real-time simulator.

[0092] The controller is an integrated controller for the airborne power generation system, connected to the output adapter module via a dedicated cable. The controller collects feedback signals from the output adapter module, runs internal control algorithms, calculates in real time, and outputs coordinated control commands to adjust the power generation system state. Simultaneously, it monitors whether feedback signals exceed limits, and executes corresponding protection strategies when simulated over-temperature or over-voltage faults are detected.

[0093] An oscilloscope, as an external independent observation device, is connected in parallel to the analog output port of the output adapter module. It is used to capture and display high-frequency dynamic waveforms of key nodes such as motor phase current and bus voltage ripple in real time, to help verify the quality of the controller output control commands and the system response characteristics under transient conditions, and to provide independent physical waveforms for test results.

[0094] The specific connection relationships are as follows: (1) Control signal transmission: The digital input / output (DIO) interface of the controller is connected to the DIO interface of the output adapter module through a signal line. This is used to transmit the coordinated control commands (including PWM drive signals, pump / valve control signals, etc.) generated by the controller in real time to the FPGA solver inside the simulator in real time to drive the power generation system model.

[0095] (2) Multi-physics domain state feedback: The analog input / output (AIO) interfaces of the output converter module are connected in parallel to the AIO interface of the controller and an external oscilloscope. The multi-physics domain feedback state calculated by the internal model of the real-time simulator is converted into a standard analog voltage signal. On the one hand, it is used as a feedback signal input to the controller to form a complete hardware-in-the-loop closed loop; on the other hand, it is transmitted to the oscilloscope for testers to observe and analyze the waveform in real time.

[0096] (3) Human-computer interaction and task management: The host computer communicates bidirectionally with the real-time simulator FPGA through the Ethernet interface to issue flight mission profile instructions and receive simulation process data in real time.

[0097] Step 5: Conduct system-level energy-thermal coupling synergy testing and performance evaluation Based on the completion of the above hardware-in-the-loop test platform, this step loads the preset flight mission profile and operating parameters into the real-time simulator through the host computer software to conduct hardware-in-the-loop test of the airborne power generation system's thermal coupling multi-mission.

[0098] This section uses PMSG's "motor temperature rise characteristic test" and "pulse load profile test" as specific examples to illustrate the test process and phenomena. It should be noted that the above two test cases are only used to illustrate the effectiveness of the test method of this invention and are not intended to limit the type of motor or the test scope of this invention. Based on the test platform and multi-task collaborative architecture constructed by this invention, it is possible to freely configure and execute, according to actual needs, tests including but not limited to: full flight envelope traversal tests, cooling medium boundary limit tests, and extreme fault injection tests such as oil pump stoppage and low fuel level.

[0099] 5.1 Motor Temperature Rise Characteristics Test (1) Temperature rise test under rated load conditions Test conditions: The PMSG operates at its rated speed of 60,000 r / min, and its three-phase output is rectified to 540V DC via PWM. The DC bus carries a rated load (120kW) with a load current of 222A. The PMSG stator is cooled by lubricating oil, with a fuel system flow rate of 1.34 kg / s. The rotor is air-cooled at a wind speed of 10 m / s, and the initial temperature is 25℃ room temperature. The safe operating temperature of the motor windings is 180℃, and the safe operating temperature of the permanent magnets is 120℃. The temperature rise characteristics of the PMSG at different locations under rated load are tested.

[0100] Test results are as follows Figure 8 The diagram illustrates the temperature rise of the permanent magnet, stator teeth, stator yoke, and stator windings of the motor under rated load. The temperature rise of different parts of the motor, from highest to lowest, is: stator windings, stator teeth, permanent magnets, and stator yoke. The stator windings have the highest temperature, balancing at 80.7℃, while the stator yoke has the lowest temperature, balancing at 73.8℃. Therefore, under rated load, the temperatures of all parts of the motor are within their safe operating temperatures.

[0101] (2) Temperature rise test under 2 times overload condition Test conditions: The PMSG operates at its rated speed of 60,000 r / min, and its three-phase output is rectified to 540V DC via PWM. The DC bus carries a load of twice the rated load (240kW), with a load current of 444A. The PMSG stator is cooled by lubricating oil, with a fuel system flow rate of 1.34 kg / s. The rotor is air-cooled at a wind speed of 10 m / s, and the initial temperature is 25℃ room temperature. The safe operating temperature of the motor windings is 180℃, and the safe operating temperature of the permanent magnets is 120℃. The temperature rise characteristics of the PMSG at different locations under a 2x overload were tested, and the test results are as follows: Figure 9 As shown in (a). With the fuel flow rate changed to 2.68 kg / s and other operating conditions remaining unchanged, the temperature rise curve of the PMSG under 2x overload was tested as follows. Figure 9 As shown in (b).

[0102] Depend on Figure 9(a) It can be seen that under the baseline heat dissipation conditions, if the generator experiences a 2x overload, the stator winding will reach 180°C after 34 seconds, and the permanent magnet temperature will reach 95°C. To increase the overload operating time of the motor, the fuel circulation flow rate can be increased to improve the motor's heat dissipation capacity. Figure 9 (b) It can be seen that when the fuel flow rate is increased to 2.68 kg / s, the stator winding of the generator reaches 180°C after 52 seconds under a 2x overload condition, and the permanent magnet reaches 113°C. This shows that coordinated regulation between the generator and the fuel thermal management system can significantly improve the generator's overload operating time, which is of great significance for enhancing the system's short-term high-power energy supply capacity.

[0103] 5.2 Pulse Load Profile Test This test uses a pulsed load task profile to drive the PMSG thermal coupling model and the fuel thermal management system model in coordinated operation. The profile duration is 1400s, the bus load power is 120kW, the pulsed load is applied starting at 500s and removed at 845s. The pulsed load power is 120kW, the pulse width is 5s, and the pulse interval is 10s. The PMSG operates at its rated speed of 60000r / min, and its three-phase output is rectified to 540V DC via PWM. The PMSG stator is cooled by lubricating oil, the fuel system flow rate is 1.34kg / s, and the rotor is air-cooled at a wind speed of 10m / s. The initial temperature is 25℃ room temperature. The safe temperature range for the motor windings is 180℃, and the safe temperature range for the permanent magnets is 120℃.

[0104] Figure 10 The test results for the pulsed load profile show that the generator operates under a 2x overload condition. During the pulsed load loading phase, the stator winding current increases significantly, leading to increased heat generation and a significant temperature rise. The stator teeth, stator yoke, and permanent magnets all experience noticeable temperature increases. During the pulse interval, the current returns to its rated value, and the stator temperature decreases, but the decrease is less than the increase. As the pulse continues, the stator and rotor temperatures gradually accumulate to a new equilibrium point, resulting in a significant increase in the motor's temperature rise. After the pulsed load ends, the load current returns to its rated value, and the motor recovers to its rated operating temperature after a period of time.

[0105] Depend on Figure 10 It can be seen that the thermal balance of the motor after pulse load is applied is greatly related to the pulse interval. If the pulse interval is too short, the accumulated heat of the motor cannot be dissipated in time, causing the motor temperature to rise continuously and eventually exceed the motor's safe operating range.

[0106] Keeping the pulse load power at 120kW and the pulse width at 5s constant, shortening the pulse interval to 5s, and maintaining the fuel circulation flow rate unchanged, the pulse load profile is retested, and the following results can be obtained: Figure 11 (b) shows the generator temperature rise. It can be seen that, without changing the generator's heat dissipation conditions, shortening the pulse interval to 5 seconds causes the generator temperature to rise continuously during pulsed power load operation until it exceeds its safe operating temperature. In the model, demagnetization of the permanent magnets causes a decrease in the generator's output power, and the bus voltage cannot be maintained at 540V. For example... Figure 11 As shown in (c), the bus voltage control has failed at this point.

[0107] To further verify the coordinated control effect of the generator and thermal management system under a 5-second pulse interval, the fuel circulation flow rate was increased to 2.68 kg / s, and the pulse load task profile was re-simulated under the condition that other operating conditions remained unchanged. The results show that... Figure 12 The graph shows the motor temperature rise and bus voltage variation curves. As can be seen from the graph, the larger fuel flow rate increases the generator's heat dissipation power, allowing the generator temperature to drop rapidly during the 5-second pulse interval, thus keeping the motor temperature rise within a safe range. Simultaneously, during the pulse load operation phase, the bus voltage is relatively stable, and voltage disturbances caused by the pulse load can be adjusted within a short time, ensuring that the generator's power supply is not affected by the pulse power load.

[0108] Comprehensive comparison Figure 8 - Figure 12 It is evident that the energy-thermal coupling multi-task collaborative testing method for airborne power generation systems constructed based on this invention can safely and efficiently explore collaborative control strategies between the generator and the thermal management system, thereby constructing an energy-thermal collaborative management and control method. This improves the operating time of pulsed power loads and shortens pulse intervals, thus maximizing the capabilities of airborne equipment. It can provide a low-cost, high-fidelity, and risk-free system-level collaborative testing method for verifying controllers under extreme fault conditions.

[0109] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A method for thermally coupled multi-task collaborative testing of an airborne power generation system, characterized in that, The specific steps include: Step 1, Test Preparation: Configure the pre-built airborne power generation system energy-thermal coupling model in the FPGA real-time simulator. The energy-thermal coupling model includes a motor electrothermal coupling sub-model and a fuel thermal management sub-model, and there is a two-way closed-loop coupling relationship between the two: electrothermal generation → thermal parameter change → parameter conversion to electrothermal generation. Connect the FPGA real-time simulator with the real controller, host computer, output adapter module and oscilloscope to form a hardware-in-the-loop test platform. Step 2, Test Loading: Load the preset flight mission profile and operating parameters into the FPGA real-time simulator through the host computer. The flight mission profile includes typical flight phases such as takeoff, cruise, maneuvering and landing. The operating parameters include time-varying speed commands, load power requirements and dynamic fault injection commands. Step 3, Test Run: Start the test. The energy-thermal coupling model in the FPGA real-time simulator runs in real time under the drive of operating parameters. The motor electrothermal coupling sub-model calculates the electromagnetic transient process and power loss of the motor in real time with a first simulation step size. The fuel thermal management sub-model calculates the fuel / lubricating oil thermal inertia dynamic process in real time with a second simulation step size. The two sub-models achieve bidirectional coupling data exchange through cross-clock domain synchronization. The real controller collects the feedback signal output by the FPGA real-time simulator in real time, calculates and outputs cooperative control commands according to the preset control strategy, and feeds back the cooperative control commands to the FPGA real-time simulator to drive the energy-thermal coupling model to respond. Step 4, Test Data Acquisition: Real-time acquisition and recording of the electrothermal dynamic response data output by the energy-thermal coupling model during the test operation. The electrothermal dynamic response data includes dynamic curves of multiple parameters such as motor winding temperature, permanent magnet temperature, phase current, bus voltage, lubricating oil outlet temperature, and fuel tank temperature over time. Step 5, Test and Evaluation: Compare the collected electrothermal dynamic response data with the preset safety threshold to determine the effectiveness of the thermal safety boundary and control strategy of the system under the conditions of limited cold source and multi-physical domain coupling, and generate test and evaluation results.

2. The method for thermal coupling multi-task collaborative testing of an airborne power generation system according to claim 1, characterized in that: The motor electrothermal coupling sub-model includes an electrical model, a loss model, a thermal network model, and an electrical parameter correction module, wherein: The electrical model calculates the electromagnetic transient process of the motor based on the current electrical parameters and outputs current, frequency, and flux linkage information to the loss model. The loss model calculates copper loss based on the current output by the electrical model and the stator resistance updated in real time, calculates iron loss based on frequency and magnetic flux density, and inputs the calculated total loss power as a heat source into the thermal network model. The thermal network model calculates the real-time temperature of each key node inside the motor using the thermal equivalent circuit method based on the total power loss and preset heat capacity and thermal resistance parameters, and outputs the real-time temperature to the electrical parameter correction module. The electrical parameter correction module, based on the received real-time temperature, calls a preset temperature-parameter mapping relationship to correct the thermistor electrical parameters online. The thermistor electrical parameters include stator resistance and permanent magnet flux linkage. The corrected electrical parameters are then fed back to the electrical model to replace the original parameters in the calculation of the next moment, thereby realizing information closed loop and dynamic collaboration.

3. The method for thermal coupling multi-task collaborative testing of an airborne power generation system according to claim 2, characterized in that: The fuel thermal management subsystem sub-model includes a fuel tank model, a regulating pump model, a fuel distribution valve model, and a heat exchanger model, wherein: The fuel tank model dynamically updates the temperature and remaining mass of fuel in the tank based on the return oil heat injection and fuel consumption, simulating the process of heat sink capacity gradually decreasing with fuel consumption under limited cold source conditions. The regulating pump model adjusts the fuel mass flow rate in the fuel line according to the speed control command, providing a variable fluid boundary; The heat exchanger model includes at least a fuel oil / lubricating oil heat exchanger, which is used to receive the lubricating oil heat source from the electric motor electrothermal coupling sub-model, and to perform countercurrent heat exchange using the temperature difference between fuel oil and lubricating oil, and to calculate the outlet temperature after the lubricating oil cools down and the temperature after the fuel oil heats up. The fuel distribution valve model dynamically distributes the fuel flowing through the heat exchanger model into two paths according to the engine operating condition command and temperature feedback signal: one path supplies the engine combustion chamber, and the other path returns to the fuel tank after being cooled by the fuel cooler, forming a return bypass to regulate the overall temperature of the fuel tank.

4. The method for thermal coupling multi-task collaborative testing of an airborne power generation system according to claim 3, characterized in that: The bidirectional closed-loop coupling relationship of electrothermal generation → thermal parameter change → parameter conversion to electrothermal conversion: First-stage heat transfer: The sum of stator copper loss and iron loss calculated in real time by the motor electrothermal coupling sub-model is used as the heat flow input and transferred to the lubricating oil circuit to calculate the lubricating oil temperature rise; Second-stage heat transfer: The heated lubricating oil is introduced into the fuel / lubricating oil heat exchanger model to exchange heat with the low-temperature fuel, and the lubricating oil outlet temperature and fuel temperature rise are calculated. Parameter feedback: The cooled lubricating oil outlet temperature and real-time flow rate are fed back to the motor electrothermal coupling sub-model as its cooling boundary conditions to participate in the calculation of the internal temperature rise rate of the motor. Parameter correction: Based on the real-time monitoring of the temperature of key motor nodes, the temperature change parameters, including stator resistance and permanent magnet flux linkage, are corrected online, and the updated parameters are fed back to the motor solver.

5. The method for thermal coupling multi-task collaborative testing of an airborne power generation system according to claim 4, characterized in that: The online correction of the temperature change parameters includes: The correction formula for stator resistance is: In the formula, This is the real-time resistance value of the stator winding at the current temperature. The resistance value of the stator winding at the initial temperature. To represent the temperature coefficient of resistance of copper, This represents the current temperature of the stator winding. This is the initial temperature of the stator winding; The change in the flux linkage amplitude of a permanent magnet is reflected by the remanence of the permanent magnet, and its relationship with temperature is as follows: In the formula, This represents the real-time remanence value of the permanent magnet at the current temperature. This is the initial remanence value. It represents the reversible temperature coefficient of remanence in a permanent magnet. The current temperature of the permanent magnet. This is the initial temperature of the permanent magnet; The real-time flux linkage and real-time remanence of permanent magnets are linearly positively correlated, with the following relationship: In the formula, It is the flux linkage amplitude of the permanent magnet at the current temperature. These are fixed coefficients related to the stator and rotor structure, number of winding turns, and pole arc coefficient of the motor.

6. A method for constructing an FPGA solver for an airborne power generation system thermal coupling model, wherein the FPGA solver for the airborne power generation system thermal coupling model is used to execute the multi-task collaborative testing method for airborne power generation system thermal coupling as described in any one of claims 1-5; characterized in that, The specific steps are as follows: Step 1: Build a multi-physics domain real-time model library: Establish a component-level model library that includes an electric motor electrothermal coupling sub-model and a fuel thermal management subsystem sub-model; The motor electrothermal coupling sub-model has a built-in mapping relationship between motor electromagnetic parameters and temperature, as well as a mapping relationship between electromagnetic loss and electrical state, and a reserved parameter update interface for dynamically updating electrical parameters during simulation to reflect thermal decay characteristics, while also being able to output real-time power loss. The fuel thermal management subsystem sub-model includes at least a fuel tank model, a pump and valve model, and a heat exchanger model. The fuel tank model is used to describe the dynamic temperature rise process of fuel as the return oil heat is injected. The pump and valve model is used to simulate the dynamic response characteristics of fuel and lubricating oil flow rate adjusted by speed or control command. The heat exchanger model is used to calculate the heat transfer efficiency between lubricating oil and fuel and between fuel and other branches in real time. Step 2: Establish an energy-thermal coupling interaction mechanism: Define a two-stage heat transfer path and closed-loop feedback logic between the generator heat source-lubricating oil-fuel mixture sub-model and the fuel thermal management sub-model. Step 3: Construct an electrothermal coupled FPGA solver for the motor: The electric motor electrothermal coupling sub-model and its energy-thermal coupling interaction mechanism are mapped to the first clock domain of the FPGA, and the first clock domain is configured to run at the first simulation step size. The FPGA pipeline architecture is used to realize the real-time calculation of the electrical model, loss model, thermal network model and electrical parameter correction module in the electric motor electrothermal coupling sub-model, and output the real-time power loss and motor state parameters of the motor. Step 4: Construct the FPGA solver for the fuel thermal management system: The fuel thermal management subsystem sub-model and its energy-thermal coupling interaction mechanism are mapped to the second clock domain of the FPGA, and the second clock domain is configured to run at a second simulation step size, which is larger than the first simulation step size. First, the state space of the oil tank model, pump and valve model and heat exchanger model are reconstructed to clarify the state variables, input variables and output variables of each model. The Euler method is used to transform the continuous domain differential equations of each model into discrete time domain state update equations and output equations; By utilizing the parallel pipeline technology of FPGA, the computational logic of each model is mapped to hardware logic modules that are executed in parallel. Based on the physical topology of the fuel thermal management system, the input-output connection relationships between each model are defined to form the data flow network of the overall solver; Step 5: Design a cross-clock domain synchronization module and integrate the solver: A cross-clock domain synchronization module is set up between the FPGA solver for the electric motor electrothermal coupling and the FPGA solver for the fuel thermal management system to realize deterministic data exchange between the two clock domains; The electric motor electrothermal coupling FPGA solver, the fuel thermal management system FPGA solver, and the cross-clock domain synchronization module are integrated on the same FPGA chip to form a unified energy-thermal coupling model FPGA solver. This solver can realize simultaneous, parallel, hard real-time bidirectional coupling calculation of electromagnetic transient processes and fluid thermal inertial processes during operation.

7. The method for constructing an FPGA solver for a thermally coupled model of an airborne power generation system according to claim 6, characterized in that: The generator heat source-lubricating oil-fuel two-stage heat transfer path and closed-loop feedback logic specifically include: The total power loss calculated in real time by the electric-thermal coupling sub-model of the motor is transmitted to the lubricating oil circuit as a heat flow input to calculate the lubricating oil temperature rise. The heated lubricating oil is introduced into the fuel / lubricating oil heat exchanger model to exchange heat with the low-temperature fuel; The cooled lubricating oil outlet temperature and real-time flow rate are fed back to the motor electrothermal coupling sub-model as its dynamic cooling boundary conditions. The electric motor electrothermal coupling sub-model calculates the internal temperature rise of the motor based on the boundary conditions, and calls the parameter update interface to dynamically and adaptively correct the stator resistance and permanent magnet flux linkage, forming a system-level closed-loop coupling of electrothermal generation → thermal parameter change → parameter-electrical modification.

8. The method for constructing an FPGA solver for a thermally coupled model of an airborne power generation system according to claim 7, characterized in that: The cross-clock domain synchronization module transmits the real-time power loss calculated by the motor electrothermal coupling FPGA solver to the fuel thermal management system FPGA solver as the heat source input boundary, and transmits the cooling medium state parameters calculated by the fuel thermal management system FPGA solver to the motor electrothermal coupling FPGA solver as the dynamic cooling boundary condition.

9. A thermally coupled multi-task collaborative test platform for an airborne power generation system using the method of any one of claims 6-8, characterized in that, include: An FPGA real-time simulator is configured with an onboard power generation system energy-thermal coupling model. The energy-thermal coupling model includes a motor electrothermal coupling sub-model and a fuel thermal management sub-model, and the two have a bidirectional closed-loop coupling relationship of electrothermal generation → thermal parameter change → parameter conversion to electricity. The FPGA real-time simulator is used to calculate the electrothermal dynamic process in real time with a first simulation step size and a second simulation step size, respectively, and output feedback signals. The real controller is connected to the FPGA real-time emulator and is used to collect the feedback signal, run the control strategy, and output cooperative control commands to the FPGA real-time emulator. The host computer is connected to the FPGA real-time simulator and is used to load flight mission profiles and operating parameters, and to receive, display and store the electrothermal dynamic response data collected during the test. The output adapter module is connected between the FPGA real-time emulator and the real controller and is used for signal conditioning and interface conversion. The FPGA real-time simulator, the real controller, and the host computer form a closed-loop test circuit, outputting test evaluation results.

10. A thermally coupled multi-task collaborative test platform for an airborne power generation system for the method according to claim 9, characterized in that: It also includes an oscilloscope, which is connected in parallel to the analog output port of the output adapter module. The oscilloscope is used to capture and display the high-frequency dynamic waveforms of key nodes in real time, to help verify the quality of the control commands output by the real controller and the response characteristics of the system under transient conditions, and to provide independent physical waveforms for the test results.

Images