A method and apparatus for designing a control law for a thermal management type combined power plant

Abstract

This application provides a method and apparatus for designing control laws for a thermally managed combined power unit, belonging to the field of aerospace electromechanical systems. The method includes: establishing mathematical models for the core components of the thermally managed combined power unit; constructing an initial steady-state simulation model by combining flow rate, pressure, and energy conservation relationships; revising the initial steady-state simulation model to obtain a faithful steady-state simulation model; building a transient simulation model based on the faithful steady-state simulation model, incorporating mode switching logic, fuel supply strategy, control target selection rules, and system safety boundaries, and designing a basic control law; employing a particle swarm optimization algorithm, using the fuel control PID parameters of the transient simulation model as the optimization object, and using mode switching time, system overshoot, and power turbine exhaust temperature as optimization indicators to optimize the control parameters of the transient simulation model; updating the optimized parameters to the transient simulation model to obtain a transient simulation model of the thermally managed combined power unit.

Description

Technical Field

[0001] This application relates to the field of aviation electromechanical technology, and in particular to a method and apparatus for designing control laws for thermal management type combined power units. Background Technology

[0002] Aircraft electromechanical systems, encompassing the aircraft's execution, support, and secondary energy systems, directly impact its overall performance. However, traditional aircraft electromechanical systems suffer from low power-to-weight ratios, poor energy efficiency, and complex maintenance, making them insufficient to support future aircraft performance enhancements. Thermally managed combined power plants, through highly integrated design of power generation / conversion devices and thermal management systems, can efficiently manage system waste heat and optimize power performance while providing mechanical and electrical energy. This results in improved overall energy efficiency, reliability, and lightweight design, making it a key component for future aircraft development. Therefore, designing suitable control laws for this device is crucial for fully leveraging its integrated advantages, ensuring stable and efficient operation, and is a vital step in upgrading aircraft electromechanical systems.

[0003] Currently, the design of control laws for thermal management combined propulsion systems largely relies on simulation systems built from single physical models or traditional fixed-gain control methods. While a basic architecture for aircraft adaptive power and thermal management systems has been proposed, related control law designs are only preliminary explorations of basic power and thermal management coordination, lacking refined design schemes tailored to the specific characteristics of the system. The overall approach remains primarily based on traditional simulation modeling combined with conventional control algorithms. Existing design methods are ill-suited to the operational requirements of the system, revealing numerous problems and shortcomings: First, control algorithms struggle to balance multiple control objectives such as temperature and efficiency, and the strong coupling of variables within the system means that a single physical model cannot accurately reflect the system's full-condition characteristics. Second, the system involves multiple operating modes, and existing methods lack specific design for mode switching, failing to meet the stringent requirements for overshoot and switching time during transitions. Third, fixed-gain control methods cannot adapt to the system's full-envelope operation requirements, resulting in insufficient control performance stability. Furthermore, the limited accuracy of traditional simulation modeling leads to high investment in subsequent physical experiments and testing, resulting in high R&D costs and low efficiency.

[0004] Therefore, there is an urgent need for a method to achieve refined and intelligent design of control laws for thermal management combined power units, effectively improve the adaptability of control laws to complex operating conditions of the unit, and ensure the stability and control accuracy of the unit's operation. Summary of the Invention

[0005] In view of this, this application provides a control law design method and apparatus for thermal management type combined power units, so as to realize the refined and intelligent design of control laws for thermal management type combined power units, effectively improve the adaptability of control laws to complex operating conditions of the unit, and ensure the stability and control accuracy of the unit operation.

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

[0007] The first aspect of this application provides a method for designing control laws for a thermally managed combined power unit, the method comprising:

[0008] A mathematical model is established for the core components of a thermal management type combined power unit. An initial steady-state simulation model is constructed by combining the flow rate, pressure, and energy conservation relationships. The initial steady-state simulation model is then corrected using a particle swarm optimization algorithm combined with experimental data to obtain a faithful steady-state simulation model.

[0009] Based on the aforementioned high-fidelity steady-state simulation model, a transient simulation model is built, incorporating mode switching logic, fuel supply strategy, control target selection rules, and system safety boundaries, to design the basic control law for the transient simulation model;

[0010] The particle swarm optimization algorithm is used to optimize the control parameters of the transient simulation model by taking the fuel control PID parameters of the transient simulation model as the optimization object and the mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators. The optimized parameters are then updated to the transient simulation model to obtain the transient simulation model of the thermal management type combined power unit.

[0011] A second aspect of this application provides a control law design device for a thermal management type combined power unit, the device comprising a construction module, a design module, and an optimization module;

[0012] The construction module is used to establish a mathematical model for the core components of the thermal management type combined power unit, construct an initial steady-state simulation model by combining the flow rate, pressure and energy conservation relationship, and use the particle swarm optimization algorithm combined with experimental data to correct the initial steady-state simulation model to obtain a faithful steady-state simulation model.

[0013] The design module is used to build a transient simulation model based on the fidelity steady-state simulation model, and incorporates mode switching logic, fuel supply strategy, control target selection rules and system safety boundaries to design the basic control law of the transient simulation model.

[0014] The optimization module is used to employ a particle swarm optimization algorithm, taking the fuel control PID parameters of the transient simulation model as the optimization object, and using mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators to iteratively optimize the control parameters of the transient simulation model, and update the optimized parameters to the transient simulation model to obtain a thermal management type combined power unit transient simulation model.

[0015] The control law design method and apparatus for thermally managed combined power units provided in this application, firstly, provide precise quantitative basis at the component level for the simulation model through the mathematical model of core components, while the conservation relationships of flow rate, pressure, and energy ensure that the initial steady-state simulation model conforms to the physical mechanism of device operation, avoiding the problem of the model deviating from actual physical laws; based on this, the efficient optimization capability of the particle swarm optimization algorithm is utilized, combined with the experimental data of the physical prototype of the thermally managed combined power unit, to iteratively correct the core parameters of the initial steady-state simulation model. The resulting faithful steady-state simulation model can highly approximate the actual operating characteristics of the device and accurately reproduce the core parameters. The nonlinear response, parameter coupling relationship, and steady-state characteristics of the core components under a single mode are analyzed. By prioritizing steady-state and transient states, and ensuring model accuracy before designing control laws, a near-realistic control design environment is provided for subsequent basic control law design and parameter optimization. This effectively solves the problem of insufficient accuracy of traditional simulation models leading to a disconnect between control law design and actual experiments. It enables the subsequently designed control laws to adapt to the complex nonlinear characteristics of the device operation in advance and to handle dynamic parameter changes during mode switching. This lays a solid foundation for the engineering practicality of the control laws and avoids problems such as control failure and response lag caused by model distortion in practical applications.

[0016] Secondly, the mode switching logic, fuel supply strategy, control target selection rules, and system safety boundary together constitute the "safety barrier" of the control law design. It does not simply rely on data-driven or algorithmic optimization, but rather ensures the physical rationality and operational safety of control decisions through deterministic rule constraints: the mode switching logic clarifies the legal transition paths, specific operating steps, and anomaly handling rules between each operating mode, avoiding system instability caused by unauthorized mode switching; the fuel supply strategy combines the characteristics of different operating modes to configure a differentiated control architecture, ensuring the accuracy and stability of fuel supply and preventing combustion anomalies caused by sudden fuel supply changes; the control target selection rules clarify the core of regulation based on the functional positioning of each mode, ensuring that the control direction aligns with the device's operational needs; the system safety boundary presets the allowable range and control limits of key parameters, eliminating dangerous conditions such as over-temperature, over-pressure, and surge. By deeply integrating physical mechanism rules with the simulation model, it effectively avoids unreasonable and unsafe control commands that may be generated by purely data-driven methods due to data bias or insufficient samples, fundamentally ensuring the safety and reliability of the control law's operation and ensuring that every decision of the control law conforms to the physical laws and safety requirements of the device's operation.

[0017] Thirdly, by introducing particle swarm optimization (PSO) twice for core optimization, a dual optimization system of "model correction + parameter optimization" is formed, further improving the adaptability and robustness of the control law: Firstly, PSO is used to correct the initial steady-state simulation model, leveraging its advantages of global optimization and fast convergence speed to quickly find the model parameter combination with the smallest error compared to experimental data, efficiently improving model fidelity; secondly, PSO is used for iterative optimization of the PID parameters of the fuel control system in the transient simulation model, using mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators, targeting different operating conditions. The system intelligently optimizes PID parameters under different modes and operating conditions, enabling offline learning and adaptive adjustment of control parameters through algorithm iteration. This effectively addresses various uncertainties in the operation of thermally managed combined power units, solving the problems of traditional fixed-parameter control laws being unable to adapt to full-envelope operation requirements and lacking robustness. The optimized control law not only meets the requirements of multi-objective optimization indicators, shortens mode switching time, reduces system overshoot, and controls the exhaust temperature of the power turbine within a safe range, but also maintains stable control performance under various complex and uncertain operating conditions, exhibiting stronger robustness and adaptability. Attached Figure Description

[0018] Figure 1 A flowchart of the control law design method for a thermal management type combined power unit provided in Embodiment 1 of this application;

[0019] Figure 2 This is a schematic diagram of the control law design device for a thermal management type combined power unit provided in Embodiment 2 of this application. Detailed Implementation

[0020] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0021] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0022] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0023] The following specific embodiments are given to illustrate the technical solution of this application in detail.

[0024] Figure 1 A flowchart illustrating the control law design method for a thermal management type combined power unit provided in Embodiment 1 of this application. Please refer to... Figure 1 The method provided in this embodiment may include:

[0025] S101. Establish a mathematical model for the core components of the thermal management type combined power unit, construct an initial steady-state simulation model by combining the flow rate, pressure, and energy conservation relationship, and use the particle swarm optimization algorithm combined with experimental data to correct the initial steady-state simulation model to obtain a faithful steady-state simulation model.

[0026] Optionally, the core components include a compressor, a combustion chamber, a power turbine, and a cooling turbine, and mathematical models are established for the inlet and outlet flow rates, temperatures, pressures, and functions of each component.

[0027] Specifically, the thermal management-type combined power plant is a highly integrated composite system adapted to the upgrade of aircraft electromechanical systems. It integrates the traditional power generation / conversion device with the thermal management system in a unified design and control. While providing the aircraft with the required mechanical and electrical energy, it actively and efficiently manages the waste heat generated during system operation and utilizes kinetic energy to optimize power system performance. The core components of the thermal management-type combined power plant are key to achieving power generation and thermal management. These include the compressor, combustion chamber, power turbine, and cooling turbine. These components work collaboratively to complete core functions such as power conversion and waste heat management, forming the basic units for constructing the device's simulation model. The mathematical model is a characteristic quantification model established for each core component of the thermal management-type combined power plant: the compressor, combustion chamber, power turbine, and cooling turbine. Specifically, it constructs functional relationships between parameters based on key physical parameters such as inlet and outlet flow rates, temperature, pressure, and work / energy of each component, quantifying the operating characteristics of each core component.

[0028] In practical implementation, the core components of the thermal management type combined power unit are first identified as the compressor, combustion chamber, power turbine, and cooling turbine. Then, for each core component, the key physical parameters in its operation process are sorted out, and the core parameters such as inlet flow rate, outlet flow rate, inlet temperature, outlet temperature, inlet pressure, outlet pressure, output work or energy transfer are determined. Finally, based on the working principle and physical characteristics of each component, the functional correspondence between the above core parameters is constructed to form an independent mathematical model for each core component.

[0029] Furthermore, an initial steady-state simulation model is constructed by combining the conservation relationships of flow rate, pressure, and energy. This includes: establishing a system working equation set based on the mathematical model of the core components, in accordance with the conservation of flow rate, pressure, and energy, while simultaneously satisfying the continuous conditions of pressure, flow rate, and temperature between the components; and building an initial steady-state simulation model based on the system working equation set, and performing simulation verification for self-starting, main start-up, and cooling modes respectively.

[0030] Optionally, flow conservation includes the outlet flow rate of the power turbine being equal to the inlet flow rate of the combustion chamber; pressure conservation includes the initial inlet pressure of the system being equal to the product of the outlet pressure of the cooling turbine and the pipeline pressure loss coefficient; and energy conservation includes the sum of the output work of the power turbine and the cooling turbine being equal to the sum of the output work of the compressor and the power generation of the system.

[0031] Specifically, firstly, the connection sequence and parameter transfer relationships of the core components—compressor, combustion chamber, power turbine, and cooling turbine—in a thermally managed combined power unit are clarified, and the continuity conditions that pressure, flow rate, and temperature must meet between each component are determined (i.e., the pressure / flow rate / temperature at the outlet of the upstream component must be continuously consistent with the corresponding parameters at the inlet of the downstream component). Furthermore, quantitative rules for conservation relationships are established: at the flow rate conservation level, the outlet flow rate of the power turbine is equal to the inlet flow rate of the combustion chamber; at the pressure conservation level, the initial inlet pressure of the system is equal to the product of the outlet pressure of the cooling turbine and the pipeline pressure loss coefficient; at the energy conservation level, the sum of the output work of the power turbine and the cooling turbine is equal to the sum of the output work of the compressor and the system power generation. Based on the established mathematical models of each core component, the above quantitative rules for flow rate, pressure, and energy conservation are substituted, and the continuity conditions of pressure, flow rate, and temperature between each component are incorporated. Through mathematical derivation, a system working equation set describing the overall steady-state operating characteristics of the thermally managed combined power unit is formed. The completed system equations are converted into executable simulation code in Matlab. An initial steady-state simulation model of the thermally managed combined power unit is built in the Matlab simulation environment, and the model's basic operating parameters and boundary conditions are configured. For the three operating modes of the thermally managed combined power unit—self-starting, main-starting, and cooling—initial input parameters (such as fuel flow rate, ambient temperature, and initial speed) are set for each mode. The initial steady-state simulation model is run, and the simulation results of the core parameters of the unit (such as inlet / outlet pressure / temperature of each component, output power, etc.) for each mode are output, completing the simulation verification of the three operating modes.

[0032] For example, in one embodiment, the thermal management type combined power unit involves components including: a compressor, a combustion chamber, a power turbine, and a cooling turbine.

[0033] Compressor: ;

[0034] in, These are environmental pressure and environmental temperature, respectively. The compressor pressure ratio. This represents the shaft speed. These are the compressor's outlet flow rate, outlet temperature, outlet pressure, and power consumption, respectively.

[0035] Combustion chamber: ;

[0036] in, These are the inlet pressure, inlet temperature, inlet flow rate, and fuel quantity of the combustion chamber. These are the combustion chamber outlet flow rate, outlet temperature, and outlet pressure, respectively.

[0037] Power Turbo: ;

[0038] in, These are the inlet pressure, inlet temperature, and expansion ratio of the power turbine, respectively. This refers to the outlet flow rate, outlet temperature, outlet pressure, and output power of the power turbine.

[0039] Cooling the turbine: ;

[0040] in, These are the inlet pressure, inlet temperature, and expansion ratio of the cooling turbine, respectively. This is to monitor the turbine outlet flow rate, outlet temperature, outlet pressure, and output power.

[0041] Based on the mathematical model of the above components, and according to the three balance relationships of flow rate, pressure, and energy, as well as the continuous conditions of pressure, flow rate, and temperature between the components, the following set of working equations is obtained:

[0042] Flow conservation: ;

[0043] Pressure conservation: *P_loss;

[0044] Energy conservation: ;

[0045] in, This refers to the outlet flow rate of the power turbine. This refers to the inlet flow rate of the combustion chamber; This represents the initial inlet pressure of the system. P_loss is the outlet pressure of the cooling turbine; P_loss is the pressure loss coefficient along the pipeline. The power is generated by the turbine output. To cool the turbine output power; This is for the compressor to output power; This refers to the system's power generation.

[0046] In specific implementation, the initial steady-state simulation model is corrected using a particle swarm optimization algorithm combined with experimental data. This includes: selecting compressor pressure ratio, compressor flow rate, power turbine flow rate, power turbine expansion ratio, and cooling turbine flow rate and expansion ratio as correction parameters, with the goal of minimizing the error between experimental and simulation data, and initializing the correction parameters; after initialization, calculating the fitness function value for each particle, updating the individual historical best solution, updating the global historical best position, and synchronously updating the particle velocity vector and position vector, and iterating continuously until a preset termination condition is reached. The correction coefficients corresponding to the globally optimal correction parameter combination obtained through iteration are then updated to the initial steady-state simulation model to obtain a faithful steady-state simulation model.

[0047] Optionally, the initialization configuration includes setting the population size, learning factor coefficient, dynamic range of inertia weight, constraint range of correction parameters, and constraint range of parameter correction speed, and adopting a chaotic initialization strategy to generate an initial population of equal dimension based on the dimension of correction parameters.

[0048] Specifically, firstly, from the core component characteristic parameters of the thermal management combined power unit, six core parameters—compressor pressure ratio, compressor flow rate, power turbine flow rate, power turbine expansion ratio, cooling turbine flow rate, and cooling turbine expansion ratio—were selected as correction parameters for the initial steady-state simulation model. The physical meaning of each parameter and its corresponding position in the model were clarified. Simultaneously, experimental data from the physical prototype of the thermal management combined power unit was collected under three operating modes: self-start, main start, and cooling. This data covered key parameters such as inlet and outlet pressure, temperature, flow rate, and output power of the core components in each mode. This experimental data served as the benchmark data for model correction, and minimizing the error between the experimental data and the simulation data was determined as the sole optimization objective of the particle swarm optimization algorithm.

[0049] Further, the first step involves setting the core operating parameters of the particle swarm optimization algorithm: based on the number of correction parameters and the required optimization accuracy, the particle population size is set, the specific values ​​of the learning factor coefficients are defined, the dynamic adjustment range of the inertia weight, the value constraints of each correction parameter, and the upper and lower limits of the parameter correction speed are defined. The second step involves generating the initial population using a chaotic initialization strategy: based on the dimensions constituted by the six correction parameters, a chaotic sequence is constructed. An initial particle population that perfectly matches the dimensions of the correction parameters is generated through chaotic mapping calculations. Each particle represents an initial combination of the six correction parameters, completing all the initialization configuration work for the particle swarm optimization algorithm.

[0050] After initialization, the iterative process of particle swarm optimization is initiated. For each particle in the swarm, the following complete operations are performed in sequence: First, extract the six modified parameter combinations corresponding to the current particle and substitute them one by one into the corresponding core component characteristic modules in the initial steady-state simulation model, replacing the original initial parameter values ​​in the model; run the configured initial steady-state simulation model in the Matlab environment to simulate three operating modes: self-start, main-start, and cooling, and output the core parameter data obtained from the model simulation; match the simulation data with the corresponding physical prototype test data parameter by parameter and condition by condition, and obtain the fitness function value corresponding to the particle through the preset error calculation method. This value directly reflects the degree of fit between the model and the actual prototype under the current parameter combination. Retrieve the individual's historical best fitness value and corresponding parameter position from the previous iteration for this particle, and compare it with the currently calculated fitness function value. If the current fitness value is smaller, it indicates that the parameter combination corresponding to this particle is better. In this case, immediately update the particle's current position (i.e., correct the parameter combination) and current fitness value to the individual's historical best solution, overwriting the original record. If the current value is not superior, retain the original individual's historical best solution. Retrieve the global historical best fitness value and corresponding particle position of the entire particle population up to the previous iteration, and compare it with the current particle's fitness function value. If the current fitness value is smaller, update the particle's current position to the global historical best position of the population, and simultaneously update the global historical best fitness value. If the current value is not superior, retain the original global historical best position and fitness value. Based on the preset inertia weights (selected in real time according to the dynamic range) and learning factor coefficients, combined with the individual historical best position of the particle and the global historical best position of the population, the velocity update formula of the particle swarm optimization algorithm is substituted to calculate the new velocity vector of the particle; then, based on the new velocity vector and the particle's current position vector, the position update formula is substituted to calculate the new position vector of the particle, thus completing the position and velocity update of the particle in the iteration and obtaining a new set of correction parameters.

[0051] After completing a single iteration for all particles in the population, a termination condition check is immediately performed: It checks whether the number of iterations has reached the preset maximum number of iterations, and simultaneously determines whether the current global historical best fitness value has stabilized (i.e., the change in fitness value over multiple iterations is less than a preset threshold), or whether it has reached a preset error threshold. If either of these conditions is met, the iteration terminates; if neither is met, based on the updated particle position and velocity, the operation of "calculating fitness function value—updating individual historical best solution—updating global historical best position—synchronously updating velocity and position vectors" is repeated for all particles in the population to enter the next iteration, until the termination condition is met. When the iteration meets the termination condition, six correction parameters corresponding to the global historical best position of the population are extracted, and this optimal parameter combination is converted into correction coefficients for the characteristics of core components in the initial steady-state simulation model. Subsequently, in the Matlab environment, these correction coefficients are updated one by one into the mathematical model modules of the corresponding components such as the compressor, power turbine, and cooling turbine in the initial steady-state simulation model, replacing the original uncorrected parameters. After updating the parameters, the model was run again to verify the fit between the simulation data and the experimental data of the corrected model. Finally, a high-fidelity steady-state simulation model that has been corrected by the particle swarm optimization algorithm and is highly matched with the actual operating characteristics of the physical prototype was obtained.

[0052] For example, in one embodiment, the initial steady-state simulation model is corrected based on the particle swarm optimization algorithm. The correction parameters are six parameters: compressor pressure ratio, compressor flow rate, power turbine flow rate, power turbine expansion ratio, cooling turbine flow rate, and cooling turbine expansion ratio. The correction objective is to minimize the error between the experimental data and the simulation data. Initialization settings are performed for the above six optimization parameters, with the following specific configurations: the population size is set to 50 (N=50), and the learning factor is set to 1.3 for all parameters. The inertia weight range is set to [0.4, 0.8], the correction parameter range is limited to [0.9, 1.2], and the correction parameter velocity threshold is set to [-0.01, 0.01]. A chaotic initialization strategy is adopted to generate an initial population of equal dimensions based on the optimization parameter dimensions. After initialization, the iterative optimization stage begins. In the nth iteration, the following operations are performed on each particle i in sequence: 1) calculate the fitness value; 2) update the individual's historical best position; 3) update the global best position by comparing the local best with the global historical best; 4) update the particle velocity and position; 5) output the current global best solution and its corresponding fitness value. The process continues iterating until the termination condition is met.

[0053] S102. Based on the aforementioned high-fidelity steady-state simulation model, a transient simulation model is built, incorporating mode switching logic, fuel supply strategy, control target selection rules, and system safety boundaries, to design the basic control law for the transient simulation model.

[0054] Specifically, this application considers that a high-fidelity steady-state simulation model can only simulate the characteristics of the stable state of a thermally managed combined propulsion system under various operating modes, and cannot reflect the transient operating laws of the system during mode switching and dynamic changes in operating conditions. However, in actual flight, the system needs to frequently switch modes and adjust operating conditions, and its operating state is constantly changing. Relying solely on a high-fidelity steady-state simulation model is insufficient to design control laws that adapt to actual dynamic operating requirements. Therefore, a transient simulation model needs to be built on the high-fidelity foundation of the high-fidelity steady-state simulation model to simulate the dynamic operation of the system, providing a simulation environment that fits the actual operating scenario for control law design. The basic control law refers to the initial control rules and algorithm system adapted to the dynamic operating requirements of the thermally managed combined propulsion system.

[0055] Optionally, the mode switching logic includes the system entering a self-starting mode from standby, switching to a main start mode or a cooling mode after the self-starting mode is completed, and switching back to the self-starting mode after the main start mode or cooling mode stabilizes. There is no direct switching between the main start mode and the cooling mode. When an illegal mode conversion request is detected, the system triggers a state lock and records a fault log. When switching from the cooling mode to the self-starting mode, the system first shuts off the bleed air and does not supply fuel until the speed drops to a preset threshold. After the speed first falls below the preset threshold, the system ignites and supplies fuel according to the self-starting mode control law.

[0056] Specifically, mode switching logic is a deterministic control rule designed for thermally managed combined power units. It specifies the legal transition paths, transition operation requirements, and anomaly handling rules between standby, self-starting, main start, and cooling operating states. It serves as the core execution basis for mode switching during transient operation, clarifying whether, when, and how to switch between different modes, as well as the handling of transition anomalies. Specifically, it includes three core aspects: legal switching paths, specific switching operation requirements, and illegal switching handling. The first aspect clarifies the legal mode switching path: the system can only enter self-starting mode from standby mode; after self-starting mode is completed, it can switch to either main start mode or cooling mode; after the main start mode or cooling mode reaches a stable state, it can switch back to self-starting mode; direct switching between main start mode and cooling mode is not allowed, thus defining a unique legal path for mode transitions. The second aspect specifies the dedicated operational requirements for switching from cooling mode to self-starting mode. This switching scenario must follow specific execution steps: the system first shuts off the bleed air, and then prohibits fuel supply until the unit speed drops to a preset threshold (e.g., 10%). Once the speed first falls below this preset threshold, ignition is performed, and fuel supply is strictly in accordance with the control law of self-starting mode to ensure the smooth completion of the mode switch in this scenario. The third aspect establishes abnormal handling rules for illegal mode switching. When the system detects a mode conversion request that does not conform to the above legal path, it immediately triggers a state locking mechanism to maintain the current operating mode of the unit. At the same time, the illegal request is logged to prevent abnormal unit operation due to unauthorized switching and to ensure the safety of system operation.

[0057] Optionally, the fuel supply strategy includes a combination of acceleration closed-loop and speed closed-loop control, with differentiated PID control architectures configured for different operating modes. The PID parameters for fuel control are functions of flight altitude and / or ambient temperature, and are dynamically adjusted using lookup tables or interpolation algorithms. A fuel supply rate limit is set during the fuel supply process. In self-starting mode, a third-order composite control architecture is adopted, using open-loop fuel supply control in the first speed stage, PID acceleration closed-loop control in the second speed stage, and PID speed closed-loop control in the third speed stage. In main start-up mode, speed PID closed-loop control is adopted.

[0058] Specifically, the fuel supply strategy is a precise fuel quantity control rule and execution scheme designed for thermal management combined power units, adapting to the dynamic needs of different operating modes. It is a core component of the unit's basic control law, designed around four dimensions: closed-loop control combination method, differentiated PID control architecture, dynamic parameter adjustment, and fuel supply safety constraints. It clarifies the control method, execution process, and adjustment rules of fuel supply under different operating modes. Among them, the core closed-loop control combination method adopts a control method that combines acceleration closed-loop and speed closed-loop to solve the temperature closed-loop lag problem caused by temperature sensor inertia, meet the boundary requirements such as system exhaust temperature, and achieve precise closed-loop control of fuel quantity. The differentiated PID control architecture for different modes is configured according to the operating characteristics of self-starting and main-starting modes, adapting to the core operating requirements of each mode. The self-starting mode adopts a three-order composite control architecture, implementing differentiated control in three stages according to the unit speed: the first speed stage uses open-loop fuel supply control, the second speed stage uses PID acceleration closed-loop control, and the third speed stage uses PID speed closed-loop control; the main-starting mode directly uses speed PID closed-loop control. The dynamic adjustment rule for PID parameters involves setting the PID parameters of fuel control as a function of flight altitude and / or ambient temperature. Based on the actual operating conditions of the device, including flight altitude and ambient temperature, the PID parameters are dynamically adjusted in real time using a lookup table or interpolation algorithm to ensure the adaptability and accuracy of fuel quantity control under different operating conditions. The safety constraint on the fuel supply rate is set by limiting the fuel supply rate throughout the entire fuel supply process, defining the maximum and minimum values ​​of the fuel supply rate to prevent combustion instability caused by sudden changes in fuel supply, thus ensuring the safety and stability of the device's fuel supply and combustion process.

[0059] For example, in one embodiment, when the system is in self-starting mode, a three-order composite control architecture is adopted: Open-loop fuel supply control: During the initial ignition phase (0-7% RPM), a basic fuel quantity is provided according to a preset RPM-fuel rate mapping table (lookup table method) to ensure successful ignition; PID acceleration closed-loop control: When the system RPM is in the 7%-80% range, angular acceleration feedback (α=ΔN1 / Δt) is introduced, and the fuel supply is adjusted through an adaptive PID controller. The target acceleration α_target is determined by a bivariate interpolation algorithm of flight altitude H and current RPM; PID speed closed-loop control: When the RPM is higher than 80%, speed feedback PID control is adopted. The PID parameters are dynamically adjusted using a three-dimensional lookup table method of flight altitude H and ambient temperature T. Fuel rate limiting: A maximum fuel supply rate and a minimum fuel supply rate are set to prevent combustion instability caused by sudden changes in fuel supply. When the system is in main start-up mode, it switches to pure speed PID closed-loop control: dynamic PID parameter tuning: the controller parameters are adjusted in real time according to flight altitude and ambient temperature; fuel supply rate limiting: the maximum fuel supply rate and minimum fuel supply rate are set to prevent combustion instability caused by sudden changes in fuel supply.

[0060] Optionally, the control target selection rules include fuel quantity as the control target in the self-starting mode and the main start-up mode; and fuel quantity, compressor inlet pressure and air supply temperature as the control targets in the cooling mode, with the compressor inlet pressure and air supply temperature controlled by corresponding regulating valves using a bang-bang control method.

[0061] Specifically, the control target selection rules are based on the functional positioning of different operating modes of the thermal management combined power unit, and the fixed rules for the physical quantities that need to be regulated and the corresponding control methods in each mode are predetermined. The control target selected for the automatic start and main start modes is the system fuel quantity to ensure ignition success rate and combustion stability; the cooling mode requires coordinated control of the fuel quantity, compressor inlet pressure, and supply air temperature. In the cooling mode, the compressor inlet pressure and supply air temperature are controlled by two regulating valves, and based on actual conditions, both valves in the Simulink model use bang-bang control.

[0062] Specifically, the system safety boundary refers to the pre-set allowable operating range, control index limits, and safety constraints for key parameters during the entire operation and mode switching process of a thermally managed combined power unit, designed to prevent dangerous or abnormal operating conditions such as surge, overheating, overpressure, and control instability. These are rigid constraints that the basic control law must strictly adhere to. The system safety boundary mainly includes: the upper limit of mode switching time; the allowable range of system parameter overshoot; the maximum temperature that each component cannot exceed (overheating boundary); the upper and lower limits of key state parameters such as compressor outlet pressure and cooling turbine outlet temperature; the minimum limit of compressor stable operating margin; and the safety constraint ranges for control quantities such as speed, fuel supply rate, and actuator actuation amplitude. In the design of the basic control law, all mode switching, fuel supply, and parameter adjustment actions must not exceed the system safety boundary to ensure the safety, stability, and reliability of the unit during full-envelope, full-mode operation.

[0063] In practical implementation, within the Matlab / Simulink simulation environment, a faithful steady-state simulation model is used as the core foundation. The mathematical models of core components and the system's working equations are retained, and modules reflecting the dynamic characteristics of the device are supplemented, including a dynamic speed response module, a dynamic fuel flow adjustment module, and a mode switching timing control module, thus establishing the basic framework of the transient simulation model. The preset mode switching logic is transformed into a state machine control module in Simulink, clearly defining the state identifiers and transition trigger conditions for each mode: standby, self-start, main start, and cooling. Valid switching path rules, specific operation steps for switching from cooling to self-start, and illegal switching handling rules are written into this module, which is then integrated into the mode control layer of the transient simulation model. A fuel supply control module is built within the simulation model, incorporating control logic combining acceleration and speed closed-loop control. A third-order composite PID control architecture is configured for the self-start mode, and a pure speed PID closed-loop control architecture is configured for the main start mode. The functional relationship between PID parameters and flight altitude / ambient temperature is written into a parameter mapping table, and a lookup table / interpolation algorithm module is added to achieve dynamic parameter adjustment. Simultaneously, fuel supply rate limiting constraints are set within the module. In the model's control target allocation module, control targets are preset according to the operating mode: in self-starting and main start-up modes, only fuel quantity is set as the control target; in cooling mode, fuel quantity, compressor inlet pressure, and supply air temperature are set as control targets simultaneously. For compressor inlet pressure and supply air temperature, corresponding regulating valve control sub-modules are built, and bang-bang control logic is written to solidify the control target selection rules. A safety constraint verification module is added to the simulation model, with preset safety boundary parameters such as mode switching time threshold, system overshoot threshold, over-temperature boundary, compressor outlet pressure range, and cooling turbine outlet temperature range. This module is connected to each core control link of the model, and trigger logic is set—when parameters approach or exceed safety boundaries during simulation, protective actions such as limiting, load reduction, or mode locking are immediately triggered. Furthermore, the modules related to mode switching, fuel supply, control objectives, and safety boundaries mentioned above are deeply integrated with the basic framework of the transient simulation model to form a complete basic control law logic. Typical transient operating conditions such as self-start-main-start switching and self-start-cooling switching are set, and the transient simulation model is run to verify the execution logic of each control rule, the parameter adjustment process, and whether the safety boundary constraints are effective, thus completing the basic control law design of the transient simulation model.

[0064] S103. Using the particle swarm optimization algorithm, the fuel control PID parameters of the transient simulation model are used as the optimization object, and the mode switching time, system overshoot, and power turbine exhaust temperature are used as multi-objective optimization indicators to iteratively optimize the control parameters of the transient simulation model. The optimized parameters are then updated to the transient simulation model to obtain the transient simulation model of the thermal management type combined power unit.

[0065] Specifically, the optimization object refers to the fuel control PID parameters of the transient simulation model, specifically the proportional gain and integral gain in the fuel control PID adjustment process, along with the corresponding derivative gain. These parameters are the core objects of the particle swarm optimization algorithm for iterative optimization. Multi-objective optimization indices refer to three core performance indicators that the particle swarm optimization algorithm must simultaneously consider during iterative optimization to measure the optimization effect of control parameters. These include: mode switching time (the total time consumed by the thermal management combined power unit to switch between different operating modes), system overshoot (the maximum deviation of core operating parameters from the steady-state target value during transient processes such as mode switching and operating condition adjustment), and power turbine exhaust temperature (the exhaust temperature at the power turbine outlet).

[0066] In specific implementation, mode switching time, system overshoot, and turbine exhaust temperature are used as multi-objective optimization indicators. This includes: determining the maximum values ​​of mode switching time, system overshoot, and turbine exhaust temperature; comparing the actual mode switching time, actual system overshoot, and actual turbine exhaust temperature with their corresponding maximum values ​​to obtain the normalization factor for each indicator; using a weighted sum method to combine the three normalization factors to construct an error function, with the lowest error function value being the final objective of multi-objective optimization, and the weight coefficient of the normalization factor corresponding to turbine exhaust temperature being higher than the weight coefficients of the normalization factors corresponding to mode switching time and system overshoot.

[0067] Specifically, firstly, considering the design requirements of the thermal management combined power unit, system safety boundaries, and actual operating conditions, the maximum values ​​of mode switching time, system overshoot, and turbine exhaust temperature are determined as benchmark values ​​for normalization of each indicator. During the transient simulation model operation, the actual mode switching time, actual system overshoot, and actual turbine exhaust temperature are collected for each iteration. The actual values ​​are then compared with the maximum values ​​of their respective indicators to obtain the normalization factors for mode switching time, system overshoot, and turbine exhaust temperature. A weighted sum method is used to assign weight coefficients to the three normalization factors, with the weight coefficient for turbine exhaust temperature being higher than that for mode switching time and system overshoot. Each normalization factor is multiplied by its weight coefficient and summed to construct an error function based on this summation result. The lowest value of the error function is set as the final optimization target of the particle swarm optimization algorithm, thereby measuring the optimization effect of the fuel control PID parameters and guiding the direction of parameter iteration optimization.

[0068] For example, in the mode switching process, the maximum switching time is first determined. Maximum overshoot of the system Maximum exhaust temperature We normalize these three parameters and define three factors:

[0069] ;

[0070] ;

[0071] ;

[0072] in, This is the normalization factor for mode switching time; This refers to the actual mode switching time. This represents the maximum switching time. This is the system overshoot normalization factor; This refers to the system overshoot. This represents the maximum overshoot of the system. The normalization factor for exhaust temperature of the power turbine; This refers to the exhaust temperature. This represents the maximum exhaust temperature.

[0073] Based on the weighted sum method, the above parameters are combined. Considering that the exhaust temperature limit is relatively strict, the present invention adopts the following weighting scheme:

[0074] ;

[0075] in, It is the error function; , , These are the weights of the three factors; This is the normalization factor for mode switching time; This is the system overshoot normalization factor; This is the normalization factor for the exhaust temperature of the power turbine.

[0076] Based on the above operations, the optimization objective of the mode switching process is the error function. Take the lowest value.

[0077] For example, in one embodiment, transient simulation model optimization is performed based on particle swarm optimization. The correction parameters are the two PID parameters involved in the fuel control law: proportional gain Kp and integral gain Ki. The correction objective is to minimize the error function E while satisfying other system boundary constraints. Initialization settings are performed for the two optimization parameters as follows: population size is set to 100 (N=100), and the learning factor is 1.3 for all parameters. The inertia weight range is set to [0.4, 0.8], the correction parameter range is limited to [0.5, 2.0], and the correction parameter velocity threshold is set to [-0.01, 0.01]. A chaotic initialization strategy is used to generate an initial population of equal dimensions based on the optimization parameter dimensions. After initialization, the iterative optimization phase begins. The iterative optimization process is similar to the optimization of the initial steady-state simulation model in the above embodiment and will not be repeated here.

[0078] The method provided in this embodiment, firstly, provides precise quantitative basis at the component level for the simulation model through the mathematical model of the core components. The conservation relationships of flow rate, pressure, and energy ensure that the initial steady-state simulation model conforms to the physical mechanism of device operation, avoiding the problem of the model deviating from actual physical laws. On this basis, the efficient optimization capability of the particle swarm optimization algorithm is utilized, combined with the experimental data of the physical prototype of the thermal management combined power device, to iteratively correct the core parameters of the initial steady-state simulation model. The resulting faithful steady-state simulation model can highly approximate the actual operating characteristics of the device and accurately reproduce the nonlinear response of each core component. By studying parameter coupling relationships and steady-state characteristics under a single mode, and prioritizing steady-state and transient states while ensuring model accuracy before designing control laws, a near-realistic control design environment is provided for subsequent basic control law design and parameter optimization. This effectively solves the problem of insufficient accuracy in traditional simulation models leading to a disconnect between control law design and actual experiments. It enables the subsequently designed control laws to adapt to the complex nonlinear characteristics of the device operation in advance and to handle dynamic parameter changes during mode switching. This lays a solid foundation for the engineering practicality of control laws and avoids problems such as control failure and response lag caused by model distortion in practical applications.

[0079] Secondly, the mode switching logic, fuel supply strategy, control target selection rules, and system safety boundary together constitute the "safety barrier" of the control law design. It does not simply rely on data-driven or algorithmic optimization, but rather ensures the physical rationality and operational safety of control decisions through deterministic rule constraints: the mode switching logic clarifies the legal transition paths, specific operating steps, and anomaly handling rules between each operating mode, avoiding system instability caused by unauthorized mode switching; the fuel supply strategy combines the characteristics of different operating modes to configure a differentiated control architecture, ensuring the accuracy and stability of fuel supply and preventing combustion anomalies caused by sudden fuel supply changes; the control target selection rules clarify the core of regulation based on the functional positioning of each mode, ensuring that the control direction aligns with the device's operational needs; the system safety boundary presets the allowable range and control limits of key parameters, eliminating dangerous conditions such as over-temperature, over-pressure, and surge. By deeply integrating physical mechanism rules with the simulation model, it effectively avoids unreasonable and unsafe control commands that may be generated by purely data-driven methods due to data bias or insufficient samples, fundamentally ensuring the safety and reliability of the control law's operation and ensuring that every decision of the control law conforms to the physical laws and safety requirements of the device's operation.

[0080] Thirdly, by introducing particle swarm optimization (PSO) twice for core optimization, a dual optimization system of "model correction + parameter optimization" is formed, further improving the adaptability and robustness of the control law: Firstly, PSO is used to correct the initial steady-state simulation model, leveraging its advantages of global optimization and fast convergence speed to quickly find the model parameter combination with the smallest error compared to experimental data, efficiently improving model fidelity; secondly, PSO is used for iterative optimization of the PID parameters of the fuel control system in the transient simulation model, using mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators, targeting different operating conditions. The system intelligently optimizes PID parameters under different modes and operating conditions, enabling offline learning and adaptive adjustment of control parameters through algorithm iteration. This effectively addresses various uncertainties in the operation of thermally managed combined power units, solving the problems of traditional fixed-parameter control laws being unable to adapt to full-envelope operation requirements and lacking robustness. The optimized control law not only meets the requirements of multi-objective optimization indicators, shortens mode switching time, reduces system overshoot, and controls the exhaust temperature of the power turbine within a safe range, but also maintains stable control performance under various complex and uncertain operating conditions, exhibiting stronger robustness and adaptability.

[0081] Corresponding to the aforementioned embodiment of the control law design method for a thermal management type combined power unit, this application also provides an embodiment of a control law design apparatus for a thermal management type combined power unit.

[0082] Figure 2This is a schematic diagram of the control law design device for a thermal management type combined power unit provided in Embodiment 2 of this application. Please refer to... Figure 2 The apparatus provided in this embodiment includes a construction module 210, a design module 220, and an optimization module 230;

[0083] The construction module 210 is used to establish a mathematical model for the core components of the thermal management type combined power device, construct an initial steady-state simulation model by combining the flow rate, pressure and energy conservation relationship, and use the particle swarm optimization algorithm combined with experimental data to correct the initial steady-state simulation model to obtain a faithful steady-state simulation model.

[0084] The design module 220 is used to build a transient simulation model based on the fidelity steady-state simulation model, and incorporates mode switching logic, fuel supply strategy, control target selection rules and system safety boundary to design the basic control law of the transient simulation model.

[0085] The optimization module 230 is used to employ a particle swarm optimization algorithm, taking the fuel control PID parameters of the transient simulation model as the optimization object, and using mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators, to iteratively optimize the control parameters of the transient simulation model, and update the optimized parameters to the transient simulation model to obtain a thermal management type combined power unit transient simulation model.

[0086] The apparatus of this embodiment can be used to perform... Figure 1 The steps of the method embodiment shown are similar in principle and process, and will not be repeated here.

[0087] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0088] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0089] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for designing control laws for a thermally managed combined power unit, characterized in that, The method includes: A mathematical model is established for the core components of a thermal management type combined power unit. An initial steady-state simulation model is constructed by combining the flow rate, pressure, and energy conservation relationships. The initial steady-state simulation model is then corrected using a particle swarm optimization algorithm combined with experimental data to obtain a faithful steady-state simulation model. A transient simulation model is built based on the aforementioned high-fidelity steady-state simulation model, incorporating mode switching logic, fuel supply strategy, control target selection rules, and system safety boundaries to design the basic control law for the transient simulation model. The fuel supply strategy includes a combination of acceleration closed-loop and speed closed-loop control, with differentiated PID control architectures configured for different operating modes. The PID parameters for fuel control are functions of flight altitude and / or ambient temperature, dynamically adjusted using lookup tables or interpolation algorithms. A fuel supply rate limit is set during fuel supply. In self-starting mode, a third-order composite control architecture is adopted, using open-loop fuel supply control in the first speed stage, PID acceleration closed-loop control in the second speed stage, and PID speed closed-loop control in the third speed stage. In main start-up mode, speed PID closed-loop control is used. The particle swarm optimization algorithm is used to optimize the fuel control PID parameters of the transient simulation model. The mode switching time, system overshoot, and power turbine exhaust temperature are used as multi-objective optimization indicators to iteratively optimize the control parameters of the transient simulation model. The optimized parameters are then updated to the transient simulation model to obtain the transient simulation model of the thermal management type combined power unit. The initial steady-state simulation model was corrected using a particle swarm optimization algorithm combined with experimental data, including: Compressor pressure ratio, compressor flow rate, power turbine flow rate, power turbine expansion ratio, cooling turbine flow rate, and cooling turbine expansion ratio are selected as correction parameters. The optimization objective is to minimize the error between experimental data and simulation data. The correction parameters are then initialized. After initialization, the fitness function value is calculated for each particle sequentially, the individual historical best solution is updated, the global historical best position is updated, and the particle velocity vector and position vector are updated synchronously. The iteration continues until the preset termination condition is reached. The correction coefficients corresponding to the global optimal correction parameter combination obtained by the iteration are updated to the initial steady-state simulation model to obtain a faithful steady-state simulation model. The optimization metrics include mode switching time, system overshoot, and turbine exhaust temperature, among others. Determine the maximum value of mode switching time, the maximum value of system overshoot, and the maximum value of power turbine exhaust temperature respectively; The normalization factor for each indicator is obtained by comparing the actual mode switching time, actual system overshoot, and actual power turbine exhaust temperature with their corresponding maximum values. The three normalization factors are combined using a weighted sum method to construct an error function. The lowest value of the error function is taken as the final objective of multi-objective optimization. The weight coefficient of the normalization factor corresponding to the exhaust temperature of the power turbine is higher than the weight coefficient of the normalization factor corresponding to the mode switching time and the system overshoot.

2. The control law design method for thermal management type combined power unit according to claim 1, characterized in that, An initial steady-state simulation model is constructed by combining the conservation relationships of flow rate, pressure, and energy, including: Based on the mathematical model of the core components, and in accordance with the conservation of flow, pressure, and energy, while simultaneously satisfying the continuous conditions of pressure, flow, and temperature between the components, a set of system working equations is established. An initial steady-state simulation model was built based on the system's working equations, and simulation verifications were performed for self-starting, main-starting, and cooling modes, respectively.

3. The control law design method for thermal management type combined power unit according to claim 2, characterized in that, The core components include a compressor, a combustion chamber, a power turbine, and a cooling turbine. Mathematical models are established for the inlet and outlet flow rates, temperatures, pressures, and work / energy related parameters of each component. Flow conservation includes ensuring that the outlet flow rate of the power turbine is equal to the inlet flow rate of the combustion chamber. Pressure conservation includes ensuring that the initial inlet pressure of the system is equal to the product of the outlet pressure of the cooling turbine and the pipeline pressure loss coefficient. Energy conservation includes ensuring that the sum of the output work of the power turbine and the cooling turbine is equal to the sum of the output work of the compressor and the power generation of the system.

4. The control law design method for thermal management type combined power unit according to claim 1, characterized in that, The initialization configuration includes setting the population size, learning factor coefficient, dynamic range of inertia weight, constraint range of correction parameters, and constraint range of parameter correction speed. A chaotic initialization strategy is used to generate an initial population of equal dimensions based on the dimension of the correction parameters.

5. The control law design method for thermal management type combined power unit according to claim 1, characterized in that, The mode switching logic includes the system entering the self-starting mode from standby, switching to the main start mode or cooling mode after the self-starting mode is completed, and switching back to the self-starting mode after the main start mode or cooling mode stabilizes. There is no direct switching between the main start mode and the cooling mode. When an illegal mode conversion request is detected, the system triggers a state lock and records a fault log. When switching from the cooling mode to the self-starting mode, the system first shuts off the bleed air and does not supply fuel until the speed drops to a preset threshold. After the speed first drops below the preset threshold, the system ignites and supplies fuel according to the self-starting mode control law.

6. The control law design method for thermal management type combined power unit according to claim 1, characterized in that, The control target selection rules include the control target being fuel quantity in the self-starting mode and the main start-up mode; and the control targets being fuel quantity, compressor inlet pressure and air supply temperature in the cooling mode, with the compressor inlet pressure and air supply temperature controlled by corresponding regulating valves using a bang-bang control method.

7. A control law design apparatus for a thermal management type combined power unit, wherein the control law design apparatus for the thermal management type combined power unit is applied to the control law design method for the thermal management type combined power unit according to any one of claims 1-6, characterized in that, The device includes a construction module, a design module, and an optimization module; The construction module is used to establish a mathematical model for the core components of the thermal management type combined power unit, construct an initial steady-state simulation model by combining the flow rate, pressure and energy conservation relationship, and use the particle swarm optimization algorithm combined with experimental data to correct the initial steady-state simulation model to obtain a faithful steady-state simulation model. The design module is used to build a transient simulation model based on the fidelity steady-state simulation model, and incorporates mode switching logic, fuel supply strategy, control target selection rules and system safety boundaries to design the basic control law of the transient simulation model. The optimization module is used to employ a particle swarm optimization algorithm, taking the fuel control PID parameters of the transient simulation model as the optimization object, and using mode switching time, system overshoot, and power turbine exhaust temperature as multi-objective optimization indicators to iteratively optimize the control parameters of the transient simulation model, and update the optimized parameters to the transient simulation model to obtain a thermal management type combined power unit transient simulation model.

Images