An intelligent parameter optimization method for thermal management combined power unit

Abstract

The present invention belongs to the field of aviation electromechanical technology, and discloses an intelligent parameter optimization method for a thermal management type combined power device, determines the input and output parameters of each component of the thermal management type combined power device, mathematically models the above input and output parameters, establishes a system common working equation based on each input and output parameter, establishes a set of equations and a simulation model based on the system balance relationship, and solves to obtain the output parameter results of each component; determines the functional objectives, performance objectives, input and output parameters, optimization parameters and system constraints of the device according to the system requirements of the current working mode; optimizes the simulation model using a particle swarm intelligent algorithm, and finally obtains the system output parameters with the best energy efficiency. The present invention establishes an optimization evaluation system with energy consumption as the evaluation index and system requirements as the constraint conditions for this device, and then obtains the optimal input parameters that meet the optimization evaluation system through parameter optimization, thereby reducing energy consumption and improving system efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of aviation electromechanical technology, and relates to a parameter optimization method for a power device, and specifically to an intelligent parameter optimization method for a thermal management type combined power device. Background Art

[0002] The thermal management combined power unit integrates the aircraft environmental control system and the secondary power system. The same compressor can provide compressed air for the environmental control system and the auxiliary power system, and the generator is designed coaxially with the compressor and turbine. The mass and volume of the electromechanical system are reduced, which is in line with the development trend of integrated and multi-electrified electromechanical systems. At the same time, it also reduces the overall mass of the aircraft and improves the aircraft's combat performance. It is the focus of future aircraft electromechanical technology research. The functionally integrated thermal management combined power unit improves the system integration, but due to the nonlinearity and strong coupling of the system, the resulting control and system optimization configuration problems are prominent.

[0003] With the continuous development of aviation electromechanical technology, how to improve the energy utilization of electromechanical systems while meeting the functional requirements of the system has become an urgent problem to be solved. By improving the efficiency of the system, the energy cost of the aircraft can be saved, the impact on the engine can be reduced, the combat range can be increased, and the overall performance of the aircraft can be improved. Therefore, it is necessary to carry out energy optimization research. The commonly used system energy optimization method is to select several groups of typical design parameters through experience or calculation, compare the solution results of different parameter allocations to select the optimal solution, and improve the system efficiency by optimizing the parameter allocation, but there are the following problems:

[0004] 1. For complex systems with multiple inputs and outputs, empirical selection of parameters is no longer suitable. It is difficult to give design point parameters, and since the design parameters are nonlinearly coupled, the empirical selection effect is poor.

[0005] 2. The parameter allocation method selected by expert experience is highly subjective, has low precision, selects a small number of design parameters, and the optimization method is accidental.

[0006] 3. Thermal management combined power devices have the characteristics of nonlinearity and strong coupling. Using exhaustive methods to calculate a large number of design parameters will greatly increase the calculation cost and reduce efficiency.

[0007] The patent "Energy Management Strategy for an Aircraft Adaptive Power and Thermal Management System" (publication number CN106970523B) introduces an energy management strategy for a power and thermal management system, which uses machine training samples to obtain the energy distribution of a neural network controller control system to achieve energy optimization management. However, this method requires a large number of data samples for training. Neural networks have the characteristics of complex calculations, and it is currently impossible to obtain a large number of data samples required for machine training. Therefore, it is necessary to seek a simpler and more efficient energy optimization method. Summary of the invention

[0008] Purpose of the invention: In order to solve the above problems, the present invention provides an intelligent parameter optimization method for a thermal management combined power device, which can screen out the optimal input parameters to ensure that the system as a whole obtains the optimal energy utilization rate, improve the optimization effect and reduce the optimization operation time.

[0009] The technical solution of the present invention is as follows:

[0010] An intelligent parameter optimization method for a thermal management type combined power device, which determines the input and output parameters of each component of the thermal management type combined power device, mathematically models the input and output parameters, establishes a system common working equation based on each input and output parameter, establishes an equation group and a simulation model based on the relationship between the system flow balance, power balance, and pressure balance, and solves to obtain the output parameter results of each component;

[0011] The functional objectives, performance objectives, input and output parameters, optimization parameters and system constraints of the device are determined according to the system requirements of the current working mode; the particle swarm intelligent algorithm is used to optimize the simulation model, and finally the system output parameters with optimal energy efficiency are obtained.

[0012] Furthermore, the input and output parameters of each component of the thermal management type combined power device are as follows:

[0013] Input and output parameters of compressors, power turbines and refrigeration turbines, including gas flow, pressure, temperature and fuel flow;

[0014] Combustion chamber input and output parameters, including gas flow, pressure, temperature and fuel flow;

[0015] Generator input and output parameters, including shaft power and power generation;

[0016] The input and output parameters of the heat exchanger include cold / hot end flow, pressure and temperature.

[0017] Furthermore, the method of mathematically modeling the input and output parameters of the compressor is:

[0018] [Q 1_out , T 1_out , P1_out , W 1_out ]=f1(n,P 1_in , T 1_in ,π1)

[0019] Where P 1＿in is the environmental pressure, T 1＿in is the ambient temperature, π1 is the compressor pressure ratio, and n is the shaft speed of the thermal management combined power unit. 1_out is the compressor outlet temperature, P 1_out is the outlet pressure of the compressor, W 1_out is the power consumed by the compressor.

[0020] Furthermore, the method of mathematically modeling the input and output parameters of the combustion chamber is:

[0021] [Q 2_out , T 2_out , P 2_out ]=f2(P 2_in , T 2_in , Q 2_in , Q fuel )

[0022] Where P 2_in is the inlet pressure of the combustion chamber, T 2_in is the inlet temperature of the combustion chamber, Q 2_in is the inlet flow rate of the combustion chamber, Q fuel is the amount of fuel in the combustion chamber, Q 2_out is the outlet flow rate of the combustion chamber, T 2_out is the outlet temperature of the combustion chamber, P 2_out is the outlet pressure of the combustion chamber.

[0023] Furthermore, the mathematical modeling method for the input and output parameters of the power turbine is:

[0024] [Q 3_out , T 3_out , P 3_out , W 3_out ]=f3(n,P 3_in , T 3_in ,π3)

[0025] Where P 3_in is the inlet pressure of the power turbine, T 3_in is the inlet temperature of the power turbine, π3 is the expansion ratio of the power turbine, Q 3_out Power turbine outlet flow, T 3_out is the outlet temperature of the power turbine, P 3_out is the outlet pressure of the power turbine, W 3_out is the output work of the power turbine.

[0026] Furthermore, the method of mathematically modeling the input and output parameters of the generator is:

[0027] [W 5_out ]=f5(n,W 5_in )

[0028] Where W 5_in is the input work of the generator, W 5_out is the output work of the generator.

[0029] Furthermore, the specific method of establishing a set of equations and a simulation model according to the relationship between system flow balance, power balance, and pressure balance is:

[0030] The steady-state joint working equation of the mode is established, and the flow and pressure loss in the pipeline are temporarily ignored. According to the interface continuity conditions and coaxial matching relationship of each component, the flow balance, pressure balance and energy balance equations are established; the interface continuity conditions of each component include: the corresponding parameters of the compressor inlet and the atmospheric environment are equal, the corresponding parameters of the compressor outlet and the combustion chamber inlet are equal, and the corresponding parameters of the combustion chamber outlet and the power turbine inlet are equal; according to the coaxial matching relationship, the work of the compressor, the power turbine and the generator is balanced, and finally the following joint working equation group is established:

[0031] Flow balance equation:

[0032] Q 1_out =Q 2_in

[0033] Q2_ out =Q 3_out

[0034] Pressure balance equation:

[0035] P 2_in =P 1_out

[0036] P 2_out =P 3_in

[0037] P 3_out =P 1_in *σ

[0038] Where σ is the expansion coefficient of the nozzle, which is determined according to the nozzle characteristics.

[0039] Energy balance equation:

[0040] T 1_out =T 2_in

[0041] T 2_out =T 3_in

[0042] W 5_in =W 3_OUT -W 1_OUT .

[0043] Furthermore, a method for solving and obtaining the output parameter results of each component is: using the Newton-Raphson numerical solution method to solve the common working equation group to obtain the parameter results of each component.

[0044] Furthermore, in the method of optimizing the simulation model using the particle swarm intelligence algorithm, chaos initialization, adaptive mutation and inertia weight adaptive adjustment improvement methods are also used.

[0045] Furthermore, the above method specifically comprises the following steps:

[0046] Step 1: Parameter initialization, including population size, learning factor, inertia weight, maximum number of iterations, particle search space range and movement speed, etc.;

[0047] Step 2: Adopt the chaos initialization strategy to randomly generate an initial population of the same dimension according to the dimension of the input parameter to be optimized;

[0048] Step 3: Combined with the simulation model, the fitness value of each particle in the population is calculated according to the fitness evaluation method in the optimization evaluation system;

[0049] Step 4: Screen out particles that do not meet the constraints according to the optimization evaluation system, and compare and update individual extreme values ​​and group extreme values;

[0050] Step 5: Substitute the above results into the particle swarm motion equation, adjust the inertia weight according to the adaptive strategy, then update the speed and position of the population, and adopt the mutation strategy to generate the next generation of population.

[0051] Step 6: Determine whether the maximum number of iterations has been reached. If not, repeat steps 3 to 5. If the maximum number of iterations has been reached, the algorithm ends and the optimization process ends.

[0052] Step 7: Output the global optimal particle, i.e. the optimization result.

[0053] Beneficial effects of the present invention:

[0054] 1) Simulate and model the thermal management combined power unit and establish a simulation model of the thermal management combined power unit with high accuracy and fast convergence

[0055] 2) For this device, an optimization evaluation system is established with energy consumption as the evaluation index and system requirements as the constraint condition. The particle swarm algorithm process is used to optimize the parameters of the simulation model of the thermal management combined power device to obtain the optimal input parameters that meet the optimization evaluation system. Compared with the traditional method of selecting parameters based on expert experience, it reduces energy consumption and improves system efficiency.

[0056] 3) The parameters of the power and thermal management devices are optimized at a relatively fast speed, which reduces the computational and time costs compared to the exhaustive calculation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a diagram of the thermal management type combined power device architecture used in the method of the present invention;

[0058] Figure 2 Schematic diagram of the optimization evaluation system of thermal management combined power unit;

[0059] Figure 3 Particle swarm optimization algorithm flow chart;

[0060] Figure 4 Flowchart of optimization simulation of thermal management combined power unit;

[0061] Among them, 1- compressor, 2- combustion chamber, 3- power turbine, 4- refrigeration turbine, 5- generator, 6- heat exchanger. DETAILED DESCRIPTION

[0062] This section is an embodiment of the present invention, which is used to explain and illustrate the technical solution of the present invention.

[0063] An intelligent parameter optimization method for a thermal management type combined power device, which determines the input and output parameters of each component of the thermal management type combined power device, mathematically models the input and output parameters, establishes a system common working equation based on each input and output parameter, establishes an equation group and a simulation model based on the relationship between the system flow balance, power balance, and pressure balance, and solves to obtain the output parameter results of each component;

[0064] The functional objectives, performance objectives, input and output parameters, optimization parameters and system constraints of the device are determined according to the system requirements of the current working mode; the particle swarm intelligent algorithm is used to optimize the simulation model, and finally the system output parameters with optimal energy efficiency are obtained.

[0065] The input and output parameters of each component of the thermal management combined power unit are as follows:

[0066] Input and output parameters of compressors, power turbines and refrigeration turbines, including gas flow, pressure, temperature and fuel flow;

[0067] Combustion chamber input and output parameters, including gas flow, pressure, temperature and fuel flow;

[0068] Generator input and output parameters, including shaft power and power generation;

[0069] The input and output parameters of the heat exchanger include cold / hot end flow, pressure and temperature.

[0070] The method of mathematical modeling of the input and output parameters of the compressor is:

[0071] [Q 1_out , T 1_out , P 1_out , W 1_out ]=f1(n,P 1_in , T 1_in ,π1)

[0072] Where P 1_in is the environmental pressure, T 1_in is the ambient temperature, π1 is the compressor pressure ratio, n is the shaft speed of the thermal management combined power unit; Q 1_out is the outlet flow rate of the compressor, T 1_out is the compressor outlet temperature, P 1_out is the outlet pressure of the compressor, W 1_out is the power consumed by the compressor.

[0073] The method for mathematical modeling of the input and output parameters of the combustion chamber is:

[0074] [Q 2_out , T 2_out , P 2_out ]=f2(P 2_in , T 2_in , Q 2_in , Q fuel )

[0075] Where P 2_in is the inlet pressure of the combustion chamber, T 2_in is the inlet temperature of the combustion chamber, Q 2_in is the inlet flow rate of the combustion chamber, Q fuel is the amount of fuel in the combustion chamber, Q 2_out is the outlet flow rate of the combustion chamber, T 2_out is the outlet temperature of the combustion chamber, P 2_out is the outlet pressure of the combustion chamber.

[0076] The mathematical modeling method for the input and output parameters of the power turbine is:

[0077] [Q 3_out , T 3_out , P 3_out , W3_out ]=f3(n,P 3_in , T 3_in ,π3)

[0078] Where P 3_in is the inlet pressure of the power turbine, T 3_in is the inlet temperature of the power turbine, π3 is the expansion ratio of the power turbine, Q 3_out Power turbine outlet flow, T 3_out is the outlet temperature of the power turbine, P 3_out is the outlet pressure of the power turbine, W 3_out is the output work of the power turbine.

[0079] The method of mathematical modeling of the input and output parameters of the generator is:

[0080] [W 5_out ]=f5(n,W 5_in )

[0081] Where W 5_in is the input work of the generator, W 5_out is the output work of the generator.

[0082] The specific method of establishing a set of equations and a simulation model based on the relationship between system flow balance, power balance, and pressure balance is as follows:

[0083] The steady-state joint working equation of the mode is established, and the flow and pressure loss in the pipeline are temporarily ignored. According to the interface continuity conditions and coaxial matching relationship of each component, the flow balance, pressure balance and energy balance equations are established; the interface continuity conditions of each component include: the corresponding parameters of the compressor inlet and the atmospheric environment are equal, the corresponding parameters of the compressor outlet and the combustion chamber inlet are equal, and the corresponding parameters of the combustion chamber outlet and the power turbine inlet are equal; according to the coaxial matching relationship, the work of the compressor, the power turbine and the generator is balanced, and finally the following joint working equation group is established:

[0084] Flow balance equation:

[0085] Q 1_out =Q 2_ i n

[0086] Q 2_out =Q 3_out

[0087] Pressure balance equation:

[0088] P 2_in =P 1_out

[0089] P 2_out =P 3_in

[0090] P 3_out =P 1_in *σ

[0091] Where σ is the expansion coefficient of the nozzle, which is determined according to the nozzle characteristics.

[0092] Energy balance equation:

[0093] T 1_out =T 2_in

[0094] T 2_out =T 3_in

[0095] W 5_in =W 3_OUT -W 1_OUT .

[0096] The method for solving the output parameter results of each component is: using the Newton-Raphson numerical solution method to solve the common working equation group to obtain the parameter results of each component.

[0097] In the method of optimizing the simulation model using the particle swarm intelligence algorithm, chaos initialization, adaptive mutation and inertia weight adaptive adjustment improvement methods are also used.

[0098] The above method specifically comprises the following steps:

[0099] Step 1: Parameter initialization, including population size, learning factor, inertia weight, maximum number of iterations, particle search space range and movement speed, etc.;

[0100] Step 2: Adopt the chaos initialization strategy to randomly generate an initial population of the same dimension according to the dimension of the input parameter to be optimized;

[0101] Step 3: Combined with the simulation model, the fitness value of each particle in the population is calculated according to the fitness evaluation method in the optimization evaluation system;

[0102] Step 4: Screen out particles that do not meet the constraints according to the optimization evaluation system, and compare and update individual extreme values ​​and group extreme values;

[0103] Step 5: Substitute the above results into the particle swarm motion equation, adjust the inertia weight according to the adaptive strategy, then update the speed and position of the population, and adopt the mutation strategy to generate the next generation of population.

[0104] Step 6: Determine whether the maximum number of iterations has been reached. If not, repeat steps 3 to 5. If the maximum number of iterations has been reached, the algorithm ends and the optimization process ends.

[0105] Step 7: Output the global optimal particle, i.e. the optimization result.

[0106] Another embodiment of the present invention is further described below with reference to the accompanying drawings.

[0107] Figure 1 It is a thermal management type combined power device used in the present invention, and the device includes a compressor 1, a combustion chamber 2, a power turbine 3, a refrigeration turbine 4, a generator 5, and a heat exchanger 6.

[0108] Taking the optimization of the ground start main engine mode of the thermal management combined power unit as an example, the particle swarm intelligent algorithm is used to optimize the parameters.

[0109] The working parts in the main engine start mode are the compressor 1, the combustion chamber 2, the power turbine 3 and the generator 5. The compressor inhales air from the outside and compresses it. The heated and pressurized air is ignited and burned in the combustion chamber and then enters the power turbine to expand and do work, driving the generator to generate electricity, and starting the main engine by electric starting.

[0110] Mathematical modeling is carried out according to the characteristics of the selected compressor, and the mathematical function form of the compressor model is [Q 1_out , T 1_out , P 1_out , W 1_out ]=f1(n,P 1_in , T 1_in ,π1) where P 1_in , T 1_in are respectively the ambient pressure and ambient temperature, π 1_out is the compressor pressure ratio, and n is the shaft speed of the thermal management combined power unit. 1_out , T 1_out , P 1_out , W 1_out They are the outlet flow, outlet temperature, outlet pressure and power consumption of the compressor respectively.

[0111] Mathematical modeling is performed based on the characteristics of the selected combustion chamber, and the mathematical function form of the combustion chamber model is [Q 2_out , T 2_out , P 2_out ]=f2(P 2_in , T 2_inn , Q 2_inn , Q fuel )

[0112] Where P 2_in , T 2_in , Q 2_in , Q fuel are the inlet pressure, inlet temperature, inlet flow rate and fuel quantity of the combustion chamber, Q 2_out , T 2_out , P2_out are the outlet flow rate, outlet temperature and outlet pressure of the combustion chamber respectively.

[0113] Mathematical modeling is performed based on the characteristics of the selected power turbine, and the mathematical function form of the power turbine model is established as follows:

[0114] [Q 3_out , T 3_out , P 3_out , W 3_out ]=f3(n,P 3_inn , T 3_inn ,π3)

[0115] Where P 3_in , T 3_in , π3 are the inlet pressure, inlet temperature and expansion ratio of the power turbine, Q 3_out , T 3_out , P 3_out , W 2_out are the power turbine outlet flow, outlet temperature, outlet pressure and output work.

[0116] Mathematical modeling is performed based on the characteristics of the selected generator, and the mathematical function form of the generator model is established as follows:

[0117] [W 5_out ]=f5(n,W 5_in )

[0118] Where W 3_in is the input work of the generator, W 3_out is the output work of the generator.

[0119] After establishing the models of each component, the steady-state joint working equation of the mode is established, temporarily ignoring the flow and pressure loss in the pipeline, and establishing the flow balance, pressure balance, and energy balance equations based on the continuity conditions of the interfaces of each component and the coaxial matching relationship.

[0120] The interface continuity conditions of each component include: the corresponding parameters of the compressor inlet and the atmospheric environment are equal, the corresponding parameters of the compressor outlet and the combustion chamber inlet are equal, and the corresponding parameters of the combustion chamber outlet and the power turbine inlet are equal. According to the coaxial matching relationship, the compressor, power turbine and generator perform work in balance. The following common working equations are established for solution.

[0121] Flow balance equation:

[0122] Q 1_out =Q 2_in

[0123] Q 2_out =Q 3_out

[0124] Pressure balance equation:

[0125] P 2_in =P 1_out

[0126] P 2_out =P 3_in

[0127] P 3_out =P 1_in *σ

[0128] Where σ is the expansion coefficient of the nozzle, which is determined according to the nozzle characteristics.

[0129] Energy balance equation:

[0130] T 1_out =T 2_in

[0131] T 2_out =T 3_in

[0132] W 5_in =W 3_OUT -W 1_OUT

[0133] After establishing the equation group, the Newton-Raphson numerical solution method is used to solve the equation group, and the overall input-output function model of the thermal management type combined power device is obtained as follows:

[0134] [W 5_out , T 3_out ]=f(n,P 1_in , T 1_in , Q fuel )

[0135] Through the function model of the thermal management type combined power device, the environmental parameters, the rotation speed, and the fuel quantity are inputted to output the power generation and the exhaust temperature of the thermal management type combined power device.

[0136] According to the system requirements of the thermal management combined power unit in the main engine starting mode, an optimization evaluation method is established. The optimization goal is determined to be the lowest fuel consumption, and the constraints are that the power generation meets the demand and the power turbine outlet does not overheat, and the input parameter combination with the lowest fuel consumption is found. The optimization objective function expression is as follows:

[0137]

[0138] Where × represents the fuel consumption and speed of the variables to be optimized in the active start mode, y represents other input variables such as flow, temperature, pressure, etc. c represents the power generation demand of the system, and the remaining inequalities represent the search interval of the system input parameters.

[0139] Combining the particle swarm optimization process, the optimization steps of the thermal management combined power unit in the active start mode can be deduced as follows: Figure 4 shown.

[0140] Step 1: Determine the parameter range to be optimized for the input variables fuel quantity and speed, and use random functions to initialize the variables to obtain the initialized particle swarm.

[0141] Step 2: Use the constraints to establish the optimization objective function, and substitute the obtained particles into the optimization objective function of the active start mode to obtain the fitness value of the particles.

[0142] Step 3: Obtain the global optimal particle and the local optimal particle at this time according to the obtained fitness value. The system constraints are that the power generation meets the demand, the system operates stably, and the power turbine outlet does not overheat. If the constraints are not met, the particles that do not meet the constraints are screened out.

[0143] Step 4: Substitute the particle's motion state equation to obtain the state vector of the next generation of particles in the particle swarm.

[0144] Step 5: Determine the termination condition. If the termination condition is not met, repeat steps 2 to 4 to update the iterative global optimal particle. If the termination condition is met, stop the iteration and output the global optimal particle at this time.

[0145] Taking the sea level starting condition as the optimization object, the optimization simulation is carried out, the ambient temperature is 15℃, the ambient pressure is 101kPa, and the shaft speed is optimized for different power generation requirements. The optimized results are compared with the results of the simulation model without optimization to verify whether the optimization effect is produced. The verification results are shown in Table 1. The fuel volume on the left is the optimized fuel volume, and the fuel volume on the right is the comparison fuel volume at the rated speed. The results show that for different power generation requirements, the fuel volume is significantly reduced after optimization by the particle swarm intelligent algorithm, and the optimization effect is obvious.

[0146] Table 1 Main engine starting mode optimization table

[0147]

Claims

1. A method for intelligent parameter optimization of a thermal management combined power unit, characterized in that: Determine the input and output parameters of each component of the thermal management combined power unit, mathematically model the above input and output parameters, establish the system common working equation based on each input and output parameter, establish a group of equations based on the relationship between system flow balance, power balance, and pressure balance, establish a simulation model, and solve to obtain the output parameter results of each component; Determine the functional objectives, performance objectives, input and output parameters, optimization parameters and system constraints of the device according to the system requirements of the current working mode; use the particle swarm intelligence algorithm to optimize the simulation model and finally obtain the system output parameters with the best energy efficiency; The specific method of establishing the equation group and the simulation model according to the relationship between the system flow balance, power balance and pressure balance is as follows: establishing the steady-state common working equation of the current working mode, temporarily ignoring the flow and pressure loss in the pipeline, and establishing the flow balance, pressure balance and energy balance equations according to the interface continuity conditions and coaxial matching relationship of each component; The continuity conditions of the interfaces of each component include: the corresponding parameters of the compressor inlet and the atmospheric environment are equal, the corresponding parameters of the compressor outlet and the combustion chamber inlet are equal, and the corresponding parameters of the combustion chamber outlet and the power turbine inlet are equal; according to the coaxial matching relationship, the work of the compressor, power turbine and generator is balanced, and finally a group of common working equations is established; The input and output parameters of each component of the thermal management combined power unit are as follows: Input and output parameters of compressors, power turbines and refrigeration turbines, including gas flow, pressure and temperature; Combustion chamber input and output parameters, including gas flow, pressure, temperature and fuel flow; Generator input and output parameters, including shaft power and power generation; The input and output parameters of the heat exchanger include the cold / hot end flow, pressure and temperature; among them, through the function model of the thermal management combined power unit, the input environmental parameters, speed and fuel volume can output the power generation and exhaust temperature of the thermal management combined power unit. According to the system requirements of the thermal management combined power unit in the main engine starting mode, an optimization evaluation method is established to determine the optimization target as the lowest fuel consumption, and the constraints are that the power generation meets the demand and the power turbine outlet does not overheat, and find the input parameter combination with the lowest fuel consumption; the optimization objective function expression is: Where × represents the fuel consumption and speed of the variables to be optimized in the main engine starting mode, y represents the other input variables; c represents the power generation demand of the system, and the remaining inequalities represent the search interval of the system input parameters; In the method of optimizing the simulation model by using the particle swarm intelligence algorithm, the improvement method of chaos initialization, adaptive mutation and inertia weight adaptive adjustment is also used, which specifically includes the following steps: Step 1: Parameter initialization, including population size, learning factor, inertia weight, maximum number of iterations, particle search space range and movement speed; Step 2: Adopt the chaos initialization strategy to randomly generate an initial population of the same dimension according to the dimension of the input parameter to be optimized; Step 3: Combined with the simulation model, the fitness value of each particle in the population is calculated according to the fitness evaluation method in the optimization evaluation system; Step 4: Screen out particles that do not meet the constraints according to the optimization evaluation system, and compare and update individual extreme values ​​and group extreme values; Step 5: Substitute the above results into the particle swarm motion equation, adjust the inertia weight according to the adaptive strategy, then update the speed and position of the population, and use the mutation strategy to generate the next generation of population; Step 6: Determine whether the maximum number of iterations has been reached. If not, repeat steps 3 to 5. If the maximum number of iterations has been reached, the algorithm ends and the optimization process ends. Step 7: Output the global optimal particle, i.e. the optimization result.

2. The intelligent parameter optimization method of a thermal management combined power unit according to claim 1, characterized in that: The method of mathematical modeling of the input and output parameters of the compressor is: [Q 1_out ,T 1_out ,P 1_out ,W 1_out ]=f1(n,P 1_in ,T 1_in ,π1) Where P 1_in is the environmental pressure, T 1_in is the ambient temperature, π1 is the compressor pressure ratio, n is the shaft speed of the thermal management combined power unit; Q 1_out is the outlet flow rate of the compressor, T 1_out is the compressor outlet temperature, P 1_out is the outlet pressure of the compressor, W 1_out is the power consumed by the compressor.

3. The intelligent parameter optimization method of a thermal management combined power unit according to claim 2, characterized in that: The method for mathematical modeling of the input and output parameters of the combustion chamber is: [Q 2_out ,T 2_out ,P 2_out ]=f2(P 2_in ,T 2_in ,Q 2_in ,Q fuel ) Where P 2_in is the inlet pressure of the combustion chamber, T 2_in is the inlet temperature of the combustion chamber, Q 2_in is the inlet flow rate of the combustion chamber, Q fuel is the amount of fuel in the combustion chamber, Q 2_out is the outlet flow rate of the combustion chamber, T 2_out is the outlet temperature of the combustion chamber, P 2_out is the outlet pressure of the combustion chamber.

4. The intelligent parameter optimization method of a thermal management combined power unit according to claim 3, characterized in that: The method of mathematically modeling the input and output parameters of the power turbine is: [Q 3_out ,T 3_out ,P 3_out ,W 3_out ]=f3(n,P 3_in ,T 3_in ,π3) Where P 3_in is the inlet pressure of the power turbine, T 3_in is the inlet temperature of the power turbine, π3 is the expansion ratio of the power turbine, Q 3_out Power turbine outlet flow, T 3_out is the outlet temperature of the power turbine, P 3_out is the outlet pressure of the power turbine, W 3_out is the output work of the power turbine.

5. The intelligent parameter optimization method of a thermal management type combined power unit according to claim 4, characterized in that: The method of mathematical modeling of the input and output parameters of the generator is: [W 5_out ]=f5(n,W 5_in ) Where W 5_in is the input work of the generator, W 5_out is the output work of the generator.

6. The intelligent parameter optimization method of a thermal management combined power unit according to claim 5, characterized in that: The compressor, power turbine and generator perform work in balance, and the final working equations are as follows: Flow balance equation: Q 1_out =Q 2_in Q 2_out =Q 3_out Pressure balance equation: P 2_in =P 1_out P 2_out =P 3_in P 3_out =P 1_in *σ Where σ is the expansion coefficient of the nozzle, which is determined according to the nozzle characteristics; Energy balance equation: T 1_out =T 2_in T 2_out =T 3_in IN 5_in =In 3_OUT -IN 1_OUT 。

Images