Wide-speed-range aircraft multi-source energy dynamic configuration method

Abstract

The invention discloses a wide-speed-range aircraft multi-source energy dynamic configuration method, and belongs to the technical field of calculation, reckoning or counting. Aiming at the characteristics of strong nonlinearity, high dynamism and multi-peak distribution in a TBCC mode conversion stage, the method comprises the following steps: firstly, constructing an environment vector e containing a flight condition and an engine state; secondly, establishing a strong coupling nonlinear mechanism model of the TBCC engine and the multi-source electric energy generation system; further constructing a multi-target optimization model taking the thrust loss delta FN and the unit thrust oil consumption increment delta SFC as targets; a model-driven environment perception EA-NSGA-II algorithm is adopted, the environment is monitored jointly through environment vectors and fitness changes, and the Pareto frontier is dynamically tracked through local restart, population memory and environment prediction mechanisms; and finally mapping the solution set into a multi-source power distribution interval. According to the method, the optimal energy configuration scheme meeting the safety constraint can be output in a wide speed range, especially in a dynamic mode conversion working condition.

Description

Technical Field

[0001] This invention relates to integrated energy optimization and engine control technology for hypersonic vehicles, and discloses a method for dynamic configuration of multi-source energy for wide-speed-range vehicles. In particular, it relates to a dynamic configuration method for multi-source electrical energy generation and energy efficiency synergistic optimization for turbine-based combined cycle power systems, and belongs to the technical field of calculation, estimation or counting. Background Technology

[0002] The Turbine-Based Combined Cycle (TBCC) propulsion system operates in parallel with the turbofan and ramjet channels, achieving a wide speed range propulsion from subsonic to hypersonic speeds. In hypersonic missions, the onboard electrical load is large and changes rapidly with each mission phase, typically requiring the extraction of shaft work from the engine main shaft or the use of bleed air from the channels to drive an air turbine for power generation. The impact of multi-source power generation on the main propulsion exhibits strong nonlinearity and strong coupling characteristics: power extraction alters the compressor operating point, turbine inlet temperature, speed, and stability margin, thus affecting thrust and fuel consumption per unit thrust. Especially during mode transitions, the engine's operating state and constraint boundaries fluctuate dramatically with rapid changes in the physical flight environment, such as Mach number and altitude, leading to a highly dynamic and multi-peaked solution space for the optimization problem.

[0003] Existing multi-objective optimization methods often separate the optimization algorithm from the engine model, failing to establish a unified closed loop for fitness calculation and safety constraint assessment. This makes it difficult to simultaneously guarantee full envelope safety constraints and optimality under rapidly changing physical conditions. Furthermore, existing algorithms lack memory reuse and predictive parameter tuning mechanisms to address abrupt changes in the optimization solution space caused by alterations in the physical flight environment, making them prone to getting trapped in local optima or experiencing convergence lag during sudden environmental changes. Therefore, it is necessary to propose a multi-source energy dynamic configuration method with deep algorithm-model coupling, tailored to the mode transition characteristics of TBCC (Through-The-Air Compression Capacity). Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the aforementioned background technology by providing a multi-source energy dynamic allocation method for the highly nonlinear, dynamic, and multi-peak distribution characteristics of the TBCC mode transition stage. By integrating and coupling the TBCC engine mechanism model, safety margin model, and environmental perception evolutionary optimization algorithm, the invention aims to dynamically determine the optimal power allocation interval of multiple sources within the entire flight envelope while ensuring safety constraints. This solves the technical problems of existing multi-objective optimization methods, which struggle to simultaneously guarantee full envelope safety constraints and optimality in dynamic environments and are prone to getting trapped in local optima due to insufficient utilization of mode transition characteristics.

[0005] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0006] A method for dynamic configuration of multi-source energy for a wide-speed-range aircraft includes:

[0007] S1 collects flight conditions and TBCC engine status parameters, and constructs an environmental vector that includes Mach number, flight altitude, total inlet temperature / total pressure, TBCC engine modes, and stability margin characterization of compression components.

[0008] S2. Establish a unified coupled nonlinear mechanism model of the TBCC engine and its multi-source power generation system. The mechanism model obtains thrust, fuel consumption rate per unit thrust, and parameters for safety assessment based on the input environmental vector and decision variables. The decision variables include the output power of various power generation modes of the multi-source power generation system. The parameters for safety assessment include the critical section temperature, the rotational speed of rotating parts, and the stability margin of the compression components.

[0009] S3. Under the conditions of satisfying power demand and safety constraints, a multi-objective energy management optimization model is constructed with thrust loss and unit thrust fuel consumption rate increment as optimization objectives.

[0010] S4 uses a mechanism-driven environment-aware non-dominated sorting genetic algorithm II to solve the multi-objective energy management optimization model and obtain the Pareto optimal solution set.

[0011] S5, perform boundary extraction on the Pareto optimal solution set obtained in S4, obtain the multi-source power allocation interval corresponding to the discrete operating point of the full envelope, interpolate the multi-source power allocation interval according to the current environment vector, obtain the real-time interval of multi-source power allocation, and then select the optimal decision variable.

[0012] S6. The mechanism model is driven by the current environment vector and the optimal decision variables to perform a safety verification and obtain a multi-energy power allocation scheme that satisfies the full envelope safety constraints.

[0013] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, the multi-objective energy management optimization model constructed by S3 includes:

[0014] Optimization goal: , ,

[0015] Power balance constraints: ,

[0016] Temperature constraints on critical sections: ,

[0017] Rotating component speed constraint: ,

[0018] Stability margin constraints for compression components: ,

[0019] Modal feasible region constraint: turret fan modal stage constraint =0, stamping mode stage let =0 and =0;

[0020] in, For thrust loss, For the increase in fuel consumption per unit thrust, , The reference thrust and reference fuel consumption per unit thrust are given under the same environmental vector without power generation extraction. Decision variables and environment vector The downward thrust, Decision variables and environment vector Fuel consumption per unit thrust under the following conditions These represent the output power of the shaft power extraction, turbofan bleed air turbine, and ramjet bleed air turbine, respectively. For power requirements, Decision variables and environment vector Lower critical section temperature, Decision variables and environment vector Lower rotating component rotational speed , Decision variables and environment vector Lower compression component Stability margin For critical sections Maximum temperature For rotating parts Maximum speed, For compression components Minimum stability margin.

[0021] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, S4 specifically employs: Combinatorial gene coding strategies will incorporate decision variables and constraint processing auxiliary parameters Mapped to individual chromosome genes, In each generation of evolutionary iterations, the following algorithm flow is executed:

[0022] S41, based on the constraint processing auxiliary parameters, the first... The chromosome genes of each individual in the population are dynamically repaired with step size constraints and their fitness is assessed.

[0023] S42, after detecting the first When the environment of a population changes drastically, calculate the first generation. After assessing the overall environmental changes of the population, it entered S43, and after monitoring the first... Drastic environmental changes in the next generation or the first generation of population When the overall environmental change of a generation exceeds a threshold, it enters S45;

[0024] S43, according to the The overall environmental change range of the population is adaptively adjusted in the first generation. Evolutionary parameters of a generational population;

[0025] S44, Predicting the first The magnitude of the overall environmental change in the population of the first generation When the predicted value of the overall environmental change of the population exceeds the threshold, the S43 adaptive adjustment is performed. Evolutionary parameters of a generational population;

[0026] S45, for the first The generation population performs a restart operation and sends feedback to S41.

[0027] As a further optimization scheme for the dynamic configuration method of multi-source energy for wide-speed-range aircraft, in S41, the constraint processing auxiliary parameters are used to adjust the first... The chromosome genes of each individual in the population undergo dynamic step-size constraint repair, specifically as follows:

[0028] In the individual When power balance constraints are violated, the first individual of Genes dynamically adjust repair step size and update the first step. individual of Genes are , , The target deviation vector;

[0029] In the individual When the modal feasible region constraint is violated, according to The decay rate is controlled to gradually approach the modal feasible region.

[0030] As a further optimization scheme for the dynamic configuration method of multi-source energy for wide-speed-range aircraft, in S41, the constraint processing auxiliary parameters are used to adjust the first... After dynamic step-size constraint repair of the chromosome genes of each individual in the population, fitness is evaluated according to the following method: using... and the Generational population environment vector The mechanism model drives the key cross-section output by the mechanism model. Temperature T i ( e) Rotating components rotational speed or stability margin If the safety constraints are still violated, the degree of violation is calculated and a penalty function is applied.

[0031] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, in S42, after monitoring the first... When the environment of a population changes drastically, calculate the first generation. The overall environmental change range of the population is as follows:

[0032] No. Variation in the physical environment vector of the population Or the magnitude of change in the statistical properties of the algorithm's fitness When the threshold is exceeded, determine the first The environmental changes in the population generation are abnormal. , , For the first Generation population, first The environmental vector of the generation population , For the first Generation population, first Mean fitness of the generation population;

[0033] After monitoring the first Frequency of environmental change in generation population When the threshold is exceeded, determine the first The environment of the population changed drastically in each generation, and the calculation of the first generation was carried out. The range of comprehensive environmental changes in the population , , , This represents the generation in which the last environmental mutation was detected. These are the normalized weighting coefficients.

[0034] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, in S43, according to the first... The evolutionary parameters are adaptively adjusted based on the magnitude of overall environmental changes in the population, specifically:

[0035] According to the The range of comprehensive environmental changes in the population Definition of the first Adaptive adjustment parameters of generation population , , Adjust the threshold for the set environmental change range;

[0036] use Dynamic adjustment of the first Variation rate of the population With cross rate , No. When the environmental changes of a generation's population are extremely drastic, the population expands. Population size of the generation , , , , Based on the basic rate of variation, Based on the cross rate, Based on the basic population size, , , The adjustment factor is a positive number.

[0037] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, S44 predicts the first energy level using a time series model. The magnitude of comprehensive environmental changes in the population.

[0038] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, S45... The generation population will be restarted, specifically as follows:

[0039] The current Pareto frontier optimal individual and its corresponding first Generational population environment vector Store in memory bank , ;

[0040] Retrieve from memory the first Record historical environment vectors that are similar to the environment vectors of the current generation population, and search the memory database for the first generation. Environment vector record The corresponding set of historical best individuals With the Elite individuals preserved in the population and randomly generated supplementary individuals Integration, Reconstruction Generation population for If no match is found Record historical environment vectors that are similar to the current population environment vectors, perform a local population restart, and only retain the first-generation environment vector. Elite individuals of a generation The remaining individuals are randomly reset, reconstructing the first... Generation population for , For the first time after random reset Individuals to the first Individual, The total number of individuals;

[0041] For the reconstructed first Generation population Based on the adaptively adjusted evolutionary parameters determined in S43, selection, crossover, and mutation operations are performed to generate the next generation population.

[0042] As a further optimization of the multi-source energy dynamic configuration method for wide-speed-range aircraft, this involves retrieving data from the memory bank that matches the first... The method for recording historical environment vectors that are similar to the environment vectors of the next generation population is as follows: Calculate the first generation... Generational population environment vector With the first in the memory bank Environment vector record The weighted Euclidean distance.

[0043] Compared with the prior art, the present invention adopts the above-described technical solution and has at least the following beneficial effects:

[0044] 1. Through closed-loop coupling of "environmental vector - mechanism model - fitness / constraint", thrust, fuel consumption rate and safety margin are dynamically calculated within the same solution framework, which is applicable to TBCC full-coverage linewidth working conditions;

[0045] 2. To address the characteristics of strong nonlinearity, high dynamics, and multi-peak distribution in the mode transition stage, a model-driven environment-aware EA-NSGA-II algorithm is adopted. This algorithm monitors the environment by combining environment vectors and fitness changes, and utilizes local restart, population memory, and predictive parameter tuning to achieve rapid tracking of the global optimal solution set, thereby reducing the risk of getting trapped in local optima.

[0046] 3. From the Pareto front to the power distribution range The mapping output can be directly used for the allocation boundary of the control law, improving real-time performance and engineering deployability;

[0047] 4. While meeting the constraints of temperature, speed and stability margin, achieve a coordinated reduction in thrust loss and fuel consumption per unit thrust. Attached Figure Description

[0048] Figure 1 This is a flowchart of the multi-source energy dynamic configuration method for wide-speed-range aircraft proposed in this invention.

[0049] Figure 2This is a schematic diagram of a TBCC engine model provided in one embodiment of the present invention.

[0050] Figure 3 This is a schematic diagram of a turbine-based combined power multi-source energy management strategy provided in one embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only some, not all, of the embodiments of this invention.

[0052] This invention proposes a dynamic multi-source energy allocation method for wide-speed-range aircraft. The method utilizes an environment-aware multi-objective genetic algorithm to solve for the current optimal multi-source energy allocation. Based on the establishment of a nonlinear mechanism model of the TBCC engine and the multi-source power generation system, an environment-aware mechanism and a local population restart strategy are introduced. Through the coupling of environmental parameters and fitness functions, the optimal allocation interval for multi-source energy under different flight conditions is dynamically determined. Within the determined interval, the optimized solution is extracted and compared with the non-optimized power generation method to verify the effectiveness of the optimization interval obtained by the energy optimization method.

[0053] Compared with existing technologies, the innovation of this invention lies in the coupling of environmental perception and model: by using external environmental parameters such as Mach number, flight altitude, engine modes, total inlet temperature and total pressure, and stability margin to drive a strongly coupled nonlinear mechanism model of the TBCC and multi-source power generation system, thrust loss is dynamically calculated. Increment in fuel consumption per unit thrust and changes in stability margin And use it as a component of the fitness function.

[0054] like Figure 1 As shown, the multi-source energy dynamic configuration method under turbine-based combined power of the present invention includes S1 to S6.

[0055] S1, Environmental Vector Construction: Collect flight conditions and TBCC engine state parameters to construct environmental vectors. ;in, Mach number, For flight altitude, / Total temperature / total pressure of the intake manifold. This refers to the TBCC engine modes, which include: turbofan, mode transition, and ramjet. This is a characterization of the stability margin of the compression component. It can consist of the stability margins of the high-pressure compressor and fan. The environment vector is used to drive the calculation of the TBCC mechanism model and serves as part of the environment perception criterion.

[0056] S2, Establish a strongly coupled nonlinear mechanism model between TBCC and multi-source power generation system: such as Figure 2 As shown, a unified coupled nonlinear mechanism model of the TBCC engine and its multi-source power generation system is established. The subsystems within this mechanism model are coupled with each other, specifically including: the parallel aerodynamic and thermodynamic model of the TBCC turbofan channel and ramjet channel, the main shaft shaft power extraction power generation model, the turbofan channel bleed air turbine power generation model, and the ramjet channel bleed air turbine power generation model.

[0057] The input of the mechanism model is ,in, Configure decision variables for multi-source power. , These represent the output power of shaft power extraction, turbofan bleed air turbine, and ramjet bleed air turbine, respectively; the model output is thrust. Fuel consumption per unit thrust And parameters used for safety assessment, including: critical sections. temperature Rotating parts rotational speed Compression components stability margin Key section i refers to critical sections such as the combustion chamber outlet and turbine inlet; rotating component j refers to rotating components such as the high-pressure rotor and low-pressure rotor; compression component k refers to compression components such as the fan and high-pressure compressor. Key section i, rotating component j, and compression component... The type is determined based on the TBCC operating mode.

[0058] S3, in meeting power requirements Under safety constraints, construct a system based on thrust loss. and the increase in fuel consumption per unit thrust A multi-objective energy management optimization model for optimizing objectives:

[0059] thrust loss Increment in fuel consumption per unit thrust For multiple objectives: (1) (2)

[0060] Meet power requirements That is, power balance: (3)

[0061] Safety constraints refer to limiting the parameters output by the mechanistic model in S2 for safety assessment within a safe range, specifically including:

[0062] Temperature constraints on critical sections: (4)

[0063] Rotating component speed constraint: (5)

[0064] Stability margin constraints for compression components: (6)

[0065] Modal feasible region constraint: turret fan modal stage constraint =0; Stamping mode stage setting =0 and =0.

[0066] In equations (1) to (6), , No power generation extraction is performed under the same environmental vector. = = The reference thrust and reference fuel consumption per unit thrust at =0 For critical sections Maximum temperature For rotating parts Maximum speed, For compression components Minimum stability margin.

[0067] S4, using the mechanism-driven Environment Aware Non-dominated Sorting Genetic Algorithm II (EA-NSGA-II), solves the multi-objective energy management optimization model to obtain the Pareto optimal solution set. This step combines model-driven and data-driven approaches, dynamically adjusting the optimization strategy in each generation of evolution. Combinatorial gene coding strategies will incorporate decision variables and constraint processing auxiliary parameters Mapped to individual chromosome genes in evolutionary algorithms, where , The algorithm executes in each generation of evolutionary iterations according to the following process:

[0068] S41, Dynamic Constraint Repair and Fitness Assessment

[0069] For the current population, i.e., the first Generation population Before each entity in the process calls the mechanism model established by S2 to obtain thrust, fuel consumption per unit thrust, and parameters used for safety assessment, it first executes a process based on... Dynamic step size constraint repair, which includes:

[0070] Power balance constraint repair: When the first individual Violation of power balance constraints At that time, instead of directly forcing projection to the boundary, the first method is used. individual of Genes dynamically adjust the repair step size. Calculate the target deviation vector. Update # individual of Genes are : (7)

[0071] pass Co-evolution, the algorithm uses smaller values In the early stages of the search, some individuals with minor violations but superior genes can be retained to avoid premature convergence to the boundary, and gradually approach the modal feasible region as evolution progresses.

[0072] Modal feasible region constraint repair: When modal feasible region constraints are violated, such as in the stamping modal stage... At that time, according to The decay rate is controlled to gradually approach the modal feasible region.

[0073] Penalty function processing: After the above dynamic step size constraint repair, and Input the mechanism model established by S2. If the key section output by the mechanism model... Temperature T i ( e) Rotating components rotational speed or stability margin If the safety constraints are still violated, the degree of violation is calculated and a penalty function is applied, significantly reducing the penalty. The first generation of the population individual fitness .

[0074] S42, Environmental Change Monitoring

[0075] After completing the dynamic constraint repair and fitness assessment for all individuals, calculate the mean fitness of the current population. : (8)

[0076] To accurately capture the dynamic characteristics of the mode transition phase, and simultaneously calculate the... Variation in the physical environment vector of the generation population Variation in the statistical properties of the algorithm's fitness : (9) (10)

[0077] Set fitness change threshold Threshold for change in environmental vector ,when or At that time, it was believed that the first The environment of the generational population undergoes a sudden change.

[0078] If the If the environmental conditions of a population change, the magnitude and frequency of these changes need to be further assessed. Let the generation number in which the last environmental mutation was detected be... The frequency of environmental changes in the current population The calculation is as follows: (11)

[0079] Frequency of environmental change in the current population When the set threshold is reached, it indicates that the first... Dramatic changes in the environment of the population necessitate more significant adjustments to strategies.

[0080] Based on the above indicators, the weighted normalization method is used to calculate the first... The range of comprehensive environmental changes in the population This serves as a quantitative basis for judging sudden environmental changes. (12)

[0081] In equations (8) to (12), The total number of individuals, , For the first Generation population, first The environmental vector of the generation population , For the first Generation population, first Mean fitness of the generation population These are the normalized weighting coefficients.

[0082] S43, adaptive adjustment of evolutionary parameters

[0083] Based on the calculated comprehensive environmental change range Definition of the first Adaptive adjustment parameters of generation population : (13)

[0084] use Dynamically adjust the variation rate of the current generation With cross rate To maintain convergence when the environment is stable and enhance exploration when the environment changes drastically: (14) (15)

[0085] When environmental changes are extremely drastic, When the population is large, temporarily expand the population size. To increase diversity: (16)

[0086] In equation (13), Adjust the threshold to the set range of environmental changes. Based on the basic rate of variation, Based on the cross rate, Based on the basic population size, , , The adjustment factor is a positive number.

[0087] S44, Environmental Trend Forecasting

[0088] To overcome response lag, establish past Comprehensive environmental change range sequence of the era The Autoregressive Integrated Moving Average (ARIMA) model is used to predict the magnitude of the next generation's comprehensive environmental changes. : (17)

[0089] in, For model constants, Let the order be the autoregressive order. These are the autoregressive coefficients of the ARIMA model. For the first The random noise term representing the overall environmental change amplitude of a generational population.

[0090] If the predicted value If the set threshold is exceeded, the evolution parameter adaptive adjustment operation in S43 will be performed in advance in the current generation to cope with the upcoming environmental mutation.

[0091] S45, Memory Mechanism and Population Reboot

[0092] When a sudden change in the environment is detected, or If the threshold is exceeded, perform a population restart operation:

[0093] Memory update: The current Pareto frontier optimal individual and its corresponding current population environment vector Store in memory bank : (18)

[0094] Memory retrieval: Calculate the current population environment vector. With the first in the memory bank Environment vector record Weighted Euclidean distance : (19)

[0095] in, This is a diagonal matrix of environmental weights.

[0096] Population fusion: if it exists Historical records If the similarity threshold is used, then the first element in the memory is extracted. Environment vector record The corresponding set of historical best individuals Combined with the elite individuals preserved in the current generation Reconstruct the current population with randomly generated supplementary individuals. : (20)

[0097] If no similar historical environment vector record is found, a local population restart is performed, retaining only elite individuals and randomly resetting the rest: (twenty one)

[0098] S46, generating the next generation

[0099] Based on the reconstructed population Using the adaptive parameters determined in S43 Perform selection, crossover, and mutation operations to produce the next generation of population. Then return to S41) and repeat the process until the termination condition is met.

[0100] S5, Pareto leading edge to power distribution range Mapping output Boundary extraction is performed on the Pareto optimal solution set obtained from S4 to obtain the multi-source power allocation interval. The upper bound U corresponds to the maximum extractable power boundary that a multi-source power generation system can provide under all safety constraints using a single power generation method, representing the system's work limit range; the lower bound L corresponds to the unit thrust fuel consumption rate increment of a multi-source power generation system under all safety constraints using a single power generation method. The minimum, or optimal, energy efficiency boundary represents the lowest energy consumption cost required to maintain power balance under current operating conditions. This is used in offline scenarios to form discrete operating points for the entire circuit envelope. Table lookup; in online scenarios, the lookup results are interpolated based on the current environment (e) to obtain a real-time interval, and then selected according to task weight within the interval. .

[0101] S6, Closed-loop safety check and constraint repair: Output Then, the mechanism model established based on S2 is checked again for safety; if T is found i N j or SM k If a constraint violation exists, a dynamic step-size constraint repair and adjustment will be performed. If necessary, return to S4 to solve again to ensure the safety constraints of the entire envelope. Finally, obtain a multi-energy power allocation scheme that satisfies the safety constraints of the entire envelope.

[0102] Taking the mode transition phase as an example, the Mach number Ma is set to 3~3.5, the flight altitude H to 20~22 km, and the power requirement is set to... The power is 500 kW, and the decision variable is... , , By using steps S1 to S6 above, the allocation range of the three power generation methods in this stage can be obtained. And, under the premise of meeting the constraints of turbine inlet temperature, shaft speed, and stability margin, the output makes and A compromise-optimal power configuration. A schematic diagram of a turbine-based combined power multi-source energy management strategy provided in one embodiment of the present invention is shown below. Figure 3 As shown.

[0103] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the specific embodiments described above. The specific embodiments and descriptions in the specification are merely for further illustrating the principles and preparation effects of the present invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the claims and their equivalents.

Claims

1. A method for dynamic configuration of multi-source energy for a wide-speed-range aircraft, characterized in that, include: S1 collects flight conditions and TBCC engine status parameters, and constructs an environmental vector that includes Mach number, flight altitude, total inlet temperature / total pressure, TBCC engine modes, and stability margin characterization of compression components. S2. Establish a unified coupled nonlinear mechanism model of the TBCC engine and its multi-source power generation system. The mechanism model obtains thrust, fuel consumption rate per unit thrust, and parameters for safety assessment based on the input environmental vector and decision variables. The decision variables include the output power of various power generation modes of the multi-source power generation system. The parameters for safety assessment include the critical section temperature, the rotational speed of rotating components, and the stability margin of the compression components. S3. Under the conditions of satisfying power demand and safety constraints, a multi-objective energy management optimization model is constructed with thrust loss and unit thrust fuel consumption rate increment as optimization objectives. S4. The multi-objective energy management optimization model is solved by the mechanism model-driven environment-aware non-dominated sorting genetic algorithm II to obtain the Pareto optimal solution set. S5, perform boundary extraction on the Pareto optimal solution set obtained in S4, obtain the multi-source power allocation interval corresponding to the discrete operating point of the full envelope, interpolate the multi-source power allocation interval according to the current environment vector, obtain the real-time interval of multi-source power allocation, and then select the optimal decision variable. S6. The mechanism model is driven by the current environment vector and the optimal decision variables to perform a safety verification and obtain a multi-energy power allocation scheme that satisfies the full envelope safety constraints.

2. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 1, characterized in that, The multi-objective energy management optimization model constructed by S3 includes: Optimization goal: , , Power balance constraints: , Temperature constraints on critical sections: , Rotating component speed constraint: , Stability margin constraints for compression components: , Modal feasible region constraint: turret fan modal stage constraint =0, stamping mode stage let =0 and =0; in, For thrust loss, For the increase in fuel consumption per unit thrust, , The reference thrust and reference fuel consumption per unit thrust are given under the same environmental vector without power generation extraction. Decision variables and environment vector The downward thrust, Decision variables and environment vector Fuel consumption per unit thrust under the following conditions These represent the output power of the shaft power extraction, turbofan bleed air turbine, and ramjet bleed air turbine, respectively. For power requirements, Decision variables and environment vector Lower critical section temperature, Decision variables and environment vector Lower rotating component rotational speed , Decision variables and environment vector Lower compression component Stability margin For critical sections Maximum temperature For rotating parts Maximum speed, For compression components Minimum stability margin.

3. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 2, characterized in that, Specifically, S4 involves: employing... Combinatorial gene coding strategies will incorporate decision variables and constraint processing auxiliary parameters Mapped to individual chromosome genes, In each generation of evolutionary iterations, the following algorithm flow is executed: S41, based on the constraint processing auxiliary parameters, the first... The chromosome genes of each individual in the population are dynamically repaired with step size constraints and their fitness is assessed. S42, after detecting the first When the environment of a population changes drastically, calculate the first generation. After assessing the overall environmental changes of the population, it entered S43, and after monitoring the first... Drastic environmental changes in the next generation or the first generation of population When the overall environmental change of a generation exceeds a threshold, it enters S45; S43, according to the The overall environmental change range of the population is adaptively adjusted in the first generation. Evolutionary parameters of a generational population; S44, Predicting the first The magnitude of the overall environmental change in the population of the first generation When the predicted value of the overall environmental change of the population exceeds the threshold, the S43 adaptive adjustment is performed. Evolutionary parameters of a generational population; S45, for the first The generation population performs a restart operation and sends feedback to S41.

4. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 3, characterized in that, In step S41, the constraint processing auxiliary parameters are applied to the first... The chromosome genes of each individual in the population undergo dynamic step-size constraint repair, specifically as follows: In the individual When power balance constraints are violated, the first individual of Genes dynamically adjust repair step size and update the first step. individual of Genes are , , The target deviation vector; In the individual When the modal feasible region constraint is violated, according to The decay rate is controlled to gradually approach the modal feasible region.

5. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 4, characterized in that, In step S41, the constraint processing auxiliary parameters are applied to the first... After dynamic step-size constraint repair of the chromosome genes of each individual in the population, fitness is evaluated according to the following method: using... and the Generational population environment vector The mechanism model drives the key cross-section output by the mechanism model. Temperature T i ( e) Rotating components rotational speed or stability margin If the safety constraints are still violated, the degree of violation is calculated and a penalty function is applied.

6. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 5, characterized in that, In S42, when the first... When the environment of a population changes drastically, calculate the first generation. The overall environmental change range of the population is as follows: No. Variation in the physical environment vector of the population Or the magnitude of change in the statistical properties of the algorithm's fitness When the threshold is exceeded, determine the first The environmental changes in the population generation are abnormal. , , For the first Generation population, first The environmental vector of the generation population , For the first Generation population, first Mean fitness of the generation population; After monitoring the first Frequency of environmental change in generation population When the threshold is exceeded, determine the first The environment of the population changed drastically in each generation, and the calculation of the first generation was carried out. The range of comprehensive environmental changes in the population , , , This represents the generation in which the last environmental mutation was detected. These are the normalized weighting coefficients.

7. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 6, characterized in that, In S43, according to the first The evolutionary parameters are adaptively adjusted based on the magnitude of overall environmental changes in the population, specifically: According to the The range of comprehensive environmental changes in the population Definition of the first Adaptive adjustment parameters of generation population , , Adjust the threshold for the set environmental change range; use Dynamic adjustment of the first Variation rate of the population With cross rate , No. When the environmental changes of a generation's population are extremely drastic, the population expands. Population size of the generation , , , , Based on the basic rate of variation, Based on the cross rate, Based on the basic population size, , , The adjustment factor is a positive number.

8. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 7, characterized in that, S44 predicts the first using a time series model. The magnitude of comprehensive environmental changes in the population.

9. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 8, characterized in that, The S45 is for the first The generation population will be restarted, specifically as follows: The current Pareto frontier optimal individual and its corresponding first Generational population environment vector Store in memory bank , ; Retrieve from the memory bank the first Record historical environment vectors that are similar to the environment vectors of the current generation population, and search the memory database for the first generation. Environment vector record The corresponding set of historical best individuals With the Elite individuals preserved in the population and randomly generated supplementary individuals Integration, Reconstruction Generation population for If no match is found Record historical environment vectors that are similar to the current population environment vectors, perform a local population restart, and only retain the first-generation environment vector. Elite individuals of a generation The remaining individuals are randomly reset, reconstructing the first... Generation population for , For the first time after random reset Individuals to the first Individual, The total number of individuals; For the reconstructed first Generation population Based on the adaptively adjusted evolutionary parameters determined in S43, selection, crossover, and mutation operations are performed to generate the next generation population.

10. The method for dynamic configuration of multi-source energy for a wide-speed-range aircraft according to claim 9, characterized in that, The retrieval of the first in the memory bank The method for recording historical environment vectors that are similar to the environment vectors of the next generation population is as follows: Calculate the first generation... Generational population environment vector With the first in the memory bank Environment vector record The weighted Euclidean distance.

Images