Design optimization method and system for ducted fan of vertical take-off and landing aircraft

Abstract

The invention relates to a design optimization method and system for a ducted fan of a vertical take-off and landing aircraft, in particular to application of a Bayesian optimization algorithm based on a variable acquisition function. According to the method, the overall efficiency of the vertical take-off and landing aircraft is improved, and the fuel consumption of a given flight mission profile is minimized by optimizing the design of the ducted fan. According to the method, the Bayesian optimization algorithm is introduced, the mixed variable optimization problem in the design of the ducted fan is efficiently processed, the optimization strategy is dynamically adjusted through the variable acquisition function, comprehensive performance optimization under different task profiles is achieved, the simulation calculation cost is remarkably reduced, and the optimization efficiency and the result quality are improved.

Description

Technical Field

[0001] The present invention belongs to the field of aircraft design, and particularly relates to an optimization method and system for the design of a ducted fan of a vertical takeoff and landing aircraft, especially an optimization method based on a Bayesian optimization algorithm with a variable acquisition function. Background Art

[0002] Due to the advantages of vertical takeoff and landing capabilities and high high-speed cruise flight efficiency, vertical takeoff and landing aircraft have great application potential in both military and civilian fields. Since vertical takeoff and landing aircraft face greater challenges in terms of power and energy requirements, it is very important to improve the overall efficiency of vertical takeoff and landing aircraft. One promising method is to use the ducted fan concept because it has higher aerodynamic performance and noise shielding effects. A large number of studies have shown that duct parameters and fan parameters have an important impact on the aerodynamic performance of ducted fans, which are the key factors affecting the overall efficiency of ducted fan vertical takeoff and landing aircraft. However, the optimization of duct and fan parameters for the entire flight mission of a vertical takeoff and landing aircraft is a complex optimization problem with a large computational amount involving continuous and discrete design variables.

[0003] Complex interaction between discrete and continuous variables: In the optimization of ducted fans, continuous design variables (such as the geometry of the blade, tip clearance, inlet radius of the duct, length, thickness, outlet expansion angle) and discrete design variables (such as the number of blades, selection of fan airfoil) are involved simultaneously. The interaction between these variables makes the optimization problem very complex, and it is impossible to obtain gradient information during the model training process. Existing optimization methods are prone to falling into local optimal solutions during optimization, and have low computational efficiency. Especially when dealing with large-scale, non-linear complex systems, the optimization results are often restricted by computational resources and computational time.

[0004] The optimization of ducted fans involves a large amount of calculations: In aircraft design, the aerodynamic performance of fans and ducts is a dynamic and complex process, and the design of ducted fans needs to consider multiple working conditions simultaneously. Solving this type of optimization problem with multiple types of design variables requires a large number of simulation calculations. Especially when it is necessary to consider the aerodynamic performance of the aircraft at multiple flight stages (such as takeoff, hover, cruise, etc.), the computational amount and time cost are extremely large. Traditional optimization methods are often inefficient and difficult to obtain an optimized solution in a short time. Therefore, existing optimization methods usually perform optimization under a single working condition, such as only optimizing the flight performance at a certain stage (for example, only optimizing the hover or cruise state).

[0005] Contradiction between aerodynamic performance and multi-task stages: The design of ducted fans needs to balance the aerodynamic performance requirements at different flight stages. A large thrust is required during takeoff and hover stages, while low energy consumption and high aerodynamic efficiency are required during the cruise stage. Existing optimization methods often focus on optimization under a single working condition and are difficult to balance the comprehensive aerodynamic performance under multi-task stages.

[0006] Therefore, the present invention provides a design optimization method for the ducted fan of a vertical takeoff and landing aircraft, which can comprehensively optimize the aerodynamic performance in multiple mission phases such as takeoff, hover, acceleration, and cruise, meet the complex requirements of multiple tasks and multiple working conditions, and at the same time support the fusion optimization of continuous and discrete design variables and adapt to the interaction relationship between complex variables. Summary of the Invention

[0007] Aiming at the key technical problems existing in the design of the ducted fan of the existing vertical takeoff and landing aircraft, the present invention proposes a design method based on the Bayesian optimization algorithm with a variable acquisition function. This method aims to solve the problems of complex interaction between discrete and continuous design variables, large computational amount and low optimization efficiency, contradiction in comprehensive optimization performance in multiple mission phases, optimization of black-box objective functions, and insufficient adaptability to dynamic complex environments. By introducing the Bayesian optimization algorithm, it efficiently processes the hybrid optimization problem in the design of the ducted fan, and uses the Gaussian process model and variable acquisition function to dynamically adjust the optimization strategy to achieve comprehensive performance optimization under multiple objectives and multiple mission phases, significantly reducing the simulation calculation cost, improving the optimization efficiency and result quality, and thus providing a better design solution for the application of vertical takeoff and landing aircraft.

[0008] The present invention uses the Bayesian optimization algorithm to efficiently process the hybrid optimization problem in the design of the ducted fan. The Bayesian optimization algorithm is mainly based on the Gaussian process model, models the objective function through the existing sample data, and calculates the posterior probability distribution of the existing data points, so as to provide predictions for the mean and variance of the objective function in the unsampled region.

[0009] The technical solution of the present invention is as follows: A design optimization method for the ducted fan of a vertical takeoff and landing aircraft, which uses the Bayesian optimization algorithm with a variable acquisition function to solve the hybrid design variables in the design of the ducted fan, specifically including: Setting the hybrid design variables of the ducted fan of the vertical takeoff and landing aircraft and their value ranges; Constructing an objective function based on the hybrid design variables, optimization objectives and constraint conditions; Obtaining initial data points through simulation calculation and initializing the sampling data set; Constructing an analysis model based on the sampling data set for fitting the objective function; Using the variable acquisition function to iteratively optimize the analysis model to obtain the optimal values of the hybrid design variables; Wherein, the hybrid design variables include fan design variables, duct design variables and fan-duct interaction design variables; The sampling data set is composed of data points; the data points include a set of hybrid design variable values and their corresponding objective function values; The variable acquisition function includes a parameter exploration weight κ; the exploration weight κ decays linearly with the increase of the number of iterations.

[0010] The present invention dynamically adjusts the optimization strategy through the Gaussian process model and the variable acquisition function. The core of the variable acquisition function is to dynamically adjust the weights of exploration and development during the optimization process to achieve a comprehensive search and focused development of the design variable space. In the early stage of optimization, a larger exploration weight is tended to be selected to fully search the design space; in the later stage of optimization, the exploration weight is gradually reduced, more attention is paid to development, and high potential solutions near known points are concentrated to quickly converge to the global optimal solution.

[0011] Furthermore, in the design optimization method of the present invention, the use of a variable acquisition function to iteratively optimize the analysis model to obtain the optimal value of the hybrid design variable specifically includes: Construct variable acquisition functions based on analytical models; A set of mixed design variable evaluation values ​​are obtained by maximizing the variable acquisition function value; The corresponding objective function value is obtained through simulation calculation; Adding the hybrid design variable evaluation value and the corresponding objective function value as new data points to the sampling data set; updating the analytical model based on the sampled data set; When the absolute value of the difference between the variable acquisition function value in this iteration and the variable acquisition function value in the previous iteration is less than the set threshold, the iterative optimization ends, and the evaluation value of the hybrid design variable in this iteration is the optimal value of the hybrid design variable; otherwise, repeat the above steps to continue the iterative optimization.

[0012] Furthermore, in the design optimization method of the present invention, the fan design variables include the twist angle of the fan blade at 75% radius and the fan airfoil; the fan airfoil is selected from one or more of a set of predetermined mainstream fan airfoils.

[0013] Furthermore, in the design optimization method of the present invention, the duct design variables include CST parameters of the duct airfoil.

[0014] Furthermore, in the design optimization method of the present invention, the fan-duct interaction design variables include the chordwise distance from the duct leading edge to the fan disk, and the gap between the duct and the fan tip.

[0015] Furthermore, in the design optimization method of the present invention, the optimization objectives include minimizing fuel consumption and power of the vertical take-off and landing aircraft in a given mission profile; the mission profile includes specific operating conditions and a complete flight mission.

[0016] Furthermore, in the design optimization method of the present invention, the specific operating conditions include cruising and hovering.

[0017] Further, in the design optimization method of the present invention, the constraint conditions include thrust requirements and tip Mach number limitations.

[0018] Further, in the design optimization method of the present invention, the simulation calculation includes obtaining the corresponding objective function value based on a set of hybrid design variable values, combining the constraint conditions and the mission profile, and through fan aerodynamic performance analysis and ducted fan performance analysis.

[0019] The design optimization process of the ducted fan in the present invention is mainly divided into two target scenarios. One is to optimize for specific operating conditions (cruise or hover), with the goal of achieving the minimum power consumption of the ducted fan under these specific conditions; the other is to optimize for the entire flight mission profile of the ducted fan, with the goal of minimizing the energy consumption of the ducted fan when completing the selected flight mission. Through this phased and multi-objective optimization method, not only can the performance requirements of the ducted fan in different mission stages be met, but also the overall energy efficiency of the ducted fan can be improved.

[0020] The present invention also provides a ducted fan design optimization system for a vertical takeoff and landing aircraft, which can implement the ducted fan design optimization method of the present invention, including: Variable design module: used to set the hybrid design variables of the ducted fan of the vertical takeoff and landing aircraft and their value ranges; Objective function construction module, used to construct the objective function based on the hybrid design variables, optimization objectives and constraint conditions; Sampling data set initialization module: used to obtain the initial data points through simulation calculation and initialize the sampling data set; Analysis model construction module: used to construct an analysis model based on the sampling data set for fitting the objective function; Iterative optimization module: used to iteratively optimize the analysis model using the variable acquisition function to obtain the optimal values of the hybrid design variables; Among them, the hybrid design variables include fan design variables, duct design variables and fan-duct interaction design variables; The sampling data set is composed of data points; the data points contain a set of hybrid design variable values and their corresponding objective function values; The variable acquisition function includes a parameter exploration weight κ; the exploration weight κ linearly decays as the number of iterations increases.

[0021] The Bayesian optimization algorithm based on the variable acquisition function provides an efficient and intelligent optimization tool for ducted fan design. By dynamically adjusting the acquisition function and adapting to complex variable types and multi-task working condition requirements, this method not only provides an innovative solution for the ducted fan design of vertical takeoff and landing aircraft, significantly improves the design efficiency, but also provides strong support for realizing the multi-stage performance optimization of vertical takeoff and landing aircraft.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Complex variable fusion optimization, which supports the fusion optimization of continuous and discrete design variables, adapts to the interaction relationship between complex variables, and provides an effective tool for solving multi-variable optimization problems in practical engineering.

[0023] 2. Optimization efficiency and performance improvement. By dynamically adjusting the exploration and exploitation weights in the variable acquisition function, this method can quickly converge to near the global optimal solution within fewer iterations, effectively avoiding getting stuck in local optimal solutions during the optimization process. This not only improves the optimization efficiency but also significantly reduces the computational cost, especially suitable for computationally complex problems such as ducted fan design, enabling the efficient completion of optimization tasks with limited computational resources.

[0024] 3. Multi-task and multi-working condition adaptability. The present invention can comprehensively consider the aerodynamic performance requirements of vertical takeoff and landing aircraft in multiple mission stages such as takeoff, hover, acceleration, and cruise, and meet the complex requirements of multi-tasks and multi-working conditions. This makes the ducted fan design not only perform well under a single working condition but also maintain high efficiency throughout the flight mission, improving the overall applicability and flexibility of the aircraft. Description of the Drawings

[0025] By describing the embodiments of the present application in more detail in combination with the drawings, the above and other objects, features, and advantages of the present application will become more obvious. The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. They are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation to the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0026] Figure 1 It is the design optimization flow chart of the ducted fan under specific working conditions in the embodiments of the present invention.

[0027] Figure 2 It is the design optimization flow chart of the ducted fan for the complete flight mission in the embodiments of the present invention.

[0028] Figure 3 It is the schematic diagram of the set of predetermined mainstream fan airfoils in the embodiments of the present invention.

[0029] Figure 4This is an iterative schematic diagram of design variables and optimization objectives during the hover optimization process in an embodiment of the present invention.

[0030] Figure 5 This is an iterative schematic diagram of design variables and optimization objectives during the cruise optimization process in an embodiment of the present invention.

[0031] Figure 6 This is a comparison schematic diagram of ducted airfoils before and after optimization during cruise and hover in an embodiment of the present invention.

[0032] Figure 7 This is a comparison schematic diagram of the initial case and the optimized case during cruise and hover in an embodiment of the present invention.

[0033] Figure 8 This is a three - dimensional view of the NASA UAM tilt - ducted eVTOL aircraft in an embodiment of the present invention.

[0034] Figure 9 This is a flight mission profile diagram in an embodiment of the present invention.

[0035] Figure 10 This is an iterative schematic diagram of design variables and optimization objectives during the complete flight mission optimization process in an embodiment of the present invention.

[0036] Figure 11 This is a comparison diagram of ducted airfoils before and after the complete flight mission optimization in an embodiment of the present invention. Detailed implementation manners

[0037] Next, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0038] Embodiment 1 The Bayesian Optimization (BO) algorithm based on variable acquisition function is an efficient algorithm for optimizing black - box functions (objective functions without explicit expressions), and is suitable for dealing with complex optimization problems where continuous and discrete design variables coexist. In the design of ducted fans for vertical take - off and landing aircraft, this algorithm can quickly find the global optimal solution by dynamically adjusting the balance between exploration and exploitation.

[0039] In an embodiment of the present invention, FIGS. 1 and 2 show the complete process of ducted fan design using the Bayesian optimization algorithm based on variable acquisition function, as well as the specific implementation paths in different working conditions and flight mission optimizations: First, set the hybrid design variables and their value ranges for the ducted fan of a vertical takeoff and landing aircraft, including fan design variables, duct design variables, and fan-duct interaction design variables; the fan design variables include fan airfoil and the twist angle of the fan blade at 75% radius ( ), the fan airfoil is selected from 12 mainstream fan airfoils and represented by discrete digital variables; the duct design variables include duct airfoil, and the class-shape transformation (CST) parameterization method is used to generate various duct geometries by changing the CST parameters to fully study the influence of duct shape on the performance of the ducted fan; the fan-duct interaction design variables include the chordwise distance from the leading edge of the duct to the fan disk ( ), and the gap between the duct and the fan tip ( ), which are used to describe the relationship between the fan and the duct.

[0040] Then, construct an objective function based on the hybrid design variables, optimization objectives, and constraint conditions; the optimization objectives and constraint conditions are related to the given mission profile. One is to optimize for specific operating conditions (cruise or hover), and the goal is to achieve the minimum power consumption of the ducted fan under these specific conditions; the other is to optimize for the entire flight mission of the ducted fan, and the goal is to minimize the energy consumption of the ducted fan when completing the selected flight mission.

[0041] Obtain the initial data points through simulation calculations and initialize the sampling data set; among them, the aerodynamic performance of the fan is obtained through XFOIL calculations; the overall performance of the ducted fan is obtained through the DFDC calculation method.

[0042] Construct an analysis model based on the sampling data set to fit the objective function; Bayesian optimization mainly uses the Gaussian Process (GP) model. Without knowing the explicit expression of the objective function, it uses the existing sample data points to model the objective function, and uses Gaussian Process Regression (GP, Regression) to calculate the posterior probability distribution of the existing data points, so as to provide predictions for the mean (μ) and variance (σ²) of the objective function in the unsampled region.

[0043] Iteratively optimize the analysis model using a variable acquisition function to obtain the optimal values of the mixed design variables; the acquisition function balances exploration and exploitation to determine the location of the next sampling point. Exploration attempts to select points far from the known points as the reference points for the next iteration, that is, attempts to explore unknown regions, where the distribution of points will be as uniform as possible. Exploitation selects points as close as possible to the known points as the reference points for the next iteration, that is, attempts to mine points around the known points, and the distribution of points will present as a dense area, easily falling into local maxima. Since the choice of exploration and exploitation methods has a great impact on the acquisition function and the optimization process, as the optimization process progresses, appropriately adjust the exploration and exploitation weights to quickly find the global optimal solution. Therefore, the embodiments of the present invention adopt a variable acquisition function to dynamically adjust the weights of exploration and exploitation during the optimization process. In the initial stage of optimization, it is inclined to select a larger exploration weight (κ is larger) to fully search the design space and cover the unexplored regions; in the later stage of optimization, gradually reduce the exploration weight (κ is smaller), pay more attention to exploitation, and concentrate on mining high-potential solutions near the known points, so as to quickly converge to the global optimal solution.

[0044] In another embodiment of the present invention, the calculation formula of the variable acquisition function is: where κ is the exploration weight, used to adjust the exploration intensity. The larger the κ value, the more inclined to explore unknown regions.

[0045] In another embodiment of the present invention, a linear decay function is used to describe the change of the exploration weight κ with the number of iterations t, and the mathematical expression is as follows: Exploration stage: In the first 100 times, κ = 10, for extensive exploration.

[0046] Transition stage: From 100 to 400 times, as the number of iterations increases, κ linearly decays from 10 to 0.5, gradually shifting from exploration to exploitation.

[0047] Exploitation stage: From 400 to 500 times, κ = 0.5, for fine exploration to find the optimal solution.

[0048] By dynamically adjusting the acquisition function mechanism for exploration and exploitation, the Bayesian optimization (BO) method can achieve efficient search and optimization of complex combinations of design variables, effectively balancing the weights of global search and local optimization. In the initial stage of optimization, it tends to explore areas that have not been fully studied, paying more attention to the uncertainty of sampling points. In the later stage of optimization, it gradually tilts towards exploiting areas with known potential excellent solutions, focusing on the expected benefits of sampling points. Through the design of variable acquisition functions, the BO method can avoid falling into local optimal solutions and quickly converge to the global optimal solution while exploring the entire design space.

[0049] Example 2 In another embodiment of the present invention, specific parameters are used to optimize specific operating conditions and a complete flight mission respectively. Table 1 lists the complete optimization formulas. β 0.75 、 X disκ 、 T gap The number of fan airfoils and the CST parameters of the ducted airfoil are used as design variables, the thrust requirement and the tip Mach number are used as design constraints, and the two optimization objectives are the minimum power (i.e., maximum efficiency) of the working thrust of the ducted fan and the minimum fuel consumption for a given flight mission.

[0050] Table 1 Formulation of the ducted fan optimal problem Twelve classic airfoils were selected for fan optimization, as Figure 3 shown. The candidate airfoils include NACA 0012, 0015, 0018, the NACA 64 series airfoils already used in the tiltrotor XV-15, as well as the NACA 24 and 44 series, arad10, clarκ-y, e851, hq07, raf6.

[0051] In the optimized design of the ducted fan, in the initial stage of design, the geometric parameter combinations of the fan and the duct are widely explored to ensure coverage of the global search space. A large κ will be selected to explore the entire design space (the number of iterations t ≤ 100, κ = 10). In the middle stage of design, κ linearly decays as the number of iterations increases (100 < t ≤ 400, ), and in the later stage of design, the high-potential areas are concentratedly optimized to improve the aerodynamic performance and energy efficiency of the ducted fan. κ will be adjusted to a smaller value (the number of iterations t > 400, κ = 0.05) to find the optimal solution.

[0052] 1) Optimization results of the ducted fan under specific operating conditions The results shown in Table 2 indicate that a total ducted fan efficiency of over 70% can be achieved in both the cruise scenario and the hover scenario. The ducted power decreases by 24.4% and 12.2% in the cruise and hover states respectively. None of the optimized design variables reach the boundary, indicating that the selection of the optimization boundary is reasonable.

[0053] Table 2 Results of ducted fan optimization during cruise and hover Figure 4 and Figure 5 respectively show the iteration of design parameters and design objectives in the hover and cruise optimization cases. In the early stage of optimization, making the BO more exploratory enables the calculation points to better fill the design space, effectively avoiding the results from falling into local optima. In the later stage of optimization, the BO becomes more exploitative, exploring in more detail the regions where the global optimum may appear and comprehensively investigating the performance of the ducted fan.

[0054] Due to the large difference in the free takeoff and landing speeds between the cruise and hover states, there are also significant differences in the optimized ducted airfoils in the two cases. Figure 6 A comparison of the ducted airfoils before and after optimization is given.

[0055] Figure 7 Using DFDC, a comparison diagram of the surface pressure vectors of the ducted fan under cruise and hover conditions before and after optimization is drawn to more intuitively understand the changes in the performance of the ducted fan. In the hover and cruise stages, although the optimized ducted lip curvature radius is smaller, resulting in a loss of part of the pressure difference between the upper and lower surfaces, the large-thickness ducted airfoil increases the pressure difference between the upper and lower surfaces in the last three-quarters section of the duct, increasing the total ducted thrust. Due to the change of design parameters, the air velocity in the duct changes, directly affecting the pressure distribution on the surface of the ducted fan, bringing an increase in the ducted thrust for the optimization of the ducted airfoil.

[0056] 2) Overall optimization of a vertical takeoff and landing aircraft considering flight missions Taking the NASA UAM tilt-ducted eVTOL aircraft as a reference vehicle to complete the benchmark vertical takeoff and landing of flight mission optimization. Figure 8 The geometric layout of the reference tilt-ducted vertical takeoff and landing aircraft is shown, and Table 3 gives the top-level aircraft parameters of the reference VTOL aircraft.

[0057] Table 3 Top-level parameters of the NASA UAM tilt-ducted eVTOL vertical takeoff and landing reference vehicle The mission profile of the tilt-ducted vertical takeoff and landing aircraft is based on the data of the NASA UAM tilt-ducted eVTOL reference vehicle. The aircraft takes off at the maximum takeoff weight. The mission profile diagram is shown in Figure 9, the mission profile consists of two identical takeoff and landing segments (1 - 9 and 10 - 18) and a 20 - minute cruise reserve range. According to the flight state of the aircraft, the power and energy consumption of the ducted fan are calculated in Table 4.

[0058] Table 4 Mission Segment Attributes Table 5 presents the optimization results of the ducted fan considering the entire flight mission profile. When minimizing the fuel consumption of the entire flight mission is set as the design goal, the efficiency of the optimized ducted fan during cruise and hover is increased to 77.7% and 81.7% respectively. The power of the ducted fan is significantly reduced, the energy consumption is reduced by 18.1%, and the total energy used is reduced from 2763.9 MJ to 2263.0 MJ.

[0059] Table 5 Optimization Results of the Ducted Fan Considering the Mission Profile Figure 10 The iteration of the design parameters and the optimized design goals during the optimization process is shown. Figure 11 The ducted airfoil shapes before and after optimization are shown. The optimization effect of the ducted fan is lower than that of optimizing a single cruise or hover scenario, but the overall performance in the entire mission is improved compared to the cruise and hover scenarios, achieving improvements in both hover and cruise performance. In the considered mission, the cruise phase occupies the main part of the mission, and the optimization results actually tend more towards the cruise optimization results.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for optimizing the design of a ducted fan for a vertical take-off and landing aircraft, characterized in that: The Bayesian optimization algorithm based on variable acquisition function is used to solve the mixed design variables in the ducted fan design, including: Setting the hybrid design variables and their value ranges of the ducted fan of a vertical take-off and landing aircraft; Construct an objective function based on mixed design variables, optimization objectives, and constraints; Obtain initial data points through simulation calculation and initialize the sampling data set; Build an analytical model based on the sampled data set to fit the objective function; The analysis model is iteratively optimized using variable acquisition functions to obtain the optimal values ​​of hybrid design variables; Wherein, the hybrid design variables include fan design variables, duct design variables and fan-duct interaction design variables; The sampling data set is composed of data points; the data points include a set of mixed design variable values ​​and their corresponding objective function values; The variable acquisition function includes a parameter exploration weight κ; the exploration weight κ decays linearly with the increase of the number of iterations.

2. The design optimization method according to claim 1, characterized in that: The analysis model is iteratively optimized using variable acquisition functions to obtain the optimal values ​​of the hybrid design variables, including: Construct variable acquisition functions based on analytical models; A set of mixed design variable evaluation values ​​are obtained by maximizing the variable acquisition function value; The corresponding objective function value is obtained through simulation calculation; Adding the hybrid design variable evaluation value and the corresponding objective function value as new data points to the sampling data set; updating the analytical model based on the sampled data set; When the absolute value of the difference between the variable acquisition function value in this iteration and the variable acquisition function value in the previous iteration is less than the set threshold, the iterative optimization ends, and the evaluation value of the hybrid design variable in this iteration is the optimal value of the hybrid design variable; otherwise, repeat the above steps to continue the iterative optimization.

3. The design optimization method according to claim 1, characterized in that: The fan design variables include a twist angle of the fan blade at 75% radius and a fan airfoil; the fan airfoil is selected from one or more of a set of predetermined mainstream fan airfoils.

4. The design optimization method according to claim 1, characterized in that: The duct design variables include CST parameters of the duct airfoil.

5. The design optimization method according to claim 1, characterized in that: The fan-duct interaction design variables include the chordwise distance from the duct leading edge to the fan disk and the gap between the duct and the fan tip.

6. The design optimization method according to claim 1, characterized in that: The optimization goal is to minimize the fuel consumption or power of the vertical take-off and landing aircraft in a given mission profile; the mission profile includes specific operating conditions and a complete flight mission.

7. The design optimization method according to claim 6, characterized in that: The specific operating conditions include cruising and hovering.

8. The design optimization method according to claim 1, characterized in that: The constraints include thrust requirements and blade tip Mach number limits.

9. The design optimization method according to claim 1, characterized in that: The simulation calculation includes obtaining corresponding objective function values ​​through fan aerodynamic performance analysis and ducted fan performance analysis based on a set of mixed design variable values, combined with constraint conditions and mission profiles.

10. A vertical take-off and landing aircraft ducted fan design optimization system, which can implement the vertical take-off and landing aircraft ducted fan design optimization method according to any one of claims 1 to 9, characterized in that: include: Variable design module: used to set the hybrid design variables and their value ranges of the ducted fan of the vertical take-off and landing aircraft; An objective function building module is used to build an objective function based on mixed design variables, optimization objectives and constraints; Sampling data set initialization module: used to obtain initial data points through simulation calculation and initialize the sampling data set; Analysis model building module: used to build an analysis model based on the sampled data set to fit the objective function; Iterative optimization module: used to iteratively optimize the analysis model using variable acquisition functions to obtain the optimal values ​​of hybrid design variables; Wherein, the hybrid design variables include fan design variables, duct design variables and fan-duct interaction design variables; The sampling data set is composed of data points; the data points include a set of mixed design variable values ​​and their corresponding objective function values; The variable acquisition function includes a parameter exploration weight κ; the exploration weight κ decays linearly with the increase of the number of iterations.

Images