Multi-airfoil combined axial flow fan blade design method and system based on CST parameterization

Abstract

The invention provides a multi-airfoil combined axial flow fan blade design method and system based on CST parameterization, and relates to the technical field of axial flow fan blade optimizing.The multi-airfoil combined axial flow fan blade design method comprises the following steps that S1, a plurality of characteristic sections are arranged in the radial direction of a blade, and the airfoil corresponding to each characteristic section is determined; s2, geometric modeling is conducted on the airfoils of all the characteristic sections through a CST parameterization method, and description parameter sets corresponding to all the airfoils are obtained; s3, performing aerodynamic performance simulation on the airfoils through the description parameter group to determine the attack angle of each airfoil; s4, the airfoils are combined at the corresponding attack angles, smooth transition cambered surfaces are generated between the airfoils, and an initial blade model is obtained; and S5, optimizing the initial blade model to obtain a target blade model by adjusting the description parameter group of each airfoil by taking the maximization of the lift-drag ratio as a target. Compared with a traditional single arc thin plate blade profile design, the pneumatic performance of the blade is more optimized, and the problem that a single blade profile is difficult to cope with complex pneumatic conditions where different sections are located is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of axial flow fan blade optimization, and in particular to a multi-airfoil combined axial flow fan blade design method and system based on CST parameterization. Background Art

[0002] Axial flow fans are widely used in industrial and residential ventilation systems due to their high air volume and low energy consumption. With increasing demands for energy efficiency and environmental protection, how to further improve the aerodynamic performance and operating efficiency of axial flow fans through optimized design has become a key technical issue.

[0003] Existing axial flow fan blade designs usually use empirical formulas or traditional aerodynamic methods, especially traditional blade shapes are mostly arc thin plate blades or single airfoils. Although these methods can meet basic air volume requirements, they have certain limitations under high-efficiency operation and complex working conditions. In traditional designs, blades usually adopt arc thin plate or single airfoil design, and maintain the same or similar airfoil along the radial direction of the blade. However, since the aerodynamic conditions of the blades at different radial positions vary significantly, this design cannot effectively adapt to the different aerodynamic environments of each cross-section, resulting in flow separation and increased vortex, which in turn affects the aerodynamic performance of the fan. Under high-efficiency working conditions, the lift-to-drag ratio of the blades is unstable and cannot continuously provide ideal air volume and efficiency. Summary of the Invention

[0004] In view of this, the present invention proposes a multi-airfoil combined axial flow fan blade design method and system based on CST parameterization. By dividing the blade into four different cross-sections along the radial direction, a combination of multiple airfoils is designed to more precisely adapt to the aerodynamic environment at different locations, and the airfoil shape of each cross-section is optimized through the CST parameterization method. Each cross-section adopts a different airfoil design, so that the entire blade can achieve the best lift-to-drag ratio and aerodynamic performance at different radial positions. Compared with the traditional single arc thin plate blade design, the blade aerodynamic performance is further optimized, effectively avoiding the problem that a single airfoil is difficult to cope with the complex aerodynamic conditions of different cross-sections.

[0005] The technical solution of the present invention is achieved as follows:

[0006] In one aspect, the present invention provides a method for designing blades of a multi-airfoil combined axial flow fan based on CST parameterization, comprising the following steps:

[0007] S1 sets multiple characteristic sections along the radial direction of the blade and determines the airfoil corresponding to each characteristic section;

[0008] S2 uses the CST parameterization method to geometrically model the airfoils of all characteristic sections and obtain the corresponding descriptive parameter groups for each airfoil;

[0009] S3 simulates the aerodynamic performance of the airfoil by describing the parameter group to determine the angle of attack of each airfoil;

[0010] S4 combines the airfoils at corresponding angles of attack and generates smooth transition cambered surfaces between the airfoils to obtain the initial blade model.

[0011] S5 takes the lift-to-drag ratio maximization as the goal and optimizes the initial blade model by adjusting the descriptive parameter groups of each airfoil to obtain the target blade model.

[0012] On the basis of the above technical solution, preferably, the number of the characteristic cross sections is four, and the first and last characteristic cross sections are located at the blade root and the blade top respectively.

[0013] Based on the above technical solution, preferably, step S3 includes the following sub-steps:

[0014] S31 builds the simulation environment;

[0015] In the simulation environment, S32 performs CFD scans at multiple attack angles for each airfoil, and calculates and stores lift-to-drag ratio and pressure distribution data in parallel;

[0016] S33 determines the angle of attack of each airfoil based on the lift-to-drag ratio and the pressure distribution data.

[0017] Based on the above technical solution, preferably, step S4 includes the following sub-steps:

[0018] S41 arranges each airfoil in the radial direction of the blade and performs twist positioning according to the angle of attack;

[0019] S42 uses B-spline curves between adjacent airfoils to generate a smoothly transitioned blade surface to connect the airfoils and obtain the intermediate blade model;

[0020] S43 verifies the surface continuity of the intermediate blade model and corrects local mutations to obtain the initial blade model.

[0021] Based on the above technical solution, preferably, step S5 includes, with the goal of maximizing the lift-to-drag ratio, using an optimization algorithm, through multiple rounds of iterations, gradually adjusting the CST parameters so that each airfoil achieves optimal performance at the corresponding angle of attack.

[0022] On the basis of the above technical solution, preferably, step S6 is further included to verify the aerodynamic performance of the target blade model, and based on the simulation results, evaluate the lift-to-drag ratio, flow separation and resistance characteristics of the optimized blade under different working conditions.

[0023] On the basis of the above technical solution, preferably, the CST parameterization method defines the airfoil profile by the product of the class function and the shape function, which is expressed as:

[0024] y(ψ)=C(ψ)·S(ψ)+Δy te ·ψ

[0025] Where ψ = x / c, ψ represents the normalized chord-wise position, x is the chord-wise position, c is the chord length, y(ψ) is used to define the airfoil profile, C(ψ) is the class function, S(ψ) is the shape function, and Δy te is the trailing edge thickness parameter.

[0026] Further preferably, the class function C(ψ) is expressed as:

[0027]

[0028] Among them, N1 and N2 are the leading edge index and the trailing edge index respectively.

[0029] Further preferably, the shape function S(ψ) is expressed as:

[0030]

[0031] Among them, B i is the weight parameter, B n,i is the Bernstein basis function.

[0032] On the other hand, the present invention provides a multi-airfoil combined axial flow fan blade design system based on CST parameterization, which is embedded with the above-mentioned multi-airfoil combined axial flow fan blade design method.

[0033] The CST parameterized multi-airfoil combined axial flow fan blade design method and system of the present invention have the following beneficial effects compared with the prior art:

[0034] (1) By dividing the blade into four different sections along the radial direction, a combination of multiple airfoils is designed to more precisely adapt to the aerodynamic environment at different locations, and the airfoil shape of each section is optimized through the CST parameterization method. Each section adopts a different airfoil design, so that the entire blade can achieve the best lift-to-drag ratio and aerodynamic performance at different radial positions. Compared with the traditional single arc thin plate blade design, the blade aerodynamic performance is more optimized, effectively avoiding the problem that a single airfoil is difficult to cope with the complex aerodynamic conditions of different sections.

[0035] (2) By simulating and optimizing the airfoil of each blade section, the shape and angle of attack of each airfoil are automatically adjusted to optimize the overall aerodynamic efficiency. Through the simulation and optimization process, the present invention can optimize the airfoil shape on each section at different angles of attack, ensuring that the lift-to-drag ratio of the blade can be maximized under various working conditions, reducing energy consumption and improving the working efficiency of the fan. Compared with traditional arc thin-plate blades and single airfoil designs, the present invention has significant advantages in improving aerodynamic performance, adapting to complex working conditions, improving design efficiency and flexibility, etc., and can effectively improve the overall performance and energy efficiency of axial flow fans. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 Schematic diagram of the steps of the CST parameterized multi-airfoil combined axial flow fan blade design method of the present invention;

[0038] Figure 2 Schematic diagram comparing the original arc thin plate type and multi-airfoil combined impeller structures of the multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0039] Figure 3 A schematic diagram of a multi-airfoil combined blade of a multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0040] Figure 4 A schematic diagram showing the chord length and angle of attack of the multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0041] Figure 5 The four-airfoil distribution diagram of the combined blade of the multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0042] Figure 6 A comparison diagram of the original airfoil and the optimized airfoil of the multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0043] Figure 7 The surface pressure cloud diagrams of the original blade impeller and the multi-airfoil combined blade impeller of the multi-airfoil combined axial flow fan blade design method based on CST parameterization of the present invention;

[0044] Figure 8These are the speed and cloud diagrams of the original blade impeller and the multi-airfoil combined blade impeller of the CST parameterized multi-airfoil combined axial flow fan blade design method of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] There are three problems in the existing technology: 1. The limitations of arc thin plate blade profile and single airfoil design; 2. The complexity of blade design; 3. The low efficiency of aerodynamic calculation and optimization.

[0047] Regarding question 1, traditional blade designs typically employ thin, circular plates or a single airfoil, maintaining the same or similar airfoil profile along the blade's radial direction. However, because the aerodynamic conditions at different radial positions vary significantly, this design cannot effectively adapt to the varying aerodynamic environments across each section, leading to flow separation and increased vorticity, which in turn affects the fan's aerodynamic performance. Under high-efficiency operating conditions, the blade's lift-to-drag ratio exhibits instability, making it impossible to consistently deliver the desired air volume and efficiency.

[0048] Regarding question 2, traditional methods for optimizing blade design under different operating conditions typically rely on experience and manual adjustment of geometric parameters, resulting in a complex and inefficient design process. When designing multiple cross-sections and airfoils, manual parameter adjustment is not only time-consuming and labor-intensive, but the optimization results are also easily limited by the designer's experience, making it difficult to achieve globally optimal aerodynamic performance.

[0049] Regarding question 3, in traditional blade design, aerodynamic analysis tools have low computational efficiency, and the optimization process often requires numerous iterative experiments. This repeated calculation and adjustment not only consumes time and resources, but also exponentially increases computational complexity, especially in complex designs with multiple cross-sections and airfoil combinations, severely limiting design and optimization efficiency.

[0050] like Figure 1-8 As shown, the CST parameterized multi-airfoil combined axial flow fan blade design method of the present invention solves the above three problems, and specifically includes steps S1-S6.

[0051] Step S1: setting a plurality of characteristic sections along the radial direction of the blade, and determining the airfoil corresponding to each characteristic section.

[0052] In this step, the characteristic sections of the blades are divided and the airfoils are selected. Specifically, in some embodiments, the blades can be divided radially into four characteristic sections according to the geometric shape of the axial fan blades. The end of the axial fan blade close to the rotating axis is the blade root, and the end away from the rotating axis is the blade top. The axial fan blades are divided into four blade sections along the direction from the blade root to the blade top. These characteristic sections will be used for subsequent optimization and aerodynamic analysis. For the sake of convenience, the four airfoils set in this embodiment all adopt BE airfoils, and the four airfoils are respectively the first airfoil, the second airfoil, the third airfoil and the fourth airfoil from the blade root to the blade tip.

[0053] In a specific embodiment, the blade is divided into four different characteristic sections along the radial direction, and airfoil selection is performed respectively according to the aerodynamic conditions of each characteristic section, wherein:

[0054] a) The first and second sections are located in the low-speed region near the blade root and use the same BE6308B airfoil design to optimize lift-to-drag ratio;

[0055] b) The third and fourth sections are located in the high-speed area of the blade near the tip and use different airfoil designs, BE6306B and BE7404B respectively, to reduce drag.

[0056] In addition, in this step, it is necessary to divide the angle of attack range of each airfoil. The angle of attack range can be selected according to actual needs. This setting is to improve the subsequent angle of attack selection and airfoil optimization efficiency. Specifically, it can be set to the first airfoil angle of attack of -5-15°, the second airfoil angle of attack of 20-60°, the third airfoil angle of attack of 15-50°, and the fourth airfoil angle of attack of 10-40°.

[0057] Dividing the blade into four sections effectively simplifies design complexity. Compared to more sections, four sections provide a better balance between design and optimization efficiency. It can finely adapt to aerodynamic changes without significantly increasing computational complexity and design workload.

[0058] Step S2: Use the CST parameterization method to perform geometric modeling on the airfoils of all characteristic sections to obtain the corresponding descriptive parameter groups of each airfoil.

[0059] Airfoil parameterization is a set of mathematical models used to describe and define airfoil shapes. It has widespread applications in fields such as aviation, aerospace, and wind power generation. The Class-Shape Transformation (CST) method is a popular airfoil parameterization technique that decomposes the airfoil geometry into a series of control parameters, allowing designers to generate new airfoil shapes by adjusting these parameters. The advantages of this method lie in its high flexibility, computational efficiency, and excellent control over design variables.

[0060] The main purpose of airfoil parameterization is to simplify the airfoil shape through mathematical expressions so that it can be controlled and adjusted with a small number of parameters. This not only facilitates the analysis and modification of the airfoil during computer-aided design and optimization, but also improves design efficiency and quality.

[0061] Building a parametric model typically involves two steps: defining a base airfoil and applying a shape transformation function. The base airfoil is typically selected from an existing airfoil library, using a suitable airfoil as a starting point. The shape transformation function then adjusts and modifies the airfoil's shape by introducing parameters.

[0062] Control parameters play a key role in parametric methods. They determine the specific geometric characteristics of the airfoil, such as thickness distribution, camber distribution, leading edge radius, and trailing edge shape. Different combinations of parameter values can generate a variety of airfoil shapes, providing a basis for subsequent performance analysis and optimization.

[0063] Specifically, the CST parameterization method defines the airfoil profile by the product of the class function and the shape function, which is expressed as follows:

[0064] y(ψ)=C(ψ)·S(ψ)+Δy te ·ψ

[0065] Where ψ = x / c, ψ represents the normalized chord-wise position, x is the chord-wise position, c is the chord length, y(ψ) is used to define the airfoil profile, C(ψ) is the class function, S(ψ) is the shape function, and Δy te It is the trailing edge thickness parameter, which is used to fine-tune the thickness of the trailing edge when it is closed. It is usually the actual thickness value of the original trailing edge of the airfoil.

[0066] The class function C(ψ) is expressed as:

[0067]

[0068] Among them, N1 and N2 are the leading edge index and trailing edge index respectively, which are usually fixed at N1=0.5 and N2=1.0. The power exponents N1 and N2 are used to control the leading edge and trailing edge shapes to define the reference shape of the airfoil. N1 controls the sharpness of the leading edge. The larger N1 is, the smoother the leading edge is. N2 controls the closing characteristics of the trailing edge. N2=1.0 can ensure smooth convergence of the trailing edge.

[0069] The shape function S(ψ) is expressed as:

[0070]

[0071] Among them, the shape function S(ψ) is a Bernstein polynomial (the coefficients are design variables), B iis a weight parameter that determines the camber, thickness distribution and local curvature of the airfoil and is used to determine the amplitude of the shape correction. The polynomial order n is usually 5-8, corresponding to 6-9 coefficients, and B n,i is the Bernstein basis function.

[0072] The descriptive parameter group in this embodiment is a set of mathematical parameters used to uniquely determine the geometric shape of the airfoil, which may include six upper surface parameters, six lower surface parameters, N1, N2, and Δy te , Δy te Can be set to 0 or a minimum value, that is, the parameter group = {N1, N2, B0, B1, ..., B n ,Δy te}.

[0073] By adjusting B i The camber and thickness distribution of the airfoil can be changed. For example, increasing B2 can increase the camber of the middle section of the upper surface, improving lift, while decreasing B4 can reduce the thickness of the rear section, reducing pressure drag.

[0074] The CST parameter set simplifies complex airfoil geometry into a small number of adjustable parameters through mathematical modeling. This process requires high-precision fitting of the original airfoil using an optimization algorithm. These parameters are not only the core of the geometric description but also serve as design variables for subsequent aerodynamic optimization, directly impacting key blade performance, such as lift-to-drag ratio and stall characteristics.

[0075] Step S3: performing aerodynamic performance simulation on the airfoil by describing the parameter group to determine the angle of attack of each airfoil.

[0076] In this step, the CST parameterized airfoil of each section needs to be imported into the isight optimization platform and integrated with the XFOIL aerodynamic analysis software for aerodynamic performance simulation and analysis. Aerodynamic performance analysis is performed on each section separately, especially calculating key performance indicators such as lift-to-drag ratio and pressure distribution, so as to determine the optimal angle of attack, i.e., the installation angle, required for each airfoil on the corresponding characteristic section for different operating conditions.

[0077] Specifically including steps S31-S33.

[0078] Step S31: Constructing a simulation environment.

[0079] First, you need to configure the software environment, create an optimization project in the isight platform, and define the design variables (CST parameters), objectives (lift-to-drag ratio), and constraints (pressure gradient limits). Create a conversion module in isight that converts CST parameters to airfoil coordinates and outputs a standard format airfoil file. At the same time, you need to set up parallel tasks in the isight platform so that it can calculate airfoils of multiple sections simultaneously.

[0080] Step S32: In the simulation environment, perform multi-angle CFD scanning on each airfoil, and calculate and store lift-to-drag ratio and pressure distribution data in parallel.

[0081] For each cross-section of the CST airfoil, an XFOIL task is automatically generated and submitted, the angle of attack range is scanned, the task status is monitored in real time, and in case of failure, the angle of attack parameters are automatically retried or adjusted, such as reducing the angle of attack step size. Key data are extracted from the XFOIL output file and stored in the isight database, associated with the corresponding CST parameter group, where the key data include lift-to-drag ratio and pressure distribution data. For each angle of attack of a single airfoil, calculations are performed, and the lift-to-drag ratio and pressure distribution data can be recorded by drawing curves. The lift-to-drag ratio L / D=C is calculated for each angle of attack. l / C d , identify the maximum value (L / D) max .

[0082] Step S33: determining the angle of attack of each airfoil based on the lift-to-drag ratio and the pressure distribution data.

[0083] This step can be implemented through the built-in optimization algorithm of the isight platform. Specifically, the built-in algorithm (such as Hooke-Jeeves) can be called to fine-tune the angle of attack, and a local search can be performed near (L / D)max to obtain the optimal angle of attack. In this way, the corresponding setting angles of attack of the four airfoils can be obtained. In a specific embodiment, for the airfoil setting in step S1, the installation angle of the first airfoil is set to 0 degrees, the angle of attack of the second airfoil is set to 46 degrees, the installation angle of the third airfoil is set to 26 degrees, and the installation angle of the fourth airfoil is set to 18 degrees.

[0084] Step S4: Combine the airfoils at corresponding angles of attack and generate smooth transition cambered surfaces between the airfoils to obtain an initial blade model.

[0085] Based on the blade's cross-sectional design, a multi-airfoil combination was employed. The first and second sections employed the same airfoil design to ensure aerodynamic continuity and structural stability. The third and fourth sections, however, employed different airfoil designs, both from the BE series. This multi-airfoil combination maximized the aerodynamic efficiency of each section, resulting in the initial blade model.

[0086] Specifically including steps S41-S43.

[0087] Step S41: Arrange the airfoils according to the radial direction of the blade and perform twist positioning according to the angle of attack.

[0088] First, based on the aerodynamic load distribution and structural strength requirements of the blade, the specific position of each characteristic section is determined along the radial direction. Through aerodynamic center coordinate calculation and three-dimensional streamline analysis, each airfoil is positioned along the blade height, and spatial twisting is performed based on the optimized angle of attack obtained in step S3. Specifically, a local coordinate system is established for each airfoil so that the aerodynamic center line of each section forms a continuous torsion axis. In this process, the theoretical value of the angle of attack can be corrected in combination with the structural deformation prediction model under the centrifugal force field to avoid dynamic stall during blade rotation. For example, the angle of attack in the tip area needs to be reduced by an additional 0.5° to compensate for the bending-torsion coupling effect caused by the centrifugal force, while the root area uses finite element analysis to verify the matching of the chord length and thickness to ensure the continuity of the radial load gradient. Finally, the three-dimensional spatial posture of each airfoil is fitted by a non-uniform rational B-spline (NURBS) curve to generate a preliminary blade space skeleton model.

[0089] Step S42: using B-spline curves between adjacent airfoils to generate a smoothly transitioned blade surface to connect the airfoils and obtain an intermediate blade model.

[0090] The core of step S42 is to achieve seamless connection of multiple airfoil sections through parametric surface modeling technology. First, the CST parameters of adjacent airfoils are interpolated to generate four intermediate control sections along the radial direction. The chord length and thickness distribution of each section are transitioned through a cubic B-spline function. Then, the bicubic B-spline surface algorithm is used to construct surface control grids in the chord direction (XY plane) and span direction (Z-axis direction), respectively. The leading edge point, maximum thickness point and trailing edge point of each section are used as key control points, and the local curvature of the surface is adjusted by the weight coefficient. In particular, high-order continuity constraints are used in the leading edge area, and the curvature matching algorithm is used to eliminate geometric mutations in the trailing edge area. For high curvature areas (such as the transition section from the root to the middle of the blade), an adaptive node insertion strategy is introduced to automatically add control nodes when the curvature change rate exceeds a threshold (such as 0.15 / mm) to ensure the smoothness of the surface. The intermediate blade model is finally generated.

[0091] Step S43: Verify the surface continuity of the intermediate blade model and correct local mutations to obtain the initial blade model.

[0092] In step S43, the intermediate model is strictly verified using a surface quality assessment system based on differential geometry. First, the Gaussian curvature analysis and absolute curvature comb tools are used to detect whether the G1 (tangent continuity) and G2 (curvature continuity) of the suction and pressure surfaces of the blades meet the standards, focusing on the curvature mutation points in the mid-blade-tip transition zone (such as areas with a curvature change rate of >30%). For areas that do not meet the continuity requirements, an energy-based surface optimization algorithm is used to minimize the surface strain energy, and local smoothing is achieved by adjusting the weight vectors of the B-spline control points. At the same time, combined with CFD pre-analysis results (such as surface pressure gradient distribution), geometric defects that may cause flow separation (such as a sudden change in the leading edge radius or excessive contraction of the trailing edge) are identified, and the parameter backpropagation method is used to correct the CST coefficient. For example, if the second derivative of the pressure coefficient (Cp) is discontinuous on the suction surface of a certain section, the order of the Bernstein polynomial in that area needs to be re-interpolated, and the parameter correlation matrix of the adjacent sections needs to be updated synchronously. The corrected model must pass the surface zebra crossing test in the VDA4955 standard, and the maximum deviation must be controlled within 0.2% of the chord length. The final output is an initial blade model that meets the requirements of aerodynamic-structural collaborative design.

[0093] In addition, the surface generation between adjacent airfoils can also be achieved directly using the software's built-in functions, such as the lofting tools in the 3D software.

[0094] Step S5: With the lift-to-drag ratio maximization as the goal, the initial blade model is optimized by adjusting the descriptive parameter groups of each airfoil to obtain a target blade model.

[0095] With the goal of maximizing the lift-to-drag ratio, an optimization algorithm is used to gradually adjust the CST parameters through multiple rounds of iterations so that each airfoil achieves the optimal performance at the corresponding angle of attack. A parallel multi-objective genetic algorithm based on an agent model can be used to globally optimize the initial blade model.

[0096] In a specific embodiment, the parameters of the initial blade model are optimized using an optimization strategy based on a combination of a Kriging proxy model and an NSGA-II multi-objective genetic algorithm.

[0097] First, a three-dimensional design space containing all CST parameters, angle of attack, and radial position was established. 200 groups of sample points were selected through Latin hypercube experimental design. The lift-to-drag ratio, pressure distribution, and flow separation characteristics of each sample point were calculated using a CFD solver. A Kriging surrogate model was then constructed and cross-validation was performed to ensure the model prediction accuracy (R 2 >0.95).

[0098] The NSGA-II algorithm was then used for multi-objective optimization, with a population size of 100 and 50 generations. The objective function was to maximize the weighted lift-to-drag ratio at each operating point, with constraints including blade strength (maximum equivalent stress <300 MPa) and processability (leading edge curvature radius >0.5 mm). Each generation of elite solutions was directly refined through CFD calculations. After obtaining the Pareto frontier solution, the TOPSIS decision-making method was used to select the parameter combination with the best overall performance.

[0099] Through parameter sensitivity analysis, this process identified the leading edge radius coefficient and maximum thickness position as key control parameters, with coefficients of variation reaching 0.32 and 0.28, respectively, which significantly affect the flow separation characteristics.

[0100] Step S6: Verify the aerodynamic performance of the target blade model, and evaluate the lift-to-drag ratio, flow separation, and drag characteristics of the optimized blade under different operating conditions based on the simulation results.

[0101] In a specific embodiment, a bidirectional fluid-structure coupling method is used to verify aerodynamic performance. First, a three-dimensional transient CFD model containing a rotating coordinate system is constructed. The SST k-ω turbulence model and sliding grid technology are used, and five characteristic operating conditions (40%-120% of the design flow) are set. For each operating condition, a transient calculation of 10 rotation cycles is performed, and the time step corresponds to 1 degree of impeller rotation angle. Simultaneously, a structural dynamics model considering the anisotropy of the composite material is established, and the first 6 natural frequencies are obtained through modal analysis (avoiding the 4-6 times rotation frequency resonance region). In the CFD calculation, the surface pressure distribution is extracted in real time and mapped to the structural grid for stress / strain analysis. The deformation is fed back to the fluid domain for mesh reconstruction. Verification indicators include aerodynamic efficiency (isentropic efficiency deviation <2%), stall margin (≥15%), and dynamic stress amplitude (<200MPa). Especially for transient operating conditions (70-90% flow), vortex dynamics methods are used to analyze secondary flow losses, and the tip leakage vortex intensity is quantified through Q-criterion vortex identification (a decrease of 18% in the Q value peak is considered qualified). Finally, the explicit relationship between design parameters and performance indicators was established through response surface methodology to verify the robustness of the S5 optimization results.

[0102] The surrogate model training data from step S5 directly serves as the base sample space for the S6 validation condition. Both stages share the same CFD solver settings and turbulence model to ensure data consistency. The flow separation characteristics discovered in the high-load region of S6 are fed back into the constraints of S5, forming a closed-loop optimization system. Both stages use the dimensionless pressure coefficient distribution (Cp) and limiting streamlines as common evaluation criteria to ensure consistency between optimization objectives and validation metrics.

[0103] The CST-parameterized multi-airfoil combined axial flow fan blade design system of the present invention can use a computer or other equipment, and embed the software and tools used in the above method, or embed a script that can implement the above method, so as to realize the CST-parameterized multi-airfoil combined axial flow fan blade design method.

[0104] In a real experimental scenario, Figure 2 As shown in the figure, the effectiveness of the present invention's optimization method is verified by comparing the original circular arc thin plate impeller and the optimized multi-airfoil combination blade. The upper figure shows the original circular arc thin plate impeller, and the lower figure shows the multi-airfoil combination blade designed using the present invention. The original airfoil has poor aerodynamic performance, especially unstable at different angles of attack. However, after using the multi-airfoil combination, the optimized blade has a higher lift-to-drag ratio and better aerodynamic efficiency.

[0105] like Figure 3 The figure shows the geometry of the entire blade and the distribution of its different cross-sections. The multi-airfoil composite blade is divided into four sections, each with a different airfoil: BE6308B, BE6306B, and BE7404B. These airfoils are applied at different radial locations on the blade: the first and second airfoils, located near the blade root where airflow is low, use the BE6308B airfoil to optimize lift; the third airfoil, located in the middle, uses the BE6306B airfoil for moderate airflow; and the fourth airfoil, located near the blade tip, uses the BE7404B airfoil to reduce drag caused by high-speed airflow. This multi-airfoil design achieves optimal aerodynamic performance at different locations, ensuring overall blade aerodynamic efficiency and stability.

[0106] like Figure 4 and Figure 5 As shown in the figure, the angle of attack setting and airfoil distribution of each section are further explained, indicating that by combining different angles of attack and airfoils, the different aerodynamic requirements of each section can be flexibly responded to, ensuring the optimization of the overall aerodynamic performance.

[0107] like Figure 6 The comparison of the original and optimized airfoils in the embodiment of the present invention, shown in Figure 2, intuitively demonstrates the geometric differences between the airfoils before and after optimization using the CST parametric optimization method. The optimized airfoil shape exhibits smoother contours and more rational aerodynamic curves in all cross-sections, ensuring superior aerodynamic performance under various operating conditions. This comparison highlights the significant advantages of the optimization method of the present invention in improving blade performance and energy efficiency.

[0108] like Figure 7 and Figure 8As shown in the figure, numerical simulations of the original thin-plate blade and the multi-airfoil composite blade, while ensuring consistent operating and boundary conditions, show that the multi-airfoil composite blade exhibits more uniform pressure gradients and a more stable velocity field across the blade surface, significantly reducing the generation of secondary flows and vortices, and improving airflow stability and energy efficiency. Compared to the original blade, the multi-airfoil composite blade reduces power consumption by approximately 8.3% and increases fan efficiency by 12.98%, demonstrating its excellent energy-saving effects and optimization potential.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-airfoil combined axial flow fan blade design method based on CST parameterization, characterized in that: The following steps are involved: S1 sets multiple characteristic sections along the radial direction of the blade and determines the airfoil corresponding to each characteristic section; S2 uses the CST parameterization method to geometrically model the airfoils of all characteristic sections and obtain the corresponding descriptive parameter groups for each airfoil; S3 simulates the aerodynamic performance of the airfoil by describing the parameter group to determine the angle of attack of each airfoil; S4 combines the airfoils at corresponding angles of attack and generates smooth transition cambered surfaces between the airfoils to obtain the initial blade model. S5 takes the lift-to-drag ratio maximization as the goal and optimizes the initial blade model by adjusting the descriptive parameter groups of each airfoil to obtain the target blade model.

2. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The number of the characteristic cross sections is four, and the first and last characteristic cross sections are located at the blade root and the blade top respectively.

3. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The step S3 includes the following sub-steps: S31 builds the simulation environment; In the simulation environment, S32 performs CFD scans at multiple attack angles for each airfoil, and calculates and stores lift-to-drag ratio and pressure distribution data in parallel; S33 determines the angle of attack of each airfoil based on the lift-to-drag ratio and the pressure distribution data.

4. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The step S4 includes the following sub-steps: S41 arranges each airfoil in the radial direction of the blade and performs twist positioning according to the angle of attack; S42 uses B-spline curves between adjacent airfoils to generate a smoothly transitioned blade surface to connect the airfoils and obtain the intermediate blade model; S43 verifies the surface continuity of the intermediate blade model and corrects local mutations to obtain the initial blade model.

5. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The step S5 includes, with the goal of maximizing the lift-to-drag ratio, using an optimization algorithm, through multiple rounds of iterations, gradually adjusting the CST parameters so that each airfoil achieves optimal performance at a corresponding angle of attack.

6. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The method further includes step S6 of verifying the aerodynamic performance of the target blade model and evaluating the lift-to-drag ratio, flow separation, and drag characteristics of the optimized blade under different operating conditions based on the simulation results.

7. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 1, characterized in that: The CST parameterization method defines the airfoil profile by the product of the class function and the shape function, which is expressed as: y(ψ)=C(ψ)·S(ψ)+Δy te ·ψ Where ψ = x / c, ψ represents the normalized chord-wise position, x is the chord-wise position, c is the chord length, y(ψ) is used to define the airfoil profile, C(ψ) is the class function, S(ψ) is the shape function, and Δy te is the trailing edge thickness parameter.

8. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 7, characterized in that: The class function C(ψ) is expressed as: Among them, N1 and N2 are the leading edge index and the trailing edge index respectively.

9. The CST parameterized multi-airfoil combined axial flow fan blade design method according to claim 7, characterized in that: The shape function S(ψ) is expressed as: Among them, B i is the weight parameter, B n,i is the Bernstein basis function.

10. A multi-airfoil combined axial flow fan blade design system based on CST parameterization, characterized in that: The invention is embedded with the multi-airfoil combined axial flow fan blade design method described in any one of claims 1 to 9.

Images