Rotor wing type multi-objective optimization design method based on biological inspiration algorithm and rotor wing type
By applying the multi-objective mantis search algorithm and kriging agent model in rotor wing design, the problem of convergence difficulties in rotor wing multi-objective optimization design is solved, and efficient multi-objective optimization is achieved to meet the multiple helicopters in different flight states Design requirements.
Patent Information
- Application Number
- CN202411258426.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-26
AI Technical Summary
The prior art has problems of difficulty and slow convergence in the multi-objective optimization design of rotor wing type, and it is difficult to meet the multi-faceted design requirements of helicopters under different flight states.
The multi-objective mantis search algorithm based on biological heuristic algorithm is adopted, combined with CST parameterization and kriging proxy model, and the multi-objective mantis search algorithm is optimized and designed, which improves optimization efficiency and results.
The efficient multi-objective optimization design of the rotor wing type is realized, which meets the multi-task requirements under different flight conditions, and improves optimization efficiency and result quality.
Smart Images

Figure CN119026251A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of rotor optimization design, and in particular to a rotor airfoil multi-objective optimization design method based on a bio-inspired algorithm and a rotor airfoil. Background Art
[0002] The rotor is the core aerodynamic component of the helicopter, which directly affects the flight envelope, speed, range, noise, vibration, payload and other flight performance indicators of the helicopter. Excellent rotor design is the basis for improving helicopter performance. As the primary consideration of rotor design, rotor airfoil design presents a typical multi-objective optimization problem. The rotor airfoil optimization design must meet the multi-faceted design requirements of different flight states such as helicopter maneuvers, hovering, and forward flight. As the number of targets increases, the contradictions between different targets become more prominent, that is, the improvement of the performance of one optimization target is often accompanied by the decline of the performance of one or more other optimization targets. Summary of the invention
[0003] In order to solve the problems existing in the prior art, the present invention proposes a multi-objective optimization design method for a rotor airfoil based on a multi-objective mantis search algorithm, and designs an airfoil that meets the requirements of the multi-task state of the rotor based on this method. The advantages of the mantis search algorithm in terms of efficiency and robustness are fully utilized, and it is fully combined with the multi-objective optimization design of the rotor airfoil, thereby improving the optimization efficiency and the optimization results.
[0004] The technical solution of the present invention is:
[0005] A multi-objective optimization design method for a rotor airfoil based on a bio-inspired algorithm comprises the following steps:
[0006] Step 1: Determine the design requirements of the rotor airfoil, select the reference airfoil, determine the objective function and constraint function, and establish an optimization mathematical model;
[0007] Step 2: Use CST parameterization to describe the geometric shape of the airfoil and determine the number of design variables;
[0008] Step 3: Based on the design variables determined in step 2, the multi-objective mantis search algorithm is used to solve the optimization mathematical model in step 1, and the airfoil that best meets the requirements is selected from the optimization solution set as the final optimized airfoil.
[0009] Furthermore, in step 1, the optimization objective function of the rotor airfoil is:
[0010] Min(f1,f2,f3,f4,f5,f6)
[0011] in
[0012] f1=1 / C l_Ma0.4,f2=C d_Ma0.6 ,f3=|C m_Ma0.6 |
[0013] f4=C d_Ma0.8 ,f5=|C m_Ma0.8 |,f6=C d_Ma0.82 C l_Ma0.4 is the lift coefficient of the airfoil at 0.4 Ma, C d_Ma0.6 is the airfoil drag coefficient at 0.6 Ma, C m_Ma0.6 is the airfoil pitching moment coefficient at 0.6 Ma, C d_Ma0.8 is the airfoil drag coefficient at 0.8 Ma, C m_Ma0.8 is the airfoil pitching moment coefficient at 0.8 Ma, C d_Ma0.82 It is the airfoil drag coefficient at 0.82Ma.
[0014] Furthermore, the specific process of step 3 is:
[0015] Step 3.1: Construct a design space based on the benchmark airfoil, use the optimal Latin hypercube experimental design method to generate several sample points in the design space, and calculate the true response values of the sample points;
[0016] Step 3.2: Establish a proxy model of sample points and response values through sample points and their true response values;
[0017] Step 3.3: Use the multi-objective mantis search algorithm to find the best solution, select new sample points according to the point addition criteria, and calculate their response values through the proxy model and add them to the sample set;
[0018] Step 3.4: Repeat step 3.3 until the termination condition is met, terminate the optimization, and obtain the final optimized airfoil.
[0019] Furthermore, the specific process of step 3.3 is as follows:
[0020] Step 3.3.1: In the design space, randomly initialize a parent population P0 containing n individuals, where each individual represents an airfoil, and use the surrogate model to predict the aerodynamic force and moment coefficient values of each airfoil under a specific design state;
[0021] Step 3.3.2: Use the multi-objective mantis search algorithm to generate a new next-generation population Q, which contains n individuals;
[0022] Step 3.3.3: Merge populations P0 and Q to form a population R containing 2n individuals;
[0023] Step 3.3.4: Select n individuals to form a new parent population P by performing non-dominated sorting and crowding distance calculation on the population R. t ;
[0024] Step 3.3.5: Repeat steps 3.3.2 to 3.3.4 until the set termination condition is met.
[0025] Furthermore, in step 3.3.2, the control parameters of the multi-target mantis search algorithm are set as follows: the archive storing the hidden location of the ambush predator is 1.0; the probability of exchange between the exploration and predation phases is 0.5; the gravity acceleration controlling the mantis attack speed is 6.0; the probability factor controlling the mantis attack failure is 0.5; the cyclic coefficient of exchange between the pursuer and the ambusher is 2.0; and the cannibalism ratio is 0.2.
[0026] Furthermore, the point adding criterion adopted in step 3 is a criterion based on the expected improvement matrix.
[0027] Furthermore, in step 3.3.4, the process of non-dominated sorting is:
[0028] First, two key quantities are assigned to each solution p: the number of solutions n that dominate p p and the solution set S dominated by p p ;
[0029] n p Individuals with a value of 0 are classified into set F1, which represents the first-order Pareto frontier;
[0030] For individuals in F1, traverse each individual's S p , and the n of each solution p minus 1;
[0031] Again, n p Individuals with a value of 0 are classified into set F2, which represents the Pareto frontier of the second order;
[0032] This process is repeated until all individuals have been classified.
[0033] Furthermore, the crowding distance is
[0034]
[0035] Furthermore, in step 3.3.4, after the calculation of non-dominated sorting and crowding distance, individuals with smaller sorting numbers are given priority in selecting individuals; for individuals with the same sorting numbers, individuals with larger crowding distance are selected.
[0036] Using the above method, CST parameterization is used to describe the geometric shape of the airfoil, and the parameterized expression coefficient of the airfoil is:
[0037] Airfoil upper surface:
[0038] <![CDATA[b 0,up ]]> <![CDATA[b 1,up ]]> <![CDATA[b 2,up ]]> <![CDATA[b 3,up ]]> <![CDATA[b 4,up ]]> <![CDATA[b 5,up ]]> <![CDATA[b 6,up ]]> <![CDATA[b 7,up ]]> <![CDATA[Δζ te,up ]]> 0.163 0.215 0.026 0.304 -0.038 0.217 0.101 0.100 0.002
[0039] Airfoil lower surface:
[0040] <![CDATA[b 0,down ]]> <![CDATA[b 1,down ]]> <![CDATA[b 2,down ]]> <![CDATA[b 3,down ]]> <![CDATA[b 4,down ]]> <![CDATA[b 5,down ]]> <![CDATA[b 6,down ]]> <![CDATA[b 7,down ]]> <![CDATA[Δζ te,down ]]> -0.122 -0.111 -0.044 -0.169 -0.031 -0.155 -0.084 -0.034 -0.002 .
[0041] Beneficial Effects
[0042] The present invention aims at the problems of difficulty in convergence and slow convergence of traditional multi-objective optimization algorithms when performing multi-objective optimization design of rotor airfoils, and proposes a multi-objective optimization design method for rotor airfoils based on a bio-inspired algorithm. Compared with traditional optimization methods, the present invention improves the optimization efficiency and the optimization results by coupling CST parameterization, kriging proxy model, multi-objective mantis search algorithm, EIM point addition criterion and other widely used and efficient methods. At the same time, the present invention has a certain expansion capability, and the proposed multi-objective optimization method can also be extended to other optimization problems.
[0043] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:
[0045] Figure 1 : The appearance comparison diagram of opt airfoil and reference airfoil;
[0046] Figure 2 :Comparison of lift characteristics of opt airfoil and reference airfoil at Mach number Ma=0.4;
[0047] Figure 3 : Lift-to-drag ratio comparison of opt airfoil and reference airfoil at Mach number Ma=0.6;
[0048] Figure 4 : Comparison of moment characteristics between opt airfoil and reference airfoil at Mach number Ma=0.6;
[0049] Figure 5 : Comparison of drag divergence characteristics between opt airfoil and reference airfoil;
[0050] Figure 6 : Comparison of zero-lift moment characteristics of opt airfoil and reference airfoil. DETAILED DESCRIPTION
[0051] Embodiments of the present invention are described in detail below. The embodiments are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.
[0052] In the process of airfoil optimization design, the performance of the optimization algorithm determines the efficiency of optimization, therefore, it is very important to select a suitable optimization algorithm. Heuristic algorithms are based on the biological evolution rules of nature, the behavior of biological groups, physical rules and human social behavior. Some classic heuristic algorithms have been used in the field of aerodynamic optimization design, but traditional multi-objective optimization algorithms often face the challenge of convergence difficulties when dealing with multi-objective design problems due to the problems of the algorithm itself. Based on this, the present invention introduces the multi-objective mantis search algorithm into the multi-objective optimization design process of the rotor airfoil, and takes the performance requirements of the rotor airfoil in the three flight states of maneuvering, hovering and forward flight as the design target, improves the optimization efficiency, and obtains the optimized airfoil that meets the design target.
[0053] The specific steps of this embodiment are:
[0054] Step 1: Determine the design requirements of the rotor airfoil, select the reference airfoil, determine the objective function and constraint function, and establish an optimization mathematical model.
[0055] In this embodiment, the performance of the rotor airfoil in three flight states of maneuvering, hovering, and forward flight is considered, including:
[0056] At low Mach numbers, the airfoil has a larger maximum lift coefficient to avoid stall caused by airflow separation of the retreating blade;
[0057] At medium Mach numbers, the airfoil has a higher lift-to-drag ratio to improve the helicopter's hovering efficiency;
[0058] At high Mach numbers, the airfoil has a higher drag divergence Mach number to delay the drag divergence and shock wave generation of the forward blades when the helicopter flies forward;
[0059] During the entire flight process, the moment coefficient is maintained at a low level to reduce blade torque and vibration and ensure reasonable control loads.
[0060] Therefore, SC1095 is selected as the reference airfoil, and the aerodynamic performance in three different flight states of helicopter maneuvering, hovering, and forward flight is taken as the target, specifically a lift coefficient of 0.4Ma, a lift-to-drag ratio of 0.6Ma, a drag coefficient of about 0.8Ma, and a moment coefficient of 0.8Ma. The design states and targets corresponding to different flight states are shown in Table 1:
[0061] Table 1 SC1095 optimization design status and objectives
[0062]
[0063] Moreover, in order to ensure that the airfoil has certain structural performance, the maximum thickness of the airfoil should not be less than the reference airfoil.
[0064] Therefore, the established mathematical model is as follows:
[0065] Min(f1,f2,f3,f4,f5,f6)
[0066] stt>t0
[0067] in
[0068] f1=1 / C l _ Ma0.4 ,f2=C d _ Ma0.6 ,f3=|C m _ Ma0.6 |
[0070] f4=C d _ Ma0.8 ,f5=|C m _ Ma0.8 |,f6=C d _ Ma0.82
[0071] t is the maximum thickness of the optimized airfoil, and t0 is the maximum thickness of the reference airfoil.
[0072] Step 2: Use CST parameterization to describe the geometric shape of the airfoil and determine the number of design variables. The general expression of the CST function is:
[0073]
[0074] in, is a class function that generates the basic geometric shape, S(x) is a shape function that modifies the generated basic shape, xΔζ te is the trailing edge thickness term, which is the coordinate of the airfoil trailing edge relative to the y-axis.
[0075] The function expression is:
[0076]
[0077] The selection of N1 and N2 corresponds to different geometric shapes. In this example, the values are 0.5 and 1.0 respectively.
[0078] The type function expression is:
[0079]
[0080] in, is the basis function of the type function, the most commonly used is the Bernstein polynomial, n is the order of the polynomial, b i are the unknown coefficients, i.e., the design variables of the optimization problem.
[0081] The number of design variables is determined by the order of the polynomial. This example uses a 7th-order Bernstein polynomial, with a total of 18 design variables: 0,up ,b 1,up ,b 2,up ,b 3,up ,b 4,up ,b 5,up ,b 6,up ,b 7,up ,Δζ te,up and the unknown coefficient b of the lower surface 0,down ,b 1,down ,b 2,down ,b 3,down ,b 4,down ,b 5,down ,b 6,down ,b 7,down ,Δζ te,down .
[0082] Step 3: Based on the design variables determined in step 2, the optimization mathematical model in step 1 is solved using a bio-inspired algorithm, and the airfoil that best meets the requirements is selected from the optimization solution set as the final optimized airfoil. This includes the following sub-steps:
[0083] Step 3.1: Construct the design space based on the benchmark airfoil, use the optimal Latin hypercube test design method to generate 100 sample points in the design space, and calculate the true response value of the sample point. Each sample point corresponds to an airfoil, and a total of 100 airfoil samples are generated. The CFD method is used to calculate the aerodynamic force and moment coefficient corresponding to each airfoil under the design state, that is, the true response value of the sample point. The optimal Latin hypercube test design method used here is the Latin hypercube test design method that uses the maximum and minimum distance method to improve the sample distribution.
[0084] Step 3.2: Establish a kriging proxy model of sample points and response values through sample points and their true response values.
[0085] Step 3.3: Use MOMSA to search for the best solution, select new sample points according to the point addition criteria, and calculate their response values through the kriging proxy model and add them to the sample set. The specific process is as follows:
[0086] Step 3.3.1: First, in the design space, a parent population P0 containing n individuals is randomly initialized, where each individual represents an airfoil, and the objective function of each individual is evaluated, that is, the aerodynamic force and moment coefficient values of each airfoil under the specific design state are predicted by the surrogate model;
[0087] Step 3.3.2: Use MOMSA to generate a new next generation population Q, which contains n individuals, that is, use the initial airfoil set to generate a new airfoil set;
[0088] Step 3.3.3: Merge populations P0 and Q to form a population R containing 2n individuals, that is, merge the set of new airfoils with the set of initial airfoils;
[0089] Step 3.3.4: Select n individuals to form a new parent population P by performing non-dominated sorting and crowding distance calculation on the population R. t , that is, through specific screening means, n airfoils are selected to become the initial airfoil set in the next optimization process;
[0090] Step 3.3.5: Repeat steps 2 to 4 until the set termination condition is met.
[0091] For MOMSA (Multi-Objective Mantis Search Algorithm), here is the introduction: Generally speaking, the Mantis Search Algorithm mainly consists of four steps:
[0092] ① Initialization: First, create an initial population for the search. Each mantis in the population represents a potential solution to the optimization problem under consideration.
[0093] ② Exploration process: This process replicates the behavior displayed by mantises when searching for prey, aiming to locate promising solution areas. In this stage, the natural behavior of mantises in searching for prey is divided into two different types: ambush predators and pursuers.
[0094] ③ Predation process: Simulating the attack behavior of the mantis, this stage includes improving the solution based on the promising areas identified in the exploration phase.
[0095] ④ Cannibalism: Inspired by the interesting cannibalism phenomenon observed in mantis populations, this step increases the diversity and adaptability of the algorithm and enhances its ability to navigate in complex solution spaces.
[0096] The control parameters of the mantis search algorithm are: the archive storing the hidden locations of ambush predators (A); the probability of exchange between exploration and predation phases (p); the gravity acceleration that controls the attack speed of the mantis (ρ); the probability factor that controls the failure of the mantis attack (a); the cycle coefficient of exchange between pursuers and ambushers (P); the proportion of cannibalism (Pc). The new next generation of population is generated by controlling the parameters.
[0097] In this embodiment, the MOMSA parameters are set as follows: the population size (N) is 100; the number of iterations (t) is 200, the archive storing the hiding positions of ambush predators (A) is 1.0; the probability of exchange between the exploration and predation phases (p) is 0.5; the gravitational acceleration (ρ) that controls the mantis's attack speed is 6.0; the probability factor that controls the mantis's attack failure (a) is 0.5; the cyclic coefficient (P) of exchange between pursuers and ambushers is 2.0; and the proportion of cannibalism (Pc) is 0.2.
[0098] The point adding criterion used in this embodiment is the EIM criterion (criterion based on the expected improvement matrix), and the number of points added is 200 times; several commonly used EIM criteria are as follows:
[0099]
[0100] In step 3.3.4, the process of non-dominated sorting is as follows:
[0101] ①First, two key quantities are assigned to each solution p: the number of solutions n that govern p p and the solution set S dominated by p p ;
[0102] ② n p Individuals with a value of 0 are classified into set F1, which represents the first-order Pareto frontier;
[0103] ③For individuals in F1, traverse the S of each individual p , and the n of each solution p minus 1;
[0104] ④ Again, n p Individuals with a value of 0 are classified into set F2, which represents the Pareto frontier of the second order;
[0105] ⑤ Repeat this process until all individuals are classified.
[0106] The crowding distance is defined as follows:
[0107]
[0108] Among them, i is an individual in the population and m is the target number.
[0109] After the calculation of non-dominated sorting and crowding distance, individuals with smaller sorting numbers are given priority, that is, individuals that are separated first in the sorting process; for individuals with the same sorting number, individuals with larger crowding distance are selected, that is, individuals with less crowding around them.
[0110] Step 3.4: Repeat step 3.3 until the termination condition is met, terminate the optimization, and obtain the final optimized airfoil.
[0111] Through the above process, the optimized airfoil is opt, and the parameterized expression coefficient of the airfoil is:
[0112] Airfoil upper surface:
[0113] <![CDATA[b 0,up ]]> <![CDATA[b 1,up ]]> <![CDATA[b 2,up ]]> <![CDATA[b 3,up ]]> <![CDATA[b 4,up ]]> <![CDATA[b 5,up ]]> <![CDATA[b 6,up ]]> <![CDATA[b 7,up ]]> <![CDATA[Δζ te,up ]]> 0.163 0.215 0.026 0.304 -0.038 0.217 0.101 0.100 0.002
[0114] Airfoil lower surface:
[0115] <![CDATA[b 0,down ]]> <![CDATA[b 1,down ]]> <![CDATA[b 2,down ]]> <![CDATA[b 3,down ]]> <![CDATA[b 4,down ]]> <![CDATA[b 5,down ]]> <![CDATA[b 6,down ]]> <![CDATA[b 7,down ]]> <![CDATA[Δζ te,down ]]> -0.122 -0.111 -0.044 -0.169 -0.031 -0.155 -0.084 -0.034 -0.002
[0116] And the airfoil coordinate points are obtained as:
[0117] Airfoil upper surface:
[0118]
[0119]
[0120] Airfoil lower surface:
[0121]
[0122]
[0123]
[0124] The comparison results between the optimized airfoil and the reference airfoil under specific design conditions are shown in Table 2.
[0125] Table 2 Comparison of aerodynamic characteristics of SC1095 airfoil before and after optimization
[0126]
[0127] The shape comparison between the optimized airfoil opt and the reference airfoil is as follows: Figure 1 As shown, compared with the original airfoil, the leading edge radius of opt is reduced, the curvature is increased, the upper surface is flatter, and the maximum thickness position is moved backward.
[0128] Figure 2 A comparison of the lift characteristics of the opt airfoil and the benchmark airfoil at low Mach numbers is given. It can be seen that the slope of the lift line has not changed significantly, but the maximum lift coefficient of the opt airfoil has increased, and the corresponding angle of attack has increased slightly. Figure 3 and Figure 4 They are the lift-to-drag ratio comparison and moment characteristics comparison under Mach number 0.6. It is not difficult to find that the maximum lift-to-drag ratio of the opt airfoil and the benchmark airfoil are both achieved near the lift coefficient of 0.6. The lift-to-drag ratio of the opt airfoil is significantly improved compared with the benchmark airfoil, and the absolute value of the moment in this state is also reduced. Figure 5 and Figure 6The comparison of drag divergence characteristics and zero-lift moment characteristics is shown respectively. It can be seen that the zero-lift drag of the opt airfoil is reduced, the drag divergence Mach number is increased, and the moment characteristics are significantly improved.
[0129] Judging from the final optimization results, the six design objectives of the opt airfoil have been improved to varying degrees.
[0130] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and intent of the present invention.
Claims
1. A multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm, characterized by: The following steps are involved: Step 1: Determine the design requirements of the rotor airfoil, select the reference airfoil, determine the objective function and constraint function, and establish an optimization mathematical model; Step 2: Use CST parameterization to describe the geometric shape of the airfoil and determine the number of design variables; Step 3: Based on the design variables determined in step 2, the multi-objective mantis search algorithm is used to solve the optimization mathematical model in step 1, and the airfoil that best meets the requirements is selected from the optimization solution set as the final optimized airfoil.
2. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 1, characterized in that: In step 1, the optimization objective function of the rotor airfoil is: Min(f1,f2,f3,f4,f5,f6) in f1=1 / C l_Ma0.4 ,f2=C d_Ma0.6 ,f3=|C m_Ma0.6 | f4=C d_Ma0.8 ,f5=|C m_Ma0.8 |,f6=C d_Ma0.82 C l_Ma0.4 is the lift coefficient of the airfoil at 0.4 Ma, C d_Ma0.6 is the airfoil drag coefficient at 0.6 Ma, C m _ Ma0.6 is the airfoil pitching moment coefficient at 0.6 Ma, C d_Ma0.8 is the airfoil drag coefficient at 0.8 Ma, C m_Ma0.8 is the airfoil pitching moment coefficient at 0.8 Ma, C d_Ma0.82 It is the airfoil drag coefficient at 0.82Ma.
3. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 1, characterized in that: The specific process of step 3 is: Step 3.1: Construct a design space based on the benchmark airfoil, use the optimal Latin hypercube experimental design method to generate several sample points in the design space, and calculate the true response values of the sample points; Step 3.2: Establish a proxy model of sample points and response values through sample points and their true response values; Step 3.3: Use the multi-objective mantis search algorithm to find the best solution, select new sample points according to the point addition criteria, and calculate their response values through the proxy model and add them to the sample set; Step 3.4: Repeat step 3.3 until the termination condition is met, terminate the optimization, and obtain the final optimized airfoil.
4. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 3, characterized in that: The specific process of step 3.3 is: Step 3.3.1: In the design space, randomly initialize a parent population P0 containing n individuals, where each individual represents an airfoil, and use the surrogate model to predict the aerodynamic force and moment coefficient values of each airfoil under a specific design state; Step 3.3.2: Use the multi-objective mantis search algorithm to generate a new next-generation population Q, which contains n individuals; Step 3.3.3: Merge populations P0 and Q to form a population R containing 2n individuals; Step 3.3.4: Select n individuals to form a new parent population P by performing non-dominated sorting and crowding distance calculation on the population R. t ; Step 3.3.5: Repeat steps 3.3.2 to 3.3.4 until the set termination condition is met.
5. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 4, characterized in that: In step 3.3.2, the control parameters of the multi-target mantis search algorithm are set as follows: the archive storing the hiding location of the ambush predator is 1.0; the probability of exchange between the exploration and predation phases is 0.5; the gravity acceleration controlling the mantis attack speed is 6.0; the probability factor controlling the mantis attack failure is 0.5; the cyclic coefficient of exchange between the pursuer and the ambusher is 2.0; and the cannibalism ratio is 0.
2.
6. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 3, characterized in that: The point addition criterion adopted in step 3 is a criterion based on the expected improvement matrix.
7. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 4, characterized in that: In step 3.3.4, the process of non-dominated sorting is: First, two key quantities are assigned to each solution p: the number of solutions n that dominate p p and the solution set S dominated by p p ; n p Individuals with a value of 0 are classified into set F1, which represents the first-order Pareto frontier; For individuals in F1, traverse each individual's S p , and the n of each solution p minus 1; Again, n p Individuals with a value of 0 are classified into set F2, which represents the Pareto frontier of the second order; This process is repeated until all individuals have been classified.
8. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 4, characterized in that: The crowding distance is 9. The multi-objective optimization design method for rotor airfoil based on a bio-inspired algorithm according to claim 4, characterized in that: In step 3.3.4, after the calculation of non-dominated sorting and crowding distance, individuals with smaller sorting numbers are given priority when selecting individuals; for individuals with the same sorting numbers, individuals with larger crowding distance are selected.
10. A rotor airfoil obtained by the method according to any one of claims 1 to 9, characterized in that: CST parameterization is used to describe the geometric shape of the airfoil, and the parameterization expression coefficient of the airfoil is: Airfoil upper surface: Airfoil lower surface: 。
