A laboratory-scale wind turbine blade design method for physical model testing

Abstract

The present invention provides a laboratory-scale wind turbine blade design method suitable for physical model testing, including a method for constructing a model of aerodynamic loads of a fan and a method for constructing an optimal solution of a fan impeller model. Since traditional fan blade models all adopt a direct design method based on the aerodynamic coefficients of the blade airfoil, this method requires manual iterative cycle design for the design goal, which is inefficient. The method and model proposed in the present invention can use a genetic algorithm to perform optimization design under given design goals and initial conditions to obtain the required fan blade model, which is more efficient and has a better optimization design effect.

Description

Technical Field

[0001] The present invention relates to the technical field of generator design, and in particular to a laboratory-scale wind turbine blade design method suitable for physical model testing. Background Art

[0002] Offshore wind turbines can be divided into fixed and floating types according to the type of supporting foundation. At present, all the offshore wind farms built in China use fixed offshore wind turbines. As the water depth continues to increase, various traditional fixed offshore wind turbines can no longer meet the requirements of deep-sea wind energy development. According to the current level of scientific research and engineering, it is generally believed internationally that when the working water depth exceeds 50 meters, the use of offshore floating wind turbine foundations and mooring systems as a development method has better economic benefits and broader market prospects. China has very rich deep-sea wind energy resources, and floating wind turbines will become the only way to develop offshore wind power in the future.

[0003] Compared with traditional floating oil and gas platforms, floating wind turbines have an additional upper tower and blade structure. While the floating foundation is subject to hydrodynamic loads, the wind turbine impeller is also subject to large aerodynamic loads. Therefore, the safety of the use of floating wind turbines needs to be more rigorous. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention aims to provide a laboratory-scale wind turbine blade design method suitable for physical model testing, which can comprehensively consider the influence of blade loads.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A laboratory-scale wind turbine blade design method suitable for physical model testing, characterized by comprising a method for constructing a model of aerodynamic loads of a wind turbine and a method for constructing an optimal solution of a wind turbine impeller model;

[0007] The method for constructing the model of the fan aerodynamic load includes:

[0008] Determining model parameters of the fan impeller related to aerodynamic loads;

[0009] Introduce axial and tangential induction factors a and b;

[0010] Initialize the induction factors a and b; calculate the angle between the relative wind speed and the blade element and the angle of attack;

[0011] The induction factors a' and b' are obtained through the angle between wind speed and blade element and the angle of attack;

[0012] Determine whether a≈a′, b≈b′ is established. If not, return to the initialization induction factor. If established, calculate the relative wind speed and the load on the unit length circular ring control body, and finally obtain the heavy load on the impeller.

[0013] The method for constructing the optimal solution of the fan impeller model includes:

[0014] Determine the prototype of the size and performance parameters of the fan impeller, and obtain the optimal solution of the fan impeller model that meets the thrust and torque similarity through the genetic algorithm; in the selection process of the genetic algorithm, write the fitness function through the fan aerodynamic load model described in the method for constructing the fan aerodynamic load model.

[0015] Preferably, the method for constructing the model of the fan aerodynamic load includes:

[0016] S1. Determining the model parameters of the fan impeller related to the aerodynamic load, comprising the following steps:

[0017] Assume that the fan impeller radius is R, the number of blades is n, the airfoil is known, and the corresponding lift and drag coefficient curve is C L (α), C D (α), the chord length is C, the pitch angle is θ; the undisturbed incoming wind speed is v1, the speed after flowing through the impeller plane is v2, and the impeller speed is Ω; the circular annular flow tube with a radius of r and a width of dr in the impeller plane is taken as the control body, and it is assumed that: the control bodies are independent of each other; the force exerted by the blades on the fluid is evenly distributed; therefore, the impeller is divided into N circular annular control bodies, and the radii of the center lines of the circular annular control bodies are r1, r2, r3, ..., r N (r i ), the corresponding chord length is C i , pitch angle θ i , leaf solidity

[0018] S2. Introduce axial and tangential induction factors a and b.

[0019]

[0020]

[0021] S3, initialization induction factors a, b;

[0022] S4. Calculate the angle between relative wind speed and blade element

[0023] S5, determine the angle of attack α=φ-θ;

[0024] S6, according to the airfoil lift and drag curve C L (α), C D(α), determine the lift and drag coefficient C L , C D ;

[0025] S7. Calculate the axial force and tangential force coefficients:

[0026] C N =C L cosφ+C D sinφ (3)

[0027] C T =C L sinφ+C D cosφ (4)

[0028] S8. Recalculate the induction factor:

[0029]

[0030]

[0031] S9, determine whether a≈a′, b≈b′ holds true, if not, return to step S1 and iterate again, if yes, proceed to the next step of calculation;

[0032] S10. Calculate relative wind speed:

[0033]

[0034]

[0035] S11. Calculate the load (thrust, torque) on the ring control body per unit length:

[0036]

[0037]

[0038] S12. Integrate formulas (8) and (9) along the blade radial direction to obtain a total load model for the impeller; the total load model for the impeller is a model of the aerodynamic load of the fan.

[0039] Preferably, the method for constructing the optimal solution of the fan impeller model includes:

[0040] Forming an initial test population: fitting a discrete model to the blade parameters of each blade unit of the fan impeller to obtain a plurality of blade units; encoding the airfoil, chord length, and twist angle corresponding to each blade unit in the discrete model, and the airfoil, chord length, and twist angle parameters corresponding to each blade unit in the discrete model form a design variable space; randomly generating a number of initial blade units in the design variable space to form an initial test population;

[0041] Selection process: writing a fitness function according to the aerodynamic load model of the fan, calculating the individual fitness of each initial blade unit in the initial test population, and selecting excellent individuals as parents according to the individual fitness to generate offspring individuals;

[0042] Crossover process: The excellent individuals obtained in the selection process are used as parents, and the crossover operator is used to reorganize the genetic structure of the parent individuals through crossover to obtain new offspring individuals;

[0043] Mutation process: The excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to randomly change the genes of a single parent individual to generate new offspring individuals;

[0044] Replacement process: The new offspring individuals generated through the selection, crossover and mutation processes are used as the contemporary population for evaluation, selection, crossover and trading to generate the next generation of offspring individuals;

[0045] Calculation termination: Set the allowable value of the optimal solution of the fitness function. If the individual fitness is less than the allowable value during the optimization process, the individual fitness is the optimal solution, and the corresponding initial leaf unit is the best individual, and the calculation is terminated. If the preset maximum evolutionary generation is reached, the individual fitness corresponding to all the best individuals is output as the optimal solution, and the calculation is terminated.

[0046] Obtaining the optimal solution of the fan impeller model: The airfoil, chord length, and twist angle corresponding to the blade unit of the fan impeller model adopt the optimal solution, that is, obtaining the optimal solution of the fan impeller model.

[0047] Preferably, the initial blade unit is a blade unit having a different aerodynamic shape.

[0048] The beneficial effects of the present invention are:

[0049] Since traditional fan blade models all use a direct design method based on the aerodynamic coefficient of the blade airfoil, this method requires manual iterative cycle design based on the design target, which is inefficient. The method and model proposed in the present invention can use genetic algorithms to optimize the design under given design targets and initial conditions to obtain the required fan blade model, which is more efficient and has better optimization design effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The present invention is further described in detail below based on the accompanying drawings and embodiments.

[0051] Figure 1 The present invention is Figure 1 . Schematic diagram of the comparison of aerodynamic performance (thrust) between the prototype and the model of the fan (all converted to the prototype scale for comparison);

[0052] Figure 2It is a schematic diagram comparing the aerodynamic performance (torque) of the prototype and the model of the fan described in the present invention (all converted to the prototype scale for comparison). DETAILED DESCRIPTION

[0053] In order to make the technical problems solved by the present invention, the technical solutions adopted and the technical effects achieved clearer, the technical solutions of the embodiments of the present invention are further described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0054] The present invention discloses a laboratory-scale wind turbine blade design method suitable for physical model testing, including a method for constructing a model of aerodynamic loads of a wind turbine and a method for constructing an optimal solution of a wind turbine impeller model.

[0055] A method for constructing a model of aerodynamic loads of a fan comprises the following steps:

[0056] S1. Determine the model parameters of the fan impeller related to the aerodynamic load, including the following steps: Assume that the fan impeller radius is R, the number of blades is n; the airfoil is known, and the corresponding lift and drag coefficient curve is C L (α), C D (α), the chord length is C, the pitch angle is θ; the undisturbed incoming wind speed is v1, the speed after flowing through the impeller plane is v2, and the impeller speed is Ω; the circular annular flow tube with a radius of r and a width of dr in the impeller plane is taken as the control body, and it is assumed that: the control bodies are independent of each other (no radial flow); the force exerted by the blades on the fluid is evenly distributed (infinite number of blades). Therefore, the impeller is divided into N circular annular control bodies, and the radii of the center lines of the circular annular control bodies are r1, r2, r3, ..., r N (r i ), the corresponding chord length is C i , pitch angle θ i , leaf solidity

[0057] S2. Introduce axial and tangential induction factors a and b.

[0058]

[0059]

[0060] S3, initialization induction factors a, b;

[0061] S4. Calculate the angle between relative wind speed and blade element

[0062] S5, determine the angle of attack α=φ-θ;

[0063] S6, according to the airfoil lift and drag curve C L (α), C D (α), determine the lift and drag coefficient C L , C D ;

[0064] S7. Calculate the axial force and tangential force coefficients:

[0065] C N =C L cosφ+C D sinφ (3)

[0066] C T =C L sinφ+C D cosφ (4)

[0067] S8. Recalculate the induction factor:

[0068]

[0069]

[0070] S9, determine whether a≈a′, b≈b′ holds. If not, return to step (1) and iterate again. If so, proceed to the next step of calculation.

[0071] S10. Calculate relative wind speed:

[0072]

[0073] S11. Calculate the load (thrust, torque) on the ring control body per unit length:

[0074]

[0075]

[0076] S12. Integrate formulas (8) and (9) along the blade radial direction to obtain the total load model of the impeller:

[0077]

[0078]

[0079] The total load model of the impeller is a model of the aerodynamic load of the fan.

[0080] The method for constructing an optimal solution of a fan impeller model described in the present invention determines the prototype of the size and performance parameters of the fan impeller, and obtains the optimal solution of the fan impeller model that meets similar thrust and torque through a genetic algorithm; during the selection process of the genetic algorithm, a fitness function is written using the fan aerodynamic load model described in the method for constructing a model of the fan aerodynamic load.

[0081] The above construction method specifically includes the following steps:

[0082] Forming a preliminary test population: fitting a discrete model to the blade parameters of each blade unit of the fan impeller to obtain a plurality of blade units; encoding the airfoil, chord length, and twist angle (i.e., individual genes) corresponding to each blade unit in the discrete model, and the airfoil, chord length, and twist angle parameters corresponding to each blade unit in the discrete model constitute a design variable space; in the design variable space, randomly generating a number of initial blade units (as a specific implementation method, the built-in genetic algorithm function of Matlab can be used for random generation; the initial blade units are blade units with different aerodynamic shapes.), forming a preliminary test population;

[0083] Selection process: Write a fitness function based on the objective function (use the written fan aerodynamic load model to calculate the thrust and torque coefficients of each individual at different wind speeds, and compare them with the prototype (referring to the aerodynamic thrust and torque of the full-size fan) to calculate the similarity between the two):

[0084] Fitness = p1*|T-T0|+p2*|M-M0|;

[0085] p1 and p2 are weighted values, T and M are the offspring thrust and torque, and T0 and M0 are the target thrust and torque, respectively.

[0086] Calculate the individual fitness of each initial leaf unit in the initial test population, and select excellent individuals as parents according to the individual fitness to generate offspring individuals. Figure 1 to Figure 2 shown.

[0087] Crossover process: The excellent individuals obtained in the selection process are used as parents, and the crossover operator is used to reorganize the genetic structure of the parent individuals through crossover to obtain new offspring individuals;

[0088] Specifically, the excellent individuals obtained in the selection process are used as parents, and the crossover operator method is used to randomly pair individuals in the parent group. For each pair of paired individuals, the position after a gene position is randomly set as the crossover point, and according to the set crossover probability P1, part of the genes of the two individuals are exchanged at the crossover point to generate two new individuals; the expression of the above crossover probability P1 is Pc=Ne / N, N is the number of individuals in the population, and Ne is the number of exchanged individuals in the population.

[0089] Mutation process: The excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to randomly change the genes of a single parent individual to generate new offspring individuals;

[0090] Specifically: the excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to first specify the gene position of each individual as the mutation point according to the mutation probability P2, and the mutation point is changed (increase or decrease the design variable) to generate new offspring individuals; the expression of the above mutation probability P2 is P2=Ne / N, N is the number of individuals in the population, and Ne is the number of mutant individuals in the population.

[0091] Replacement process: The new offspring individuals generated through the selection, crossover and mutation processes are used as the contemporary population for evaluation, selection, crossover and trading to generate the next generation of offspring individuals;

[0092] Calculation termination: Set the allowable value of the optimal solution of the fitness function (the allowable value is less than or equal to 5%). If the individual fitness is less than the allowable value during the optimization process, the individual fitness is the optimal solution, and the corresponding initial leaf unit is the best individual, and the calculation is terminated. If the preset maximum evolutionary generation (which can be 500) is reached, the individual fitness corresponding to all the best individuals is output as the optimal solution, and the calculation is terminated.

[0093] Obtaining the optimal solution of the fan impeller model: The airfoil, chord length, and twist angle corresponding to the blade unit of the fan impeller model adopt the optimal solution, that is, obtaining the optimal solution of the fan impeller model.

[0094] The fan impeller model described in the present invention has blade parameters that satisfy the optimal solution of the individual fitness of the fan impeller model.

[0095] Example 1

[0096] When the airfoil is determined to be the blade profile of a 5MW wind turbine impeller, the blade profile includes 8 specific models: C1, C2, DU40, DU35, DU30, DU25, DU21, and NACA64.

[0097] The method proposed in the present invention can be used to obtain the corresponding model scale fan blade. The specific process is as follows:

[0098] The initial data of 8 specific models C1, C2, DU40, DU35, DU30, DU25, DU21, NACA64 are as follows:

[0099] C1 blade model aerodynamic performance data:

[0100] Angle of attack Drag coefficient Lift coefficient Pitch moment coefficient -180 0.5 0 0 0 0.5 0 0 180 0.5 0 0

[0101] C2 blade model aerodynamic performance data:

[0102] Angle of attack Drag coefficient Lift coefficient Pitch moment coefficient -180 0.35 0 0 0 0.35 0 0 180 0.35 0 0

[0103] DU21 blade model aerodynamic performance data:

[0104]

[0105]

[0106]

[0107] DU25 blade model aerodynamic performance data:

[0108]

[0109]

[0110]

[0111] DU30 blade model aerodynamic performance data:

[0112]

[0113]

[0114]

[0115] DU35 blade model aerodynamic performance data:

[0116]

[0117]

[0118]

[0119] DU40 blade model aerodynamic performance data:

[0120]

[0121]

[0122]

[0123]

[0124] NACA64 blade model aerodynamic performance data:

[0125]

[0126]

[0127]

[0128] Aerodynamic parameters of prototype fan blades:

[0129]

[0130] A1. Determine the model parameters of the fan impeller related to the aerodynamic load, including the following steps:

[0131] Assume that the fan impeller radius is R, the number of blades is n, the airfoil is known, and the corresponding lift and drag coefficient curve is C L (α), C D (α), the chord length is C, the pitch angle is θ; the undisturbed incoming wind speed is v1, the speed after flowing through the impeller plane is v2, and the impeller speed is Ω; the circular annular flow tube with a radius of r and a width of dr in the impeller plane is taken as the control body, and it is assumed that: the control bodies are independent of each other (no radial flow); the force exerted by the blades on the fluid is evenly distributed (infinite number of blades). Therefore, the impeller is divided into N circular annular control bodies, and the radii of the center lines of the circular annular control bodies are r1, r2, r3, ..., r N (r i ), the corresponding chord length is C i , pitch angle θ i , leaf solidity

[0132] Among them, the selected airfoil is the blade shape of a 5MW wind turbine impeller, which includes 8 specific models: C1, C2, DU40, DU35, DU30, DU25, DU21, and NACA64.

[0133] A2. Introduce axial and tangential induction factors a and b.

[0134]

[0135]

[0136] A3, initialization induction factors a, b;

[0137] A4. Calculate the angle between relative wind speed and blade element

[0138] A5. Determine the angle of attack α = φ - θ;

[0139] A6. According to the lift and drag curve of airfoil C L (α), C D (α), determine the lift and drag coefficient C L , C D ;

[0140] A7. Calculate the axial force and tangential force coefficients:

[0141] C N =C L cosφ+C D sinφ (3)

[0142] C T =C L sinφ+C D cosφ (4)

[0143] A8. Based on the above data, the induction factor can be recalculated:

[0144]

[0145]

[0146] A9, determine whether a≈a′, b≈b′ holds. If not, return to step A1 and iterate again. If so, proceed to the next step of calculation.

[0147] A10. Calculate relative wind speed:

[0148]

[0149] A11. Calculate the load (thrust, torque) on the circular control body per unit length:

[0150]

[0151]

[0152] A12. Integrate formulas (8) and (9) along the blade radial direction to obtain the total load model of the impeller of the selected airfoil: the total load model of the impeller is the model of the aerodynamic load of the fan.

[0153]

[0154]

[0155] Through the above steps, the model of the fan aerodynamic load is obtained, the prototype of the size and performance parameters of the fan impeller is determined, and the optimal solution of the fan impeller model that meets the thrust and torque similarity is obtained through the genetic algorithm; in the selection process of the genetic algorithm, the fitness function is written through the fan aerodynamic load model described in the construction method of the fan aerodynamic load model. Specifically, the following steps are included:

[0156] A13. Constructing a preliminary test population: fitting a discrete model to the blade parameters of each blade unit of the fan impeller to obtain a plurality of blade units; encoding the airfoil, chord length, and twist angle (i.e., individual genes) corresponding to each blade unit in the discrete model, and the airfoil, chord length, and twist angle parameters corresponding to each blade unit in the discrete model constitute a design variable space; randomly generating a plurality of initial blade units in the design variable space (as a specific implementation method, the random generation can be performed using a built-in genetic algorithm function of Matlab; the initial blade units are blade units with different aerodynamic shapes.), to form a preliminary test population;

[0157] A14. Selection process: Write a fitness function based on the objective function (use the written fan aerodynamic load model to calculate the thrust and torque coefficients of each individual at different wind speeds, and compare them with the prototype (referring to the aerodynamic thrust and torque of the actual size fan) to calculate the similarity between the two):

[0158] Fitness = p1*|T-T0|+p2*|M-M0|;

[0159] p1 and p2 are weighted values, T and M are the offspring thrust and torque, and T0 and M0 are the target thrust and torque, respectively.

[0160] The individual fitness of each initial leaf unit in the preliminary test population is calculated, and excellent individuals are selected as parents according to the individual fitness to generate offspring individuals.

[0161] A15. Crossover process: The excellent individuals obtained in the selection process are used as parents, and the crossover operator is used to cross over and reorganize the genetic structure of the parent individuals to obtain new offspring individuals.

[0162] Specifically, the excellent individuals obtained in the selection process are used as parents, and the crossover operator method is used to randomly pair individuals in the parent group. For each pair of paired individuals, the position after a gene position is randomly set as the crossover point, and according to the set crossover probability P1, part of the genes of the two individuals are exchanged at the crossover point to generate two new individuals; the expression of the above crossover probability P1 is Pc=Ne / N, N is the number of individuals in the population, and Ne is the number of exchanged individuals in the population.

[0163] A16. Mutation process: The excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to randomly change the genes of a single parent individual to generate new offspring individuals;

[0164] Specifically: the excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to first specify the gene position of each individual as the mutation point according to the mutation probability P2, and the mutation point is changed (increase or decrease the design variable) to generate new offspring individuals; the expression of the above mutation probability P2 is P2=Ne / N, N is the number of individuals in the population, and Ne is the number of mutant individuals in the population.

[0165] A17. Replacement process: The new offspring individuals generated through the selection, crossover and mutation processes are used as the contemporary population for evaluation, selection, crossover and trading to generate the next generation of offspring individuals;

[0166] A18. Calculation termination: Set the allowable value of the optimal solution of the fitness function (the allowable value is less than or equal to 5%). If the individual fitness is less than the allowable value during the optimization process, the individual fitness is the optimal solution, and the corresponding initial leaf unit is the best individual, and the operation is terminated. If the preset maximum evolutionary generation is reached (which can be 500), the individual fitness corresponding to all the best individuals is output as the optimal solution, and the calculation is terminated.

[0167] A19. Obtain the optimal solution of the fan impeller model: The airfoil, chord length, and twist angle corresponding to the blade unit of the fan impeller model adopt the optimal solution, that is, the optimal solution of the fan impeller model is obtained.

[0168] The method and model of the present invention are applied to floating wind turbines and their impellers and blades. The dynamic response analysis of floating wind turbines needs to consider the influence of the restoring force of the mooring system, the hydrodynamic load and the aerodynamic load at the same time. At present, there are some approximate theories and empirical correction models in the numerical simulation of the motion response of floating wind turbines, and the numerical simulation of strong nonlinear processes under extreme working conditions has high uncertainty. Therefore, physical model tests are needed to correct the empirical coefficients in the numerical analysis and verify the safety under extreme working conditions.

[0169] Unlike fixed wind turbines, floating wind turbines have a larger motion response, especially the coupling effect between surge and pitch motion and aerodynamic loads is significant. Therefore, how to accurately simulate aerodynamic loads and hydrodynamic loads at the same time is a key issue to ensure the authenticity and reliability of floating wind turbine model tests. Generally, the simulation of hydrodynamic loads must follow the Froude similarity criterion, while the simulation of aerodynamic loads must follow the Reynolds similarity criterion. Floating wind turbines are affected by both hydrodynamic loads and aerodynamic loads, so there is a contradiction in the selection of similarity criteria.

[0170] Current floating wind turbine model tests usually follow the Froude similarity criterion, while the simulation of the aerodynamic load on the impeller ignores the secondary load components and only simulates the main load components (aerodynamic thrust and torque), that is, the thrust and torque coefficients of the model and the prototype are the same at different wind speeds. Therefore, in order to meet the thrust and torque similarity of the model under Froude scaling, it is necessary to redesign the aerodynamic shape of the model blades and improve the aerodynamic performance of the blades at low Reynolds numbers.

[0171] The method and model proposed in the present invention can optimize the design using genetic algorithms under given design goals and initial conditions to obtain the required fan blade model, which is more efficient and has better optimization design effects. It can meet the thrust and torque similarity of the model and improve the aerodynamic performance of the blade at low Reynolds numbers. It can correct the empirical coefficients in numerical analysis and verify safety under extreme working conditions.

[0172] In the description herein, the reference terms "an embodiment", "example", etc., mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example.

[0173] In addition, it should be understood that although this specification is described according to implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0174] The technical principle of the present invention is described above in conjunction with specific embodiments. These descriptions are only for explaining the principle of the present invention and cannot be interpreted as limiting the scope of protection of the present invention in any way. Based on the explanations herein, those skilled in the art can associate other specific implementations of the present invention without paying creative labor, and these methods will fall within the scope of protection of the present invention.

Claims

1. A laboratory scale wind turbine blade design method suitable for physical model testing, characterized in that: Including a method for constructing a model of aerodynamic loads of a fan and a method for constructing an optimal solution of a fan impeller model; The method for constructing the model of the fan aerodynamic load includes: Determining model parameters of the fan impeller related to aerodynamic loads; Introduce axial and tangential induction factors a and b; Initialize the induction factors a and b; calculate the angle between the relative wind speed and the blade element and the angle of attack; The induction factors a' and b' are obtained through the angle between wind speed and blade element and the angle of attack; Determine whether a≈a′, b≈b′ is true. If not, return to the initialization induction factor. If true, calculate the relative wind speed and the load on the unit length circular ring control body, and finally get the heavy load on the impeller. The method for constructing the optimal solution of the fan impeller model includes: Determine the prototype of the size and performance parameters of the fan impeller, and obtain the optimal solution of the fan impeller model that meets the thrust and torque similarity through the genetic algorithm; in the selection process of the genetic algorithm, write the fitness function through the fan aerodynamic load model described in the method for constructing the fan aerodynamic load model; The method for constructing the optimal solution of the fan impeller model includes: Forming an initial test population: fitting a discrete model to the blade parameters of each blade unit of the fan impeller to obtain a plurality of blade units; encoding the airfoil, chord length, and twist angle corresponding to each blade unit in the discrete model, and the airfoil, chord length, and twist angle parameters corresponding to each blade unit in the discrete model form a design variable space; randomly generating a number of initial blade units in the design variable space to form an initial test population; Selection process: writing a fitness function according to the aerodynamic load model of the fan, calculating the individual fitness of each initial blade unit in the initial test population, and selecting excellent individuals as parents according to the individual fitness to generate offspring individuals; Crossover process: The excellent individuals obtained in the selection process are used as parents, and the crossover operator is used to reorganize the genetic structure of the parent individuals through crossover to obtain new offspring individuals; Mutation process: The excellent individuals obtained in the selection process are used as parents, and the mutation operator is used to randomly change the genes of a single parent individual to generate new offspring individuals; Replacement process: The new offspring individuals generated through the selection, crossover and mutation processes are used as the contemporary population for evaluation, selection, crossover and trading to generate the next generation of offspring individuals; Calculation termination: Set the allowable value of the optimal solution of the fitness function. If the individual fitness is less than the allowable value during the optimization process, the individual fitness is the optimal solution, and the corresponding initial leaf unit is the best individual, and the calculation is terminated. If the preset maximum evolutionary generation is reached, the individual fitness corresponding to all the best individuals is output as the optimal solution, and the calculation is terminated. Obtaining the optimal solution of the fan impeller model: the airfoil, chord length, and twist angle corresponding to the blade unit of the fan impeller model adopt the optimal solution, that is, obtaining the optimal solution of the fan impeller model; The initial blade units are blade units with different aerodynamic shapes.

Images