Blade shape multistage optimization design method based on intelligent algorithm and blade

Abstract

The invention belongs to the field of wind power blade design, and relates to a blade appearance multistage optimization design method based on an intelligent algorithm and a blade. Firstly, design variables and control variables, including chord length design variables and torsion angle design variables, of a blade are determined; a chord length design section, a chord length control section, a chord length of the control section, a torsion angle design section, a torsion angle control section and a torsion angle of the control section; secondly, determining an initial value, an upper limit value and a lower limit value of a design variable through intelligent optimization of a single target; a series of optimal solution sets meeting the requirements of a plurality of objective functions and constraint functions are further obtained through multi-objective intelligent optimization, the objective functions comprise annual energy output, blade root thrust loads and blade power coefficients, and the constraint functions comprise change rates of blade solidness and chord lengths; according to the method, the traditional blade shape optimization design process is simplified, the shape optimization design comprehensively considering the chord length and the torsion angle of the blade can be realized, and the method has a good application prospect.

Description

Technical Field

[0001] The present invention belongs to the field of wind turbine blade design, and in particular relates to a blade shape multi-level optimization design method based on an intelligent algorithm and a blade. Background Art

[0002] With the large-scale wind turbine blades, the traditional modification-verification design can no longer meet the design requirements of blades. Advanced design methods such as optimization based on intelligent algorithms can significantly improve the design effect and efficiency, and obtain designs that meet multiple design index requirements in the scheme stage for designers to make decisions. In the aerodynamic shape design of blades, chord length and torsion angle are two relatively key design parameters, which greatly affect the aerodynamic performance of blades. The chord length and torsion angle distribution of the full span of the blade can be fitted by the chord length and torsion angle of several sections. Fitting methods such as Bessel spline fitting are more commonly used interpolation methods, but this type of spline fitting method is greatly affected by the selection of sample points. How to determine the initial value and value range of the sample points in the spline fitting will affect the subsequent optimization efficiency and effect. In addition, from the structural and overall perspectives, the position of the maximum chord length, the maximum chord length value, and the torsion angle of some sections are also constrained. The influence of such control points also needs to be considered during the design. The optimization of multiple variables brings a large amount of calculation to the optimization based on intelligent algorithms, which greatly affects the optimization efficiency of such algorithms. Therefore, how to reasonably determine the design space of design variables to ensure both refined optimization needs and improve optimization efficiency is an urgent problem to be solved in this type of optimization. Summary of the invention

[0003] The purpose of the present invention is: in order to solve the above problems, the present invention is a multi-level optimization design method for blade shape based on intelligent algorithm, which can achieve the efficient acquisition of blade chord length and torsion angle design that meet the requirements of power generation, load and aerodynamic efficiency under certain overall parameters and structural parameters. The first level is to optimize the initial values ​​of the chord length and torsion angle design variables through multiple single-objective optimizations, so as to determine the initial values ​​and upper and lower limits of the design variables in the next level of optimization. The second level adopts a multi-objective optimization algorithm to further determine the optimal solution set of design variables that meet the requirements of multiple performance indicators of the blade under the design starting point and design space obtained by the first level of optimization.

[0004] To solve this technical problem, the technical solution of the present invention is:

[0005] On the one hand, a blade shape multi-level optimization design method based on an intelligent algorithm is provided, comprising the following steps:

[0006] S1: Determine the overall parameters of the blade;

[0007] S2: Determine the design variables and control variables of the blade:

[0008] The design variables of the blade are: the chord length of the chord length design section and the torsion angle of the torsion angle design section;

[0009] The control variables are: chord length design section, chord length control section, chord length of the control section, torsion angle design section, torsion angle control section, torsion angle of the control section.

[0010] S3: Determine the initial value and upper and lower limits of the design variables through multiple single-objective optimizations;

[0011] S4: Determine feasible solutions for design variables based on multi-objective genetic algorithm optimization.

[0012] In step S1, the overall parameters of the blade include blade length, number of blade sections, maximum chord length, maximum chord length position, blade hub radius, and cylindrical section length.

[0013] Step S3 includes the following:

[0014] S3_1. Determine the chord length distribution and torsion angle distribution of the reference blade, including the chord length and torsion angle of each cross section over the full span of the blade; the reference blade refers to an existing blade design;

[0015] S3_2, determining the upper and lower limits, number of iterations and population size of the design variables, refers to respectively giving the upper and lower limits, number of iterations and population size for different design variables;

[0016] S3_3. Single-objective optimization based on intelligent algorithms, including algorithm initialization, algorithm iteration and fitness evaluation.

[0017] Step S4 includes S4_1 to S4_7:

[0018] S4_1. Determine the objective function and constraint function, where the objective function includes annual power generation, blade root thrust load, and blade pressure coefficient, and the constraint function includes blade solidity and the rate of change of chord length;

[0019] S4_2, initializing design variables based on a multi-objective genetic algorithm, mainly including generating individuals in a population according to the type of design variables, the number of categories, and the number of variables in each category, each individual representing a design solution under a set of design variable values;

[0020] S4_3, according to the design variable values ​​and the control variables, the geometric parameters of the blade are generated based on spline fitting (specifically Bessel spline fitting), mainly including the chord length and torsion angle distribution of the corresponding blade section position in S1, and the solidity of the blade and the change rate of the chord length are calculated according to the chord length distribution;

[0021] S4_4, modify the blade performance calculation model, mainly modify the aerodynamic calculation model in the calculation model;

[0022] S4_5. According to the calculated result file, extract the corresponding data and calculate the objective function;

[0023] S4_6, updating population information, including updating the function values ​​of the objective function and constraint function of individuals in the population;

[0024] S4_7, population evaluation & genetic operations, if it is the first generation, save the first generation of preferred individuals, repeat s4_2 to s4_7, if it is not the first generation, perform selection, crossover and mutation to generate a new population.

[0025] Preferably, in S2, the chord length design section and the chord length include the coordinate value and chord length value of the section between the maximum chord length section and the blade tip, and the number of design variables according to the chord length can be uniformly selected between them;

[0026] The chord length control section and the chord length of the control section should at least include the end section of the cylindrical section of the blade, the maximum chord length section, the blade tip and the chord length values ​​on these three sections;

[0027] The torsion angle design section and torsion angle refer to the coordinate value and torsion angle value of the section from the cylindrical segment end section to the blade tip, and can be uniformly selected between them according to the number of torsion angle design variables;

[0028] The torsion angle control section includes at least two sections between the blade root and the cylindrical terminal section. These two sections serve as control sections, and the torsion angle of the control section is the starting torsion angle determined by the whole.

[0029] Preferably, in S3_1, the chord length distribution and the torsion angle distribution of the reference blade refer to providing two sets of data, the first set of data is (L i ,C i ), where i∈[N1, N2], N2 can be set according to the specific length of the blade, N1 is at least greater than the number of chord length design variables in S2, L i is the cross-sectional position, and L i The sections corresponding to the chord length design variables do not coincide; C i is the cross section L i The corresponding chord length;

[0030] Similarly, the second set of data is (L i ,θ i ), where i∈[N1, N2], N2 can be set according to the specific length of the blade, N1 is at least greater than the number of torsion angle design variables in S2, L i is the cross-sectional position, and L i The sections corresponding to the torsion angle design variables do not coincide; θ i is the cross section L i The corresponding torsion angle.

[0031] Preferably, in S3_3, a single-objective optimization is performed based on a gray wolf pack intelligent algorithm, and the fitness function is generated by the chord length or the torsion angle according to the position information of the blade. i The corresponding chord length or torsion angle The fitness function is calculated by the following formula (1) or (2), where N is L i The total number of

[0032] For the chord length design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (1) to obtain the optimal value of the chord length design variable; for the torsion angle design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (2) to obtain the optimal value of the torsion angle design variable; further preferably, based on the optimal value of each type of design variable, 20% is added as the upper limit value of the design variable, and 20% is reduced as the lower limit value of the design variable. The fitness function is expressed as follows:

[0033]

[0034] or

[0035]

[0036] in, and are the chord length and torsion angle of the corresponding section of the reference blade, which are constants; C i and θ i are the chord length and torsion angle of the corresponding section of the designed blade, which are variables.

[0037] In S4_1, the objective function is the blade performance evaluation index, including the annual power generation of the blade, the blade root thrust load and the blade power coefficient; the constraint function is the geometric constraint of the blade, including the blade solidity and the chord length change rate.

[0038] The annual power generation of the blade is the energy E absorbed by the wind turbine in a time period of one year t, and the calculation formula is:

[0039]

[0040] Among them: U ci is the cut-in wind speed, U co To cut out wind speed. is the average wind speed at the blade installation location, f(u) is the wind speed distribution density; u is the wind speed, du is the increment of u, and P(u) is the power curve;

[0041] The blade root thrust load is the component of the aerodynamic force on the blade element at the blade root along the rotation axis, which is calculated as follows:

[0042]

[0043] Where R is the aerodynamic force on the blade element, is the actual airflow direction, γ is the drag-lift angle.

[0044] Blade power coefficient C p , calculated as follows:

[0045]

[0046] Where P is power, U is ∞ is the wind speed and R is the blade tip radius.

[0047] The constraint function is specifically:

[0048] The chord length change rate is used to evaluate the monotonicity of the change in chord length from the maximum chord length section to the blade tip. If it is monotonically decreasing, the value is set to zero. If it is not monotonically decreasing, the value is set to -1.

[0049] Blade solidity σ is defined as the total blade area divided by the area of ​​the wind rotor disk. It is the main parameter that determines the performance of the wind rotor.

[0050]

[0051] Where S is the total blade area, R = L + Hub_r, L is the blade length, and Hub_r is the hub radius.

[0052] Preferably, in S4_2, a multi-objective genetic algorithm NSGA-II is used for optimization. The design variable initialization refers to generating individuals in the population according to the type, number of types, and number of variables in each type of the design variable. Each individual represents a design solution under a set of design variable values, that is, a set of chord length and torsion angle values ​​{[C1, C2, .., C m ],[θ1,θ2,…,θ n ]} constitute an individual, m, n are the chord lengths C i and the torsion angle θ i The number of .

[0053] In step S4_4, the blade performance calculation model is modified by modifying the aerodynamic calculation model, specifically:

[0054] The chord length and torsion angle in the aerodynamic model of the blade aerodynamic calculation model are replaced by a set of chord length and torsion angle values ​​{[C1, C2, .., C m ],[θ1,θ2,…,θ n ]}.

[0055] The population information file in step S4_6 records the objective function value, constraint function value, design variable value, constraint violation value, non-dominated sorting, and crowding distance of each individual in the population. Updating the population information refers to writing the specific calculated values ​​of the objective function value and constraint value defined in S4_1 into the population file.

[0056] On the other hand, a wind turbine blade designed by the method is provided, which is installed on a wind turbine generator. The wind turbine blade parameters are: blade length L = 100m to 120m, maximum chord length c = 4.6m to 5.2m, and initial torsion angle is 14 degrees to 16 degrees.

[0057] Compared with the prior art, the present invention has the following beneficial technical effects:

[0058] The multi-level optimization design method for blade shape based on intelligent algorithm disclosed in the present invention optimizes the initial values ​​of chord length and torsion angle design variables respectively by establishing an evaluation function with a reference blade, thereby determining the initial value and upper and lower limits of the design variables in the next step of optimization. In the second step, a multi-objective optimization algorithm is used to further determine the design variable values ​​that meet multiple design indicators of the blade under the design starting point and design space obtained by the first step of optimization.

[0059] The first step of optimization reduces the value space of each design variable, so that in multi-objective optimization, the combination space of the design variables is also reduced accordingly, thereby improving the design efficiency. From the optimization results, the method proposed by the present invention avoids the problem of being unable to obtain a satisfactory design due to unreasonable initial values ​​and upper and lower limits of the design variables, and also avoids the problem of being difficult to determine the upper and lower limits of the values ​​due to insufficient design experience. Compared with the prior art, multiple optimizations are required to determine the design rules, and then adjust the upper and lower limits of the design variables, or directly increase the design variable space, resulting in a long optimization calculation time. The present invention improves such problems and has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions implemented in the present invention, the following is a brief explanation of the drawings required to be used in the examples of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 A flow chart of the method proposed by the present invention;

[0062] Figure 2 It is a schematic diagram of the control section;

[0063] Figure 3It is a schematic diagram of the spatial variation of the design variable values ​​in multi-level optimization;

[0064] Figure 4 This is a flowchart based on the Gray Wolf intelligent algorithm in the first-level optimization;

[0065] Figure 5 The optimal solution of the design variables obtained for the first-level optimization;

[0066] Figure 6 This is the flowchart of the optimization based on the multi-objective genetic algorithm NSGA-II in the second-level optimization;

[0067] Figure 7 This is a diagram showing the performance results of blades optimized in two stages in the embodiment. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 work are within the scope of protection of the present invention.

[0069] The features of various aspects of the embodiments of the present invention will be described in detail below. In the following detailed description, many specific details are proposed to provide a comprehensive understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can also be implemented without these specific details.

[0070] Figure 1 The following is a flow chart of the multi-stage optimization design method of the present invention. Figure 1 and Figure 2 , in step S2:

[0071] like Figure 2 The chord length control section includes at least three sections, namely the maximum chord length (L1), the chord length rapid decrease section (L2) and the blade tip (L3). The corresponding section chord lengths are the maximum chord length, the maximum chord length and the minimum chord length at the blade tip.

[0072] Further preferably, the chord length rapid drop section selects an intermediate value between the cross section corresponding to the first design variable of the chord length and L1.

[0073] like Figure 2There are at least three sections for chord length control, namely the end point of torsion angle maintenance (L4), the torsion angle decreasing section (L5) and the blade tip (L3), that is, the initial values ​​of the torsion angle corresponding to the three sections are the starting torsion angle value, the starting torsion angle value and the minimum torsion angle value of the blade tip.

[0074] Further, the twist angle decreasing section is selected as a position corresponding to a blade length increased by 10% from the twist angle maintaining termination point.

[0075] Refer to the steps of the multi-level optimization design method of the present invention.

[0076] refer to Figure 3 , the chord length design variable and the torsion angle design variable are optimized and designed respectively by using intelligent algorithms. Preferably, 6 chord length design variables and torsion angle design variables are selected, and 3 control variables are selected. Taking the chord length design variable as an example, for the 6 chord length design variables, in the first-level optimization, due to the use of intelligent algorithms, there is no need to determine the initial value, only the upper limit (UL1) and lower limit (LL1) of each design variable need to be given. At this time, the upper and lower limits of all chord length design variables are the same. After optimization by using intelligent algorithms, the optimal solution of each design variable under the single-objective optimality (similar geometric features) is obtained. On the basis of the optimal solution, the upper and lower limits of the design variable are determined by increasing or decreasing the corresponding percentage. At this time, the upper and lower limits of each chord length design variable are different. The torsion angle variable performs the above process again, and the optimal solution and upper and lower limits of each torsion angle design variable are also obtained. Then enter the second-level optimization. The second-level optimization is to obtain the value combination of chord length and torsion angle under the optimal blade performance. After the second-level optimization, a series of optimal solution sets will be obtained, each solution is a combination of a group of chord length variables and torsion angle variables.

[0077] Specifically, in the grey wolf pack intelligent algorithm in step S3_3, the number of iterations and the population size are set to 500 and 50 respectively. Figure 4 The flowchart of the gray wolf-based group intelligence algorithm is shown in Figure 2. This method achieves the optimization purpose by simulating the predation behavior of gray wolf groups and based on the mechanism of wolf group cooperation. The fitness function is calculated by the following formula (1) or (2), where N is L i The total number of

[0078] For the chord length design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (1) to obtain the optimal value of the chord length design variable; for the torsion angle design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (2) to obtain the optimal value of the torsion angle design variable; further, based on the optimal values ​​of various design variables, 20% is added as the upper limit value of the design variable, and 20% is reduced as the lower limit value of the design variable.

[0079]

[0080] or

[0081]

[0082] in, and are the chord length and torsion angle of the corresponding section of the reference blade, which are constants; C i and θ i are the chord length and torsion angle of the corresponding section of the designed blade, which are variables.

[0083] Figure 5 The first two-dimensional distribution diagram of the optimal solution of chord length and torsion angle after the first-level optimization based on the gray wolf population intelligent algorithm. The optimized gray wolf population size is 50 and the number of iterations is 500.

[0084] Figure 6 This is the flow chart of the multi-objective genetic algorithm NSGA-II adopted in the second level, with a crossover probability Pc of 0.9, a mutation probability of 0.1, a crossover distribution index of 10, and a mutation distribution index of 20.

[0085] Further preferably, three objective functions are selected, namely, the annual power generation of the blade, the blade root thrust and the blade power coefficient. The performance evaluation is solved using the open source code fast. Its solution efficiency is suitable for the larger population evaluation calculation requirements of NSGA-II, and its accuracy can also meet the requirements of the scheme design stage.

[0086] Specifically, in S4_1, the objective function is the blade performance evaluation index, including the annual power generation of the blade, the blade root thrust load and the blade power coefficient; the constraint function is the geometric constraint of the blade, including the blade solidity and the chord length change rate.

[0087] The annual power generation of the blade is the energy E absorbed by the wind turbine in a time period of one year t, and the calculation formula is:

[0088]

[0089] Among them: U ci is the cut-in wind speed, U co To cut out wind speed. is the average wind speed at the blade installation location, f(u) is the wind speed distribution density; u is the wind speed, du is the increment of u, and P(u) is the power curve;

[0090] The blade root thrust load is the component of the aerodynamic force on the blade element at the blade root along the rotation axis, which is calculated as follows:

[0091]

[0092] Where R is the aerodynamic force on the blade element, is the actual airflow direction, γ is the drag-lift angle.

[0093] Blade power coefficient C p , calculated as follows:

[0094]

[0095] Where P is power, U is ∞ is the wind speed and R is the blade tip radius.

[0096] The constraint function is specifically:

[0097] The chord length change rate is used to evaluate the monotonicity of the change in chord length from the maximum chord length section to the blade tip. If it is monotonically decreasing, the value is set to zero. If it is not monotonically decreasing, the value is set to -1.

[0098] Blade solidity σ is defined as the total blade area divided by the area of ​​the wind rotor disk. It is the main parameter that determines the performance of the wind rotor.

[0099]

[0100] Where S is the total blade area, R = L + Hub_r, L is the blade length, and Hub_r is the hub radius.

[0101] Specifically in S4_2, as Figure 6 As shown in the figure, the multi-objective genetic algorithm NSGA-II is used for optimization. The design variable initialization refers to generating individuals in the population according to the type, number of types, and number of variables in each type of design variable. Each individual represents a design solution under a set of design variable values, that is, a set of chord length and torsion angle values ​​{[C1, C2, .., C m ],[θ1,θ2,…,θ n ]} constitute an individual, m, n are the chord lengths C i and the torsion angle θ i The number of .

[0102] In step S4_3, Bezier spline fitting adopts segmented fitting. Further, the segmented selections of chord length fitting are cylindrical segment, cylindrical-maximum chord length segment, and maximum chord length-blade tip segment; the segmented selections of torsion angle fitting are cylindrical segment, cylindrical segment-initial torsion holding segment, and initial torsion holding segment-blade tip segment. The calculation of the chord length change rate in the constraint function only calculates the chord length change from the maximum chord length to the blade tip. If it is not monotonically decreasing, the constraint condition is not met.

[0103] In step S4_4, the blade performance calculation model is modified by modifying the aerodynamic calculation model, specifically:

[0104] The chord length and torsion angle in the aerodynamic model of the blade aerodynamic calculation model are replaced by a set of chord length and torsion angle values ​​{[C1, C2, .., C m ],[θ1,θ2,…,θ n ]}.

[0105] In step S4_5, the open source wind blade simulation code FAST is used for solving, and the Euler beam and modal superposition method in FAST are used to calculate the aeroelasticity of the blade. The calculation accuracy can meet the requirements of conceptual design, and the calculation speed is fast, which can meet the efficiency requirements of large number of sample calculations during optimization.

[0106] The population information file in step S4_6 records the objective function value, constraint function value, design variable value, constraint violation value, non-dominated sorting, and crowding distance of each individual in the population. Updating the population information refers to writing the specific calculated values ​​of the objective function value and constraint value defined in S4_1 into the population file.

[0107] During the calculation, blade loss and blade hub loss are not considered, blade inclination angle and cone angle are not taken into account, and a higher blade aerodynamic efficiency can be obtained; a uniform wind field is used for calculation.

[0108] Figure 7 The figure is a result of blade performance through two-stage optimization in the embodiment. The performance index of the blade in the embodiment, i.e., the objective function of optimization, is selected as annual power generation, blade power coefficient and blade root thrust load. The optimal directions of each index are maximization, maximization and minimization, respectively, and the weight coefficients are 1, 20, and 1. In order to show the effect achieved by the first-stage optimization, only the fourth step result of the second-stage optimization is selected for comparison.

[0109] From the comparison chart ( Figure 7 ) It can be seen that the optimization can quickly obtain more satisfactory results than the existing design. From the perspective of the blade power coefficient ( Figure 7 (a)), there are multiple solutions in the optimization results, not only their power coefficient is higher than the existing design, but also their corresponding power generation is greater than the existing design.

[0110] A wind turbine generally includes components such as a wind rotor, a generator, a rectifier, a tower, a speed limiting safety mechanism, and an energy storage device. The blade obtained by the present invention is installed on the wind rotor and used in a wind turbine. Combining multiple solutions in the optimization results obtained by the multi-level optimization of the present invention, the wind turbine blade parameters can be obtained as follows: blade length L = 100m to 120m, maximum chord length c = 4.6m to 5.2m, and a starting torsion angle of 14 degrees to 16 degrees.

[0111] Through the above discussion and examples, a multi-level optimization design of blade shape based on intelligent algorithm is realized.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A multi-level optimization design method for blade shape based on intelligent algorithm, characterized in that: include: S1: Determine the overall parameters of the blade; S2: Determine the design variables and control variables of the blade: The design variables of the blade are: the chord length of the chord length design section and the torsion angle of the torsion angle design section; The control variables are: chord length design section, chord length control section, chord length of the control section, torsion angle design section, torsion angle control section, torsion angle of the control section. S3: Determine the initial value and upper and lower limits of the design variables through multiple single-objective optimizations; S4: Determine feasible solutions for design variables based on multi-objective genetic algorithm optimization.

2. The method according to claim 1, characterized in that In S1, the overall parameters of the blade include blade length, number of blade sections, maximum chord length, maximum chord length position, blade hub radius, and cylindrical section length.

3. The method according to claim 1, characterized in that S3 includes the following steps: S3_1. Determine the chord length distribution and torsion angle distribution of a reference blade, including the chord length and torsion angle of each cross section over the full span of the blade; the reference blade refers to an existing blade design; S3_2, determining the upper and lower limits, number of iterations and population size of the design variables, refers to respectively giving the upper and lower limits, number of iterations and population size for different design variables; S3_3. Single-objective optimization based on intelligent algorithms, including algorithm initialization, algorithm iteration and fitness evaluation.

4. The method according to claim 1, characterized in that: S4 includes steps S4_1 to S4_7: S4_1. Determine the objective function and constraint function, where the objective function includes annual power generation, blade root thrust load, and blade pressure coefficient, and the constraint function includes blade solidity and the rate of change of chord length; S4_2, initializing design variables based on a multi-objective genetic algorithm, mainly including generating individuals in a population according to the type of design variables, the number of categories, and the number of variables in each category, each individual representing a design solution under a set of design variable values; S4_3, according to the design variable values ​​and the control variables, the geometric parameters of the blade are generated based on spline fitting (specifically Bessel spline fitting), mainly including the chord length and torsion angle distribution of the corresponding blade section position in S1, and the solidity of the blade and the change rate of the chord length are calculated according to the chord length distribution; S4_4. Modify the blade performance calculation model: modify the aerodynamic calculation model in the calculation model; S4_5. According to the calculated result file, extract the corresponding data and calculate the objective function; S4_6, updating population information, including updating the function values ​​of the objective function and constraint function of individuals in the population; S4_7, population evaluation & genetic operations, if it is the first generation, save the first generation of preferred individuals, repeat s4_2 to s4_7, if it is not the first generation, perform selection, crossover and mutation to generate a new population.

5. The method according to claim 1, characterized in that In S2, the chord length design section and chord length include the coordinate value and chord length value of the section between the maximum chord length section and the blade tip, and the number of design variables according to the chord length can be uniformly selected between them; The chord length control section and the chord length of the control section should at least include the end section of the cylindrical section of the blade, the maximum chord length section, the blade tip and the chord length values ​​on these three sections; The torsion angle design section and torsion angle refer to the coordinate value and torsion angle value of the section from the cylindrical segment end section to the blade tip, and can be uniformly selected between them according to the number of torsion angle design variables; The twist angle control section includes at least two sections between the blade root and the cylindrical terminal section, and the twist angle of the control section is a starting twist angle determined by the whole.

6. The method according to claim 3, characterized in that In S3_1, the chord length distribution and twist angle distribution of the reference blade refer to two sets of data. The first set of data is (L i ,C i ), where i∈[N1, N2], N2 can be set according to the specific length of the blade, N1 is at least greater than the number of chord length design variables in S2, L i is the cross-sectional position, and L i The sections corresponding to the chord length design variables do not coincide; C i is the cross section L i The corresponding chord length; Similarly, the second set of data is (L i ,θ i ), where i∈[N1, N2], N2 can be set according to the specific length of the blade, N1 is at least greater than the number of torsion angle design variables in S2, L i is the cross-sectional position, and L i The sections corresponding to the torsion angle design variables do not coincide; θ i is the cross section L i The corresponding torsion angle.

7. The method according to claim 3, characterized in that In S3_3, a single-objective optimization is performed based on the gray wolf pack intelligent algorithm. The fitness function is generated by the chord length or torsion angle according to the position information of the blade. i The corresponding chord length or torsion angle The fitness function is calculated by the following formula (1) or (2), where N is L i The total number of For the chord length design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (1) to obtain the optimal value of the chord length design variable; for the torsion angle design variable, S3_1~S3_3 are executed, and the fitness function is calculated using formula (2) to obtain the optimal value of the torsion angle design variable; or in, and are the chord length and torsion angle of the corresponding section of the reference blade, which are constants; C i and θ i are the chord length and torsion angle of the corresponding section of the designed blade, which are variables.

8. The method according to claim 7, characterized in that Based on the optimal values ​​of the chord length and torsion angle design variables, an increase of 20% is used as the upper limit of the design variable, and a decrease of 20% is used as the lower limit of the design variable.

9. The method according to claim 4, characterized in that In S4_1, the objective function is the blade performance evaluation index, including the annual power generation of the blade, the blade root thrust load and the blade power coefficient; the constraint function is the geometric constraint of the blade, including the blade solidity and the chord length change rate; The annual power generation of the blade is the energy E absorbed by the wind turbine in a time period of one year t, and the calculation formula is: Among them: U ci is the cut-in wind speed, U co To cut out the wind speed, is the average wind speed at the blade installation location, f(u) is the wind speed distribution density; u is the wind speed, du is the increment of u, and P(u) is the power curve; The blade root thrust load is the component of the aerodynamic force on the blade element at the blade root along the rotation axis, which is calculated as follows: Where R is the aerodynamic force on the blade element, is the actual airflow direction, γ is the drag-lift angle. Blade power coefficient C p , calculated as follows: Where P is power, U is ∞ is the wind speed and R is the blade tip radius. The constraint function is specifically: The chord length change rate is used to evaluate the monotonicity of the change in chord length from the maximum chord length section to the blade tip. If it is monotonically decreasing, the value is set to zero. If it is not monotonically decreasing, the value is set to -1. The blade solidity σ is defined as the total blade area divided by the area of ​​the wind rotor disk, which is the main parameter that determines the performance of the wind rotor; Where S is the total blade area, R = L + Hub_r, L is the blade length, and Hub_r is the hub radius.

10. The method according to claim 4, characterized in that In S4_2, the multi-objective genetic algorithm NSGA-II is used for optimization. The design variable initialization refers to generating individuals in the population according to the type, number of types, and number of variables in each type of design variable. Each individual represents a design solution under a set of design variable values, that is, a set of chord length and torsion angle values ​​{[C1, C2, .., C m ],[θ1,θ2,…,θ n ]} constitute an individual, m, n are the chord lengths C i and the torsion angle θ i The number of .

11. A wind turbine blade, installed on a wind turbine, characterized in that: The chord length and the torsion angle of the wind turbine blade are designed according to the method described in any one of claims 1 to 10.

12. The wind turbine blade according to claim 11, characterized in that: The wind turbine blade parameters are: blade length L=100m-120m, maximum chord length c=4.6m-5.2m, and a starting torsion angle of 14 degrees-16 degrees.

Images