A method for predicting non-Gaussian wake wind speed loss distribution of yaw wind turbines

Abstract

The present invention discloses a method for predicting the distribution of non-Gaussian wake wind speed loss of a yaw wind turbine. By constructing a yaw wind turbine wake wind speed loss model and taking into account that the non-Gaussian distribution can increase or decrease the skewness, the wind turbine wake speed can be well predicted according to different wind turbine types and incoming wind conditions in a wind farm, thereby improving the authenticity at the source, improving the accuracy and optimization rate of wind resource assessment and real-time yaw control of the wind farm, and thus helping to improve the power generation efficiency of the wind farm, thereby obtaining greater power generation and economic benefits.

Description

Technical Field

[0001] The invention relates to the technical field of wind power generation, and in particular to a method for predicting the distribution of non-Gaussian wake wind speed loss of a yaw wind turbine. Background Art

[0002] The wake of a wind turbine can be mainly divided into three areas: the near wake area, the transition area, and the far wake area. The near wake area mainly refers to the area from the location of the wind turbine to about three times the diameter of the wind rotor downstream; the transition area refers to the area from three times the diameter of the wind rotor downstream to five times the diameter of the wind rotor; and the far wake area refers to the flow area after five times the diameter of the wind rotor downstream of the wind turbine. The flow field distribution in the near wake area mainly depends on the aerodynamic shape of the blades and the atmospheric flow. The reduction in wind speed downstream of the wind turbine is caused by the wind rotor absorbing energy from the incoming atmosphere and the pressure difference of the airflow upstream and downstream of the wind rotor. In the far wake area, turbulence plays a major role in the flow in this area. The flow distribution characteristics of the far wake area are the focus of much attention from scholars, but due to the increase in wind turbine power and the increase in blade size in recent years, the flow distribution characteristics of the near wake area and the transition area are also very important. (The increase in the size of the wind turbine blades means an increase in the downstream distance in units of the wind rotor diameter. For practical considerations, the wind farm must shorten the distance between wind turbines)

[0003] In fact, people have long known about the flow phenomenon of wakes, and have also realized that the wake effect of wind turbines will have a great impact on the power generation of wind farms; therefore, scholars have long started to study wind turbine wakes and wake effects. In general, there are three main methods for studying wake effects: wind tunnel tests and wind field measurements, computational fluid dynamics methods (CFD), and wake models.

[0004] Wind tunnel tests and wind field measurements were the main methods used by people to study the wake effects of wind turbines in the early days. Wind tunnel tests on wind turbine wakes can be traced back to the 1980s. In 1981, the Swedish Aeronautical Research Institute (FFA) used flow visualization methods to display the flow trajectory of the wake through wind tunnel tests, provided flow information on the wind turbine wake, and conducted a series of related studies. Chamorro and Porté-Agel also conducted wind tunnel tests on single wind turbines, studying the effects of surface roughness and atmospheric stability on the numerical distribution of wake velocity, turbulence intensity, shear stress, and its flow field structure. Wind tunnel tests and wind field measurements can provide real and effective data, but there are also many problems. For example, only existing wind turbines can be measured, and predictions cannot be made for wind turbines that have not yet been built; only partial information on the local area of ​​the wind turbine wake flow field can be obtained, and the entire flow field cannot be fully understood; the investment is large and the cycle is long.

[0005] In recent decades, the rapid improvement of computer technology has also promoted the rapid development of computational fluid dynamics, making CFD methods more and more widely used in the study of wind turbine wake effects. The CFD method simulates the wake flow field of the wind turbine by numerically modeling the wind turbine and solving the fluid control equations. First of all, there are two main types of wind turbine simulation: direct numerical simulation method and brake series method. The direct numerical simulation method refers to directly drawing the real wind turbine in the calculation domain, and then generating a calculation grid around the real blades, nacelle and tower, directly capturing the flow around the blades to obtain the distribution information of the wind turbine wake flow field; this method has extremely high requirements on the quality and quantity of the grid, and requires huge computing resources to obtain more accurate numerical results. However, the brake disc model does not represent the actual movement of the blades, and cannot capture the flow details and blade tip vortices well. The computing speed and power of the computer are an important foundation and guarantee for the CFD numerical simulation method. In general, although the CFD method has high calculation accuracy and comprehensive flow field information, it also has high requirements on computing resources and the level of computing operators. Especially when it comes to complex engineering problems, such as wind farm micro-site optimization and wind farm cluster prediction, the lengthy calculation time is unacceptable in actual engineering.

[0006] The wake model mainly refers to a mathematical model that describes the structure of the wind turbine wake. It is used to calculate the velocity distribution and turbulence distribution in the wake area. It has a relatively simple structure, acceptable calculation accuracy, and short calculation time. It is very suitable for use in wind farm research projects equipped with large wind turbine units. While ensuring acceptable engineering accuracy, the use of the wake model greatly improves the efficiency of wake distribution calculations. Crespo and other scholars have studied the efficiency of using the wake model for wind farm layout.

[0007] The wake models introduced above are all for the non-yaw situation of a single wind turbine on flat terrain. However, in a real wind farm, the atmospheric flow is constantly changing, and it is impossible to ensure that the wind turbine is always facing the flow; and when we use active yaw and other means to avoid the negative impact caused by the wind turbine wake effect and conduct micro-site selection of wind farms, accurately predicting the distribution and characteristics of the yaw wind turbine wake is a solid foundation and guarantee; if this cannot be achieved, the effects of active control and site selection will certainly be unsatisfactory. Jiménez and Crespo constructed a yaw wind turbine wake model based on the conservation of momentum, conservation of mass, and the wind turbine force correction and flow tube model considering the deflection angle. The model is divided into two parts: the wake deviation model and the wind speed loss model. The combination of the two can give the predicted yaw wake distribution. The problem with this model is similar to that of the Jensen model. The assumption that the wind speed loss distribution is a constant does not conform to the actual situation, and the wind speed error predicted by the model will be relatively large. At the same time, the prediction results of the wake deviation model constructed based on this also have a large error with the experimental results, greatly overestimating the deflection of the wake. Ishihara and Guo-Wei Qian also constructed a yaw wake wind speed model based on Gaussian distribution. It considers the wind speed loss as a Gaussian distribution, and based on this, a wake deviation model is given. Both in terms of physical theory and the accuracy of the prediction results, better results are achieved.

[0008] At present, the research on the wake model of yawed wind turbines mainly refers to the framework route of the wake model of non-yawed wind turbines, and introduces Gaussian distribution, super Gaussian distribution and double Gaussian distribution to describe the lateral distribution of downstream wake wind speed loss, all of which adopt the symmetrical lateral distribution assumption of downstream wind speed loss. However, due to the existence of the yaw angle, the actual wind turbine has a non-zero force in the yaw direction, and the average lateral velocity of the wake space is non-zero. Its far wake velocity not only has the center line offset to one side, but more importantly, the cross-sectional shape of the wake will also be deflected at the same time, with obvious asymmetry on the deflection plane. This leads to the lack of high accuracy in the prediction results of the current yawed wind turbine wake velocity model and the wrong estimation of the wake loss area, and directly causes a series of yaw control efficiency reductions and unreality, which has a huge impact on the overall power generation of wind farms. At the same time, with the continuous development of the research on the wind turbine wake effect, people have found in more and more experimental measurements, numerical simulations and theoretical analyses that the wind speed loss of the wind turbine wake is not a single top-hat distribution (the wind speed loss of the wake is distributed as a constant value on the cross section) or Gaussian distribution (the wind speed loss of the wake is distributed as a Gaussian distribution on the cross section) or double Gaussian distribution in the entire wake area, but there is a gradual change from the bimodal distribution of the near wake to the top-hat distribution, and then to the single-peak distribution in the far wake area. However, the existing wake models cannot describe and predict this point, which directly leads to the limitations and accuracy errors of the prediction of the entire wake area, especially in the area near the center of the wake where the wind speed loss is the largest. Finally, the accuracy of the wake offset prediction model also has room for improvement. Summary of the invention

[0009] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a method for predicting the distribution of non-Gaussian wake wind speed loss of a yaw wind turbine, so as to solve the problem that the yaw wake wind speed loss model used in the existing wind resource assessment and real-time yaw control technology of wind farm is insufficiently accurate, which leads to distortion of load and power assessment and low control efficiency.

[0010] To solve the above technical problems, the present invention adopts the following technical solutions:

[0011] A method for predicting the non-Gaussian wake wind speed loss distribution of a yaw wind turbine, characterized in that it comprises the following steps:

[0012] S1. Obtain the mechanical performance parameters and yaw angle parameters of the wind turbine, and obtain the wind field environment parameters at the location of the wind turbine through actual measurement;

[0013] S2. According to the mechanical performance parameters and yaw angle parameters of the wind turbine, determine the offset angle θ of the wind turbine wake center in the near wake area. n , wind turbine wake boundary and wake width expansion coefficient;

[0014] S3, according to the deviation angle θ of the wind turbine wake center in the near wake area n and the wake width expansion coefficient, the theoretical boundary position x of the near wake and far wake is obtained. 0 , and then determine the wake offset μ1(x) in the near wake area of ​​the wind turbine and the wake offset μ2(x) in the far wake area;

[0015] S4. Calculate the non-Gaussian skewness α of the yaw wind turbine wake at each downstream position of the wind turbine based on the mechanical performance parameters and yaw angle parameters of the wind turbine. 3 and the kurtosis parameter α 4 , and then obtain the process parameters of the non-Gaussian distribution model;

[0016] S5. According to the process parameters of the non-Gaussian distribution model and the wind field environment parameters where the wind turbine is located, determine the normalized standard deviation parameter σ when the non-Gaussian distribution model is transformed. N , and then get the type function of the non-Gaussian distribution model and the maximum loss function C(x), thereby determining the non-Gaussian distribution prediction model of the wake wind speed loss of the yaw wind turbine;

[0017] S6. Calculate and determine the wake wind speed loss distribution of the yaw wind turbine using the obtained wake wind speed loss non-Gaussian distribution prediction model.

[0018] Preferably, in step S1, the relevant parameters of the wind turbine include hub height and impeller diameter; the wind field environmental parameters are the incoming wind speed and direction and environmental turbulence intensity information at the location of the wind turbine obtained based on wind tower data or relevant on-site measured data.

[0019] Preferably, in step S2: the offset angle of the center of the wind turbine wake in the near wake area is calculated by the following formula:

[0020]

[0021] In the formula, θ n is the deviation angle of the wind turbine wake center in the near wake area, γ is the wind turbine yaw angle, C T is the fan thrust coefficient.

[0022] According to the mechanical performance parameters and yaw angle parameters, the fan wake boundary and wake width expansion coefficient are calculated by the following formula:

[0023]

[0024]

[0025]

[0026]

[0027] Where σ is the standard deviation of wake width, k is the wake width expansion coefficient, and D T is the diameter of the fan impeller, x is the downstream distance of the fan, ε and β are the wake boundary parameters, γ is the yaw angle of the fan, and C T is the fan thrust coefficient, I a is the ambient turbulence intensity.

[0028] Preferably, in step S3: the theoretical boundary position between the near wake and the far wake is calculated by the following formula:

[0029]

[0030]

[0031] In the formula, x 0 is the theoretical boundary position between the near wake and far wake, σ 0 is the standard deviation of the wake width at the boundary between the near wake and the far wake, D T is the diameter of the fan impeller, k is the wake width expansion coefficient, ε is the wake boundary parameter, γ is the fan yaw angle, C T is the fan thrust coefficient, θ n is the offset angle of the wind turbine wake center in the near wake area.

[0032] Preferably, in step S3: the wake offset in the near wake region of the wind turbine is calculated by the following formula:

[0033]

[0034] In the formula, μ 1 (x) is the wake deviation in the near wake area of ​​the wind turbine, θ n is the deviation angle of the wind turbine wake center in the near wake area, x is the downstream distance of the wind turbine, D T is the diameter of the fan impeller.

[0035] The wake offset in the far wake region is calculated using the following formula:

[0036]

[0037] In the formula, μ 2 (x) is the wake deviation in the far wake area of ​​the wind turbine, D T is the diameter of the fan impeller, x is the downstream distance of the fan, x 0 is the theoretical boundary position between the near wake and far wake, C T is the thrust coefficient of the fan, σ is the standard deviation of the wake width, σ 0 is the standard deviation of the wake width at the boundary between the near wake and the far wake, θ nis the offset angle of the wind turbine wake center in the near-wake area, k is the wake width expansion coefficient, and γ is the wind turbine yaw angle.

[0038] Preferably, in step S4: the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine is calculated by the following formula:

[0039]

[0040] Kurtosis parameters at each downstream location of the fan:

[0041]

[0042] In the formula, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter of each downstream position of the wind turbine, γ is the yaw angle of the wind turbine, x is the downstream distance of the wind turbine, and D T is the diameter of the fan impeller.

[0043] Preferably, in step S4: according to the wake non-Gaussian skewness α 3 , kurtosis parameter α 4 , and then the process parameters of the non-Gaussian distribution model can be calculated by the following formula:

[0044] K=[1-0.06(3-α 4 )] 1 / 3 ;

[0045]

[0046]

[0047]

[0048] In the formula, K, b 2 、b 3 、b 4 are process parameters of non-Gaussian distribution model, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter at each downstream position of the fan.

[0049] Preferably, in step S5: the normalized standard deviation parameter when the non-Gaussian distribution model is transformed is calculated by the following formula:

[0050]

[0051]

[0052]

[0053]

[0054] In the formula, σ N is the normalized standard deviation parameter, k N is the wake expansion rate of the non-Gaussian distribution model, ε N Wake expansion rate for non-Gaussian distribution model, p 1 is the process parameter, x is the downstream distance of the fan, D T is the diameter of the fan impeller, C T is the fan thrust coefficient, I a is, γ is the yaw angle of the wind turbine;

[0055] The shape function of the non-Gaussian distribution model is calculated by the following formula:

[0056]

[0057] a=b 3 / (3*b 4 );

[0058]

[0059] c=(ba 2 ) 3 ;

[0060] u=(y / D T -μ(x)) / o N ;

[0061]

[0062]

[0063] In the formula, is the shape function of the non-Gaussian distribution model, b 2 , b 3 , b 4 are process parameters of non-Gaussian distribution model, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter of each downstream position of the wind turbine, y is the crosswind distance length, a, b, c, u are intermediate parameters for simplified calculation and construction expression, σ N is the normalized standard deviation parameter, μ(x) is the wake offset of the entire wake area of ​​the yaw wind turbine wake, and ξ(u) is the intermediate conversion function of the non-Gaussian distribution function.

[0064] Preferably, in step S5: the maximum loss function of the non-Gaussian distribution model is calculated by the following formula:

[0065]

[0066]

[0067] Where C(x) is the maximum loss function of the non-Gaussian distribution model, y is the crosswind distance, and D T is the diameter of the fan impeller, C T is the thrust coefficient of the wind turbine, γ is the yaw angle of the wind turbine, σ is the standard deviation of the wake width, and p 2 is the maximum loss correction parameter for non-Gaussian distribution.

[0068] The non-Gaussian distribution prediction model of the yaw wind turbine wake wind speed loss is calculated by the following formula:

[0069]

[0070] AU=U ∞ -U;

[0071] Where U ∞ is the wind speed of the incoming flow at the hub height of the undisturbed fan, U is the wind speed of the fan wake, ΔU is the wind speed value lost after the incoming flow passes through the fan, and D T is the diameter of the fan impeller, σ is the standard deviation of the wake width, γ is the yaw angle of the fan, u is an intermediate parameter for simplifying the calculation structure expression, and p 2 is the maximum loss correction parameter of non-Gaussian distribution, C T is the fan thrust coefficient.

[0072] Preferably, in step S6: utilizing the obtained non-Gaussian distribution prediction model for wake wind speed loss, taking known wind turbine parameters and wind field parameters as input, the lateral distribution of wind speed loss at the height of the wind turbine hub is calculated through the non-Gaussian distribution prediction model for wake wind speed loss, that is, the wake wind speed loss distribution of the yaw wind turbine is obtained.

[0073] Compared with the prior art, the present invention has the following beneficial technical effects:

[0074] 1. The present invention constructs a non-Gaussian wake speed loss distribution prediction method for yaw wind turbines. The non-Gaussian distribution can well describe and predict the asymmetry of the wake velocity distribution of the yaw wind turbine by increasing or decreasing the skewness. At the same time, the gradual change of the distribution form of the wind turbine wake velocity in the entire wake area can be well described and predicted by changing the kurtosis, thereby improving the accuracy and distribution area of ​​the yaw wind turbine wake velocity prediction and improving the authenticity at the source.

[0075] 2. The present invention provides a yaw wind turbine wake wind speed loss model, which can well predict the wind turbine wake speed according to different wind turbine types and incoming wind conditions in the wind farm, improve the accuracy and optimization rate of wind resource assessment and real-time yaw control of the wind farm, and then help improve the power generation efficiency of the wind farm, thereby obtaining greater power generation and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0077] Figure 1 It is a schematic diagram of the processing flow of the method for predicting the distribution of non-Gaussian wake wind speed loss of a yaw wind turbine according to the present invention.

[0078] Figure 2 It is a schematic diagram of the distribution state of the wind turbine's wake wind speed loss when the yaw angle is 0°.

[0079] Figure 3 It is a schematic diagram of the distribution state of the wind turbine's wake wind speed loss when the yaw angle is not 0°.

[0080] Figure 4 This is a comparison diagram of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 0° in the embodiment;

[0081] Figure 5 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 0° in the embodiment.

[0082] Figure 6 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 0° in the embodiment.

[0083] Figure 7 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 10° in the embodiment.

[0084] Figure 8 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 10° in the embodiment.

[0085] Fig. 9 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 10° in the embodiment.

[0086] Fig.10 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 20° in the embodiment.

[0087] Fig.11 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 20° in the embodiment.

[0088] Fig.12 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 20° in the embodiment.

[0089] Fig.13 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 30° in the embodiment.

[0090] Fig.14 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 30° in the embodiment.

[0091] Fig.15 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 30° in the embodiment.

[0092] Fig.16 This is a comparison chart of the predicted results of the maximum wind speed loss value at each downstream position of the wind turbine when the yaw angle yaw is 10° in the embodiment.

[0093] Fig.17 This is a comparison chart of the predicted results of the maximum wind speed loss value at each downstream position of the wind turbine when the yaw angle yaw is 20° in the embodiment.

[0094] Fig.18 It is a comparison chart of the prediction results of the loss intensity at different wind speeds when the yaw angle yaw is 10° in the embodiment.

[0095] Fig.19 It is a comparison chart of the prediction results of the loss intensity at different wind speeds when the yaw angle yaw is 20° in the embodiment. DETAILED DESCRIPTION

[0096] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions 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. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0097] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0098] Due to the existence of the yaw angle, the wake cross section of the actual wind turbine has obvious asymmetry while being offset. This leads to the lack of high accuracy in the prediction results of the current yaw wind turbine wake velocity model and the wrong estimation of the wake loss area, which has a huge impact on the overall power generation of the wind farm. At the same time, the wind turbine wake wind speed loss has a gradual change phenomenon in the entire wake area, from the bimodal distribution of the near wake to the top hat distribution, and then to the single-peak distribution in the far wake area, and the accuracy of the wake offset prediction model also has room for improvement. In response to these problems and shortcomings, the present invention introduces the skewness parameter and the kurtosis parameter to construct and establish the non-Gaussian distribution prediction model of the wake wind speed loss to predict and determine the wake wind speed loss distribution of the yaw wind turbine, which can well solve and accurately describe the above-mentioned wake phenomenon, so as to ultimately improve the accuracy of the wake velocity prediction, and then improve the yaw control efficiency and the total power generation and economic benefits of the wind farm.

[0099] The method for predicting the non-Gaussian wake wind speed loss distribution of a yaw wind turbine proposed in the present invention comprises the following steps:

[0100] S1. Obtain the mechanical performance parameters and yaw angle parameters of the wind turbine, and obtain the wind field environment parameters at the location of the wind turbine through actual measurement;

[0101] S2. According to the mechanical performance parameters and yaw angle parameters of the wind turbine, determine the offset angle θ of the wind turbine wake center in the near wake area. n , wind turbine wake boundary and wake width expansion coefficient;

[0102] S3, according to the deviation angle θ of the wind turbine wake center in the near wake area n and the wake width expansion coefficient, the theoretical boundary position x of the near wake and far wake is obtained. 0 , and then determine the wake offset μ1(x) in the near wake area of ​​the wind turbine and the wake offset μ2(x) in the far wake area;

[0103] S4. Calculate the non-Gaussian skewness α of the yaw wind turbine wake at each downstream position of the wind turbine based on the mechanical performance parameters and yaw angle parameters of the wind turbine. 3 and the kurtosis parameter α 4 , and then obtain the process parameters of the non-Gaussian distribution model;

[0104] S5. According to the process parameters of the non-Gaussian distribution model and the wind field environment parameters where the wind turbine is located, determine the normalized standard deviation parameter σ when the non-Gaussian distribution model is transformed. N , and then get the type function of the non-Gaussian distribution model and the maximum loss function C(x), thereby determining the non-Gaussian distribution prediction model of the wake wind speed loss of the yaw wind turbine;

[0105] S6. Calculate and determine the wake wind speed loss distribution of the yaw wind turbine using the obtained wake wind speed loss non-Gaussian distribution prediction model.

[0106] Each link in the method of the present invention is described in further detail below.

[0107] In step S1, the mechanical performance parameters of the wind turbine obtained include hub height, impeller diameter, etc., or the yaw angle parameters of the wind turbine include the yaw angle value of the wind turbine, etc.; the wind field environmental parameters are mainly obtained based on wind tower data or relevant on-site measured data to obtain wind field condition information such as the incoming wind speed and direction at the location of the wind turbine and the environmental turbulence intensity.

[0108] In step S2, the deviation angle of the wind turbine wake center in the near wake area is calculated by the following formula:

[0109]

[0110] In the formula, θ n is the deviation angle of the wind turbine wake center in the near wake area, γ is the wind turbine yaw angle, C T is the fan thrust coefficient.

[0111] The fan wake boundary and wake width expansion coefficient are calculated by the following formula:

[0112]

[0113]

[0114]

[0115]

[0116] Where σ is the standard deviation of wake width, k is the wake width expansion coefficient, and D T is the diameter of the fan impeller, x is the downstream distance of the fan, ε and β are the wake boundary parameters, γ is the yaw angle of the fan, and C T is the fan thrust coefficient, I a is the ambient turbulence intensity.

[0117] In step S3, the theoretical boundary position between the near wake and far wake is calculated by the following formula:

[0118]

[0119]

[0120] In the formula, x0 is the theoretical boundary position between the near wake and far wake, σ 0 is the standard deviation of the wake width at the boundary between the near wake and the far wake, D T is the diameter of the fan impeller, k is the wake width expansion coefficient, ε is the wake boundary parameter, γ is the fan yaw angle, C T is the fan thrust coefficient, θ n is the offset angle of the wind turbine wake center in the near wake area.

[0121] The wake deviation in the near wake area of ​​the wind turbine is calculated by the following formula:

[0122]

[0123] In the formula, μ 1 (x) is the wake deviation in the near wake area of ​​the wind turbine, θ n is the deviation angle of the wind turbine wake center in the near wake area, x is the downstream distance of the wind turbine, D T is the diameter of the fan impeller.

[0124] The wake offset in the far wake region is calculated using the following formula:

[0125]

[0126] In the formula, μ 2 (x) is the wake deviation in the far wake area of ​​the wind turbine, D T is the diameter of the fan impeller, x is the downstream distance of the fan, x 0 is the theoretical boundary position between the near wake and far wake, C T is the thrust coefficient of the fan, σ is the standard deviation of the wake width, σ 0 is the standard deviation of the wake width at the boundary between the near wake and the far wake, θ n is the offset angle of the wind turbine wake center in the near-wake area, k is the wake width expansion coefficient, and γ is the wind turbine yaw angle.

[0127] In step S4, the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine is calculated by the following formula:

[0128]

[0129] Kurtosis parameters at each downstream location of the fan:

[0130]

[0131] In the formula, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter of each downstream position of the wind turbine, γ is the yaw angle of the wind turbine, x is the downstream distance of the wind turbine, and DT is the diameter of the fan impeller.

[0132] By determining the wake non-Gaussian skewness α 3 , kurtosis parameter α 4 , and then the process parameters of the non-Gaussian distribution model can be calculated by the following formula:

[0133] K=[1-0.06(3-α 4 )] 1 / 3 ;

[0134]

[0135]

[0136]

[0137] In the formula, K, b 2 , b 3 , b 4 are process parameters of non-Gaussian distribution model, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter at each downstream position of the fan.

[0138] In step S5, the normalized standard deviation parameter when the non-Gaussian distribution model is transformed is calculated by the following formula:

[0139]

[0140]

[0141]

[0142]

[0143] In the formula, σ N is the normalized standard deviation parameter, k N is the wake expansion rate of the non-Gaussian distribution model, ε N is the wake expansion parameter of the non-Gaussian distribution model, p 1 is the process parameter, x is the downstream distance of the fan, D T is the diameter of the fan impeller, C T is the fan thrust coefficient, I a is, γ is the yaw angle of the wind turbine

[0144] The shape function of the non-Gaussian distribution model is calculated by the following formula:

[0145]

[0146] a=b3 / (3*b 4 );

[0147]

[0148] c=(ba 2 ) 3 ;

[0149] u=(y / D T -μ(x)) / σ N ;

[0150]

[0151]

[0152] In the formula, is the shape function of the non-Gaussian distribution model, b 2 、b 3 、b 4 are the process parameters of the non-Gaussian distribution model, b 2 、b 3 、b 4 are process parameters of non-Gaussian distribution model, α 3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α 4 is the kurtosis parameter of each downstream position of the wind turbine, y is the crosswind distance length, a, b, c, u are intermediate parameters for simplified calculation and construction expression, σ N is the normalized standard deviation parameter, μ(x) is the wake offset of the entire wake area of ​​the yaw wind turbine wake, and ξ(u) is the intermediate conversion function of the non-Gaussian distribution function.

[0153] The maximum loss function of the non-Gaussian distribution model is calculated by the following formula:

[0154]

[0155]

[0156] Where C(x) is the maximum loss function of the non-Gaussian distribution model, y is the crosswind distance, and D T is the diameter of the fan impeller, C T is the thrust coefficient of the wind turbine, γ is the yaw angle of the wind turbine, σ is the standard deviation of the wake width, and p 2 is the maximum loss correction parameter for non-Gaussian distribution.

[0157] Thus, the non-Gaussian distribution prediction model of the yaw wind turbine wake wind speed loss is obtained:

[0158]

[0159] AU=U. -U;

[0160] Where U ∞ is the wind speed of the incoming flow at the hub height of the undisturbed fan, U is the wind speed of the fan wake, ΔU is the wind speed value lost after the incoming flow passes through the fan, and D T is the diameter of the fan impeller, σ is the standard deviation of the wake width, γ is the yaw angle of the fan, u is an intermediate parameter for simplifying the calculation structure expression, and p 2 is the maximum loss correction parameter of non-Gaussian distribution, C T is the fan thrust coefficient.

[0161] In step S6, the obtained wake wind speed loss non-Gaussian distribution prediction model is used, and the known wind turbine parameters and wind field parameters are taken as input. The lateral distribution of wind speed loss at the height of the wind turbine hub is calculated through the wake wind speed loss non-Gaussian distribution prediction model, that is, the wake wind speed loss distribution of the yaw wind turbine, that is, the ΔU value at different y positions is obtained.

[0162] Figure 2 and Figure 3 The distribution states of the wind turbine wake speed loss when the yaw angle is 0° and when it is not 0° are shown respectively. The non-Gaussian wake wind speed loss distribution prediction method for the yawed wind turbine provided by the present invention can be used to predict the wind turbine wake speed well according to different wind turbine types and incoming wind conditions in the wind farm, thereby improving the accuracy and optimization rate of wind resource assessment and real-time yaw control of the wind farm, thereby helping to improve the power generation efficiency of the wind farm, thereby obtaining greater power generation and economic benefits.

[0163] Example:

[0164] In order to further verify the non-Gaussian wake wind speed loss distribution prediction method for the yaw wind turbine provided by the present invention, the present invention is further illustrated by the following test case.

[0165] The model wind turbine WIRE-01 is used as the research object to test the effect of the non-Gaussian yaw wake wind speed loss prediction model. The wind turbine rotor diameter is 0.15m and the hub height is 0.125m. The incoming wind speed at the hub height is 4.88m / s and the turbulence intensity is 7%; the test yaw angles are 0 degrees, 10 degrees, 20 degrees and 30 degrees; the wind turbine thrust coefficient C T are 0.8217 (yaw=0), 0.7876 (yaw=10), 0.737 (yaw=20), and 0.665 (yaw=30).

[0166] The wake wind speed loss prediction model proposed in the present invention is used to predict the wind turbine wake distribution, and the prediction result of the model of the present invention (denoted as Non-Gauss Model) is compared with the prediction result of the traditional Jimenez wake model (denoted as Jimenez Model) and the prediction result of the two-dimensional Gaussian wake model (denoted as QI 2018). The distribution deviations of the three relative to the lateral distribution result profile of the wind speed loss simulated by the set skewness parameter kurtosis parameter (denoted as LES) are compared to verify the prediction performance of the method of the present invention. Figures 4 to 15 This is a comparison chart of the prediction results of the lateral distribution of wind speed loss predicted by the wind turbine wake model when the yaw angle is equal to 0°~30°. Figure 16 to Figure 19 This is a diagram showing the effect of model prediction evaluation. Figure 4 The comparison diagram of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 (where x is the downstream position of the wind turbine and d is the diameter of the wind turbine wheel) downstream of the wind turbine when the yaw angle yaw is 0°; Figure 5 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 0°; Figure 6 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 0°; Figure 7 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 10°; Figure 8 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 10°; Fig. 9 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 10°; Fig.10 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 20°; Fig.11 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 20°; Fig.12 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 20°; Fig.13 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=2 downstream of the wind turbine when the yaw angle yaw is 30°; Fig.14 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=4 downstream of the wind turbine when the yaw angle yaw is 30°; Fig.15 This is a comparison chart of the predicted results of the lateral distribution of wind speed loss at the position x / d=6 downstream of the wind turbine when the yaw angle yaw is 30°.

[0167] Through experimental tests and the above comparison Figures 4 to 15It can be seen that in this embodiment, the lateral distribution of the wind turbine wake wind speed loss at each yaw angle is predicted. Compared with the traditional Jimenez wake model and the two-dimensional Gaussian wake model, the model of the present invention shows higher accuracy for the simulated data by introducing the skewness parameter and the kurtosis parameter, and the gradual change phenomenon of the wind turbine wake wind speed loss in the whole wake area from the bimodal distribution of the near wake to the top hat distribution and then to the single-peak distribution in the far wake area is well reproduced and predicted. At the same time, it can be seen that the tail wind speed loss distribution prediction proposed by the present invention has better prediction effect and display for the offset of the wake and the asymmetry of the wake speed distribution at a larger yaw angle.

[0168] In addition, this embodiment also compares the prediction of the maximum wind speed loss value at each downstream position of the wind turbine using the wake wind speed loss prediction model proposed in the present invention, and compares the prediction result of the model of the present invention (denoted as Proposed) with the prediction result of the two-dimensional Gaussian wake model (denoted as Qian-Ishihara), and compares the distribution deviation of the wind speed loss lateral distribution result profile (denoted as LES) obtained by simulating the set skewness parameter kurtosis parameter and the wind tunnel test result (denoted as EXP), such as Fig.16 and Fig.17 As shown. Among them, Fig.16 This is a comparison chart of the predicted results of the maximum wind speed loss value at each downstream position of the wind turbine when the yaw angle is 10°. Fig.17 This is a comparison chart of the predicted results of the maximum wind speed loss value at each downstream position of the wind turbine when the yaw angle yaw is 20°.

[0169] like Fig.16 and Fig.17 As shown, in the prediction of the maximum wind speed loss value at each downstream position, compared with the two-dimensional Gaussian wake model (Qian-Ishihara) and the wind tunnel experimental data (EXP), the prediction results given by the model of the present invention are closer to the actual values ​​(corresponding to the wind tunnel experimental data EXP and the simulation data LES), thereby proving that the prediction results of the maximum wind speed loss value at each downstream position of the model of the present invention are more accurate.

[0170] This embodiment also compares the prediction results of the model proposed by the method of the present invention (denoted as Proposed) and the prediction results of the two-dimensional Gaussian wake model (denoted as Qian-Ishihara) for different wind speed loss intensities. Fig.18 and Fig.19 As shown. Among them, Fig.18 This is a comparison chart of the prediction results of different wind speed loss intensity when the yaw angle is 10°. Fig.19 This is a comparison chart of the predicted results of loss intensity at different wind speeds when the yaw angle yaw is 20°.

[0171] like Fig.18 and Fig.19 As shown, the data points predicted by the model (Proposed) of the present invention are compared with the two-dimensional Gaussian model (Qian-Ishihara). Through the distribution of discrete points, it can be seen that the prediction results of the present invention at different wind speed loss intensities are closer to the simulated data, indicating that the method of the present invention also has better prediction accuracy in the distribution area of ​​wind speed loss intensity. The present invention can well solve and accurately describe the phenomenon of wake wind speed loss of yaw wind turbines, and improve the prediction accuracy of wake speed, thereby improving the yaw control efficiency and the total power generation of wind farms, and improving economic benefits.

[0172] In summary, the present invention constructs a non-Gaussian wake velocity loss distribution prediction method for yaw wind turbines, and takes into account that the non-Gaussian distribution can increase or decrease the skewness, which can well describe and predict the asymmetry of the wake velocity distribution of the yaw wind turbine. At the same time, the gradual change of the distribution form of the wind turbine wake velocity in the entire wake area is well described and predicted by the change of kurtosis, including bimodal distribution, top hat distribution and unimodal distribution. Overall, compared with the existing wake model, the accuracy and distribution area of ​​the yaw wind turbine wake velocity prediction are improved, the authenticity is improved at the source, and the wind farm is helped to increase the power generation of the wind turbine through a high-accuracy yaw control strategy, and increase economic benefits and energy utilization.

[0173] Finally, it should be noted that although the present invention has been described with reference to the preferred embodiments of the present invention, it should be understood by those skilled in the art that various changes may be made in form and details without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for predicting the non-Gaussian wake wind speed loss distribution of a yaw wind turbine, characterized in that: The steps include: S1. Obtain the mechanical performance parameters and yaw angle parameters of the wind turbine, and obtain the wind field environment parameters at the location of the wind turbine through actual measurement; S2. According to the mechanical performance parameters and yaw angle parameters of the wind turbine, determine the offset angle θ of the wind turbine wake center in the near wake area. n , wind turbine wake boundary and wake width expansion coefficient; The deviation angle of the wind turbine wake center in the near wake area is calculated by the following formula: In the formula, θ n is the deviation angle of the wind turbine wake center in the near wake area, γ is the wind turbine yaw angle, C T is the fan thrust coefficient; According to the mechanical performance parameters and yaw angle parameters, the fan wake boundary and wake width expansion coefficient are calculated by the following formula: Where σ is the standard deviation of wake width, k is the wake width expansion coefficient, and D T is the diameter of the fan impeller, x is the downstream distance of the fan, ε and β are the wake boundary parameters, γ is the yaw angle of the fan, and C T is the fan thrust coefficient, I a is the ambient turbulence intensity; S3, according to the deviation angle θ of the wind turbine wake center in the near wake area n and the wake width expansion coefficient, the theoretical boundary position x0 between the near wake and the far wake is obtained, and then the wake offset μ1(x) in the near wake region and the wake offset μ2(x) in the far wake region of the wind turbine are determined; The theoretical boundary position of the near wake and far wake is calculated by the following formula: Where x0 is the theoretical boundary position of the near wake and far wake, σ0 is the standard deviation of the wake width at the boundary position of the near wake and far wake, and D T is the diameter of the fan impeller, k is the wake width expansion coefficient, ε is the wake boundary parameter, γ is the fan yaw angle, C T is the fan thrust coefficient, θ n is the offset angle of the wind turbine wake center in the near wake area; The wake deviation in the near wake area of ​​the wind turbine is calculated by the following formula: Where μ1(x) is the wake deviation in the near wake area of ​​the wind turbine, θ n is the deviation angle of the wind turbine wake center in the near wake area, x is the downstream distance of the wind turbine, D T is the diameter of the fan impeller; The wake offset in the far wake region is calculated using the following formula: Where μ2(x) is the wake deviation in the far wake area of ​​the wind turbine, D T is the diameter of the fan impeller, x is the downstream distance of the fan, x0 is the theoretical boundary position between the near wake and the far wake, C T is the thrust coefficient of the wind turbine, σ is the standard deviation of the wake width, σ0 is the standard deviation of the wake width at the boundary between the near wake and the far wake, θ n is the offset angle of the wind turbine wake center in the near wake area, k is the wake width expansion coefficient, and γ is the wind turbine yaw angle; S4, according to the mechanical performance parameters and yaw angle parameters of the fan, calculate the non-Gaussian skewness α3 and the kurtosis parameter α4 of the yaw fan wake at each downstream position of the fan, and then obtain the process parameters of the non-Gaussian distribution model; The non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine is calculated by the following formula: Kurtosis parameters at each downstream location of the fan: Where α3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α4 is the kurtosis parameter at each downstream position of the wind turbine, γ is the yaw angle of the wind turbine, x is the downstream distance of the wind turbine, and D T is the diameter of the fan impeller; According to the non-Gaussian skewness α3 and kurtosis parameter α4 of the wake, the process parameters of the non-Gaussian distribution model can be calculated by the following formula: K＝[1-0.06(3-α4)] 1 / 3 ； Where K, b2, b3, and b4 are all process parameters of the non-Gaussian distribution model, α3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, and α4 is the kurtosis parameter at each downstream position of the wind turbine; S5. According to the process parameters of the non-Gaussian distribution model and the wind field environment parameters where the wind turbine is located, determine the normalized standard deviation parameter σ when the non-Gaussian distribution model is transformed. N , and then get the type function of the non-Gaussian distribution model and the maximum loss function C(x), thereby determining the non-Gaussian distribution prediction model of the wake wind speed loss of the yaw wind turbine; The normalized standard deviation parameter when transforming the non-Gaussian distribution model is calculated by the following formula: In the formula, σ N is the normalized standard deviation parameter, k N is the wake expansion rate of the non-Gaussian distribution model, ε N Wake expansion rate of non-Gaussian distribution model, p1 is the process parameter, x is the downstream distance of the wind turbine, D T is the diameter of the fan impeller, C T is the fan thrust coefficient, I a is, γ is the yaw angle of the wind turbine; The shape function of the non-Gaussian distribution model is calculated by the following formula: a=b3 / (3*b4); c=(b-a 2 ) 3 ; u=(y / D T -μ(x)) / σ N ; In the formula, is the shape function of the non-Gaussian distribution model, b2, b3, and b4 are all process parameters of the non-Gaussian distribution model, α3 is the non-Gaussian skewness of the yaw wind turbine wake at each downstream position of the wind turbine, α4 is the kurtosis parameter at each downstream position of the wind turbine, y is the crosswind distance length, a, b, c, and u are intermediate parameters for simplifying the calculation and construction expression, and σ N is the normalized standard deviation parameter, μ(x) is the wake offset of the whole wake area of ​​the yaw wind turbine wake, ξ(u) is the intermediate conversion function of the non-Gaussian distribution function; The maximum loss function of the non-Gaussian distribution model is calculated by the following formula: Where C(x) is the maximum loss function of the non-Gaussian distribution model, y is the crosswind distance, and D T is the diameter of the fan impeller, C T is the thrust coefficient of the wind turbine, γ is the yaw angle of the wind turbine, σ is the standard deviation of the wake width, and p2 is the maximum loss correction parameter of the non-Gaussian distribution; The non-Gaussian distribution prediction model of the yaw wind turbine wake wind speed loss is calculated by the following formula: ΔU=U ∞ -U; Where U ∞ is the wind speed of the incoming flow at the hub height of the undisturbed fan, U is the wind speed of the fan wake, ΔU is the wind speed value lost after the incoming flow passes through the fan, and D T is the diameter of the fan impeller, σ is the standard deviation of the wake width, γ is the yaw angle of the fan, u is the intermediate parameter for simplifying the calculation structure expression, p2 is the maximum loss correction parameter of the non-Gaussian distribution, C T is the fan thrust coefficient; S6. Calculate and determine the wake wind speed loss distribution of the yaw wind turbine using the obtained wake wind speed loss non-Gaussian distribution prediction model.

2. A method for predicting the non-Gaussian wake wind speed loss distribution of a yaw wind turbine according to claim 1, characterized in that: In step S1, the relevant parameters of the wind turbine include hub height and impeller diameter; the wind field environmental parameters are the incoming wind speed and direction and environmental turbulence intensity information at the location of the wind turbine obtained based on wind tower data or relevant on-site measured data.

3. The method for predicting the non-Gaussian wake wind speed loss distribution of a yaw wind turbine according to claim 1, characterized in that: In step S6: using the obtained non-Gaussian distribution prediction model of wake wind speed loss, taking the known wind turbine parameters and wind field parameters as input, the lateral distribution of wind speed loss at the height of the wind turbine hub is calculated through the non-Gaussian distribution prediction model of wake wind speed loss, that is, the wake wind speed loss distribution of the yaw wind turbine is obtained.

Images