A wind turbine yaw control method and system taking load calculation into account
Through grid search method and Pareto search optimization of yaw control parameters of wind turbines, the problem of equivalent fatigue load increase in wind turbines when pursuing power generation efficiency is solved, the power generation power increase and load reduction are achieved, and the unit is extended service life.
Patent Information
- Application Number
- CN202210989753.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-18
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-08-18
AI Technical Summary
While pursuing power generation efficiency, the existing wind turbine yaw control strategy is difficult to effectively reduce the equivalent fatigue load, affecting the smooth operation of the unit throughout its life cycle.
The grid search method and Pareto optimization method are used to determine the threshold and delay time in the yaw control parameters, optimize the nacelle position of the wind turbine to minimize the equivalent fatigue load and maximize the power generation, and the rain flow counting method is used to calculate the The initial load is calculated and corrected, and the cabin position is finally adjusted for optimization.
It increases the power generation power of the wind turbine, while reducing the equivalent fatigue load and extending the service life of the unit.
Smart Images

Figure CN115288929B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wind power technology, and in particular to a yaw control method and system for wind turbines that takes load calculation into account. Background Technology
[0002] With the rapid development of the economy and society, the demand for and consumption of energy in human production and daily life are constantly increasing. The development of a green and low-carbon economy has gradually replaced the development model that sacrifices the ecological environment for economic growth, and the utilization and development of renewable energy is becoming a trend of the times. The full development and utilization of wind and solar energy is undoubtedly essential. Although wind resources in nature are inexhaustible, they also exhibit fluctuations and randomness. Therefore, it is necessary to implement certain controls on wind turbine and wind farm scheduling to improve the power generation efficiency of the units and thus achieve efficient utilization of wind resources.
[0003] The control of wind turbines aims to improve the performance of individual wind turbine units, mainly including yaw control to achieve maximum wind energy conversion efficiency, pitch control to reduce imbalance, asymmetry, and intermittent loads, and torque control to improve grid friendliness. While existing yaw control strategies and related optimization methods are effective and can improve the power generation efficiency of individual units, it is also important to pay attention to and study the impact of increased loads on the stable operation of wind turbine units throughout their entire life cycle while pursuing power generation efficiency. Summary of the Invention
[0004] The purpose of this invention is to provide a yaw control method and system for wind turbines that takes load calculation into account, which can improve the power generation of wind turbines and at the same time reduce the equivalent fatigue load.
[0005] To achieve the above objectives, the present invention provides the following solution:
[0006] A wind turbine yaw control method that takes load calculation into account includes:
[0007] Multiple sets of historical yaw data were determined using a grid search method. Each set of historical yaw data included yaw control parameters, the equivalent fatigue load of the wind turbine during yaw control under the action of the yaw control parameters, and the power generation. The yaw control parameters included a threshold and a delay time. The threshold represented the maximum allowable deviation angle between the nacelle and the wind direction. When the deviation angle between the nacelle and the wind direction exceeded the threshold and remained for a certain delay time, yaw control of the wind turbine was initiated.
[0008] Based on multiple sets of historical yaw data, with the optimization objectives of minimizing equivalent fatigue load and maximizing power generation, the Pareto optimization method is used to optimize the threshold and delay time, and obtain the Pareto optimal threshold and Pareto optimal delay time.
[0009] Collect the current inflow wind direction and the azimuth angle of the wind turbine nacelle;
[0010] The current deviation angle is determined based on the current inflow wind direction and the nacelle azimuth angle of the wind turbine.
[0011] The nacelle orientation of the wind turbine is adjusted based on the current deviation angle, the Pareto optimal threshold, and the Pareto optimal delay time.
[0012] Optionally, the wind turbine includes a generator, blades, and a tower; the equivalent fatigue load of the wind turbine includes the generator torque equivalent fatigue load, the blade root bending moment equivalent fatigue load, and the tower thrust equivalent fatigue load.
[0013] The method of using grid search to determine multiple sets of historical yaw data specifically includes:
[0014] Multiple pairs of yaw control parameters are determined using a grid search method; each pair of yaw control parameters includes a threshold and a delay time.
[0015] For any pair of yaw control parameters, under the influence of the yaw control parameters, wind speed environment, pitch angle position and rated torque, the initial power generation, generator torque stress time history curve, blade root bending moment stress time history curve, tower top thrust stress time history curve and nacelle deviation angle are obtained within a set time period; wind speed environment, pitch angle position and rated torque are pre-applied to the wind turbine.
[0016] Based on the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve, and tower top thrust stress time history curve, the rainflow counting method is used to determine the initial generator torque equivalent fatigue load, initial blade root bending moment equivalent fatigue load, and initial tower thrust equivalent fatigue load in the corresponding time period.
[0017] Based on the nacelle's deviation angle and deviation loss coefficient, the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root moment equivalent fatigue load, and the initial tower thrust equivalent fatigue load are corrected to obtain the final power generation, generator torque equivalent fatigue load, blade root moment equivalent fatigue load, and tower thrust equivalent fatigue load under the corresponding yaw control parameters; the deviation loss coefficient is determined in advance through dynamic simulation fitting.
[0018] Optionally, the step of determining the initial generator torque equivalent fatigue load, initial blade root bending moment equivalent fatigue load, and initial tower thrust equivalent fatigue load within the corresponding time period using the rainflow counting method based on the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve, and tower top thrust stress time history curve specifically includes:
[0019] Obtain the fatigue curve slopes of generator materials, blade root materials, and tower materials;
[0020] Using the rainflow counting method, the stress cycle period of the generator torque is statistically analyzed based on the generator torque stress time history curve to obtain the first statistical result;
[0021] The cumulative rainflow spectrum of the generator is obtained by linearly superimposing the first statistical results based on the linear damage accumulation rule.
[0022] Based on the cumulative rainflow spectrum of the generator and the slope of the fatigue curve of the generator material, the equivalent fatigue load of the initial generator torque in the corresponding time period is determined;
[0023] Using the rainflow counting method, the stress cycle period of the blade root is statistically analyzed based on the time history curve of the blade root bending moment stress, and a second statistical result is obtained.
[0024] The second statistical results are linearly superimposed based on the linear damage accumulation rule to obtain the leaf root cumulative rainflow spectrum;
[0025] Based on the cumulative rainflow spectrum at the blade root and the slope of the fatigue curve of the blade root material, the initial equivalent fatigue load at the blade root in the corresponding time period is determined.
[0026] Using the rainflow counting method, the stress cycle period of the tower is statistically analyzed based on the time history curve of the thrust stress at the top of the tower, and a third statistical result is obtained.
[0027] The cumulative rainflow spectrum of the tower is obtained by linearly superimposing the third statistical results based on the linear damage accumulation rule.
[0028] Based on the cumulative rainflow spectrum of the tower and the slope of the fatigue curve of the tower material, the initial equivalent fatigue load of the tower in the corresponding time period is determined.
[0029] Optionally, the equivalent fatigue load of the initial generator torque can be calculated using the following formula:
[0030]
[0031] Among them, R DEL R is the initial generator torque equivalent fatigue load. i Let n be the load amplitude during the i-th stress cycle in the cumulative rainflow spectrum of the generator. i denoted as the number of load cycles accumulated in the i-th stress cycle of the cumulative rainflow spectrum of the generator, where i>0, and m is the slope of the fatigue curve of the generator material.
[0032] Optionally, the deviation loss coefficient includes: power deviation loss coefficient, generator torque deviation loss coefficient, blade root deviation loss coefficient, and tower deviation loss coefficient;
[0033] The process involves correcting the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root moment equivalent fatigue load, and the initial tower thrust equivalent fatigue load based on the nacelle deviation angle and deviation loss coefficient, respectively, to obtain the final power generation, generator torque equivalent fatigue load, blade root moment equivalent fatigue load, and tower thrust equivalent fatigue load under the corresponding yaw control parameters. Specifically, this includes:
[0034] Based on the cabin deviation angle and the power deviation loss coefficient, the initial power generation is corrected to obtain the final power generation under the corresponding yaw control parameters; the power deviation loss coefficient is determined in advance by dynamic simulation fitting of historical power generation and historical cabin deviation angle.
[0035] Based on the nacelle deviation angle and the generator torque deviation loss coefficient, the initial generator torque equivalent fatigue load is corrected to obtain the final generator torque equivalent fatigue load under the corresponding yaw control parameters; the generator torque deviation loss coefficient is determined in advance by dynamic simulation fitting of the historical generator torque equivalent fatigue load and the historical deviation angle of the nacelle.
[0036] Based on the nacelle deviation angle and the blade root deviation loss coefficient, the initial blade root bending moment equivalent fatigue load is corrected to obtain the final blade root bending moment equivalent fatigue load under the corresponding yaw control parameters; the blade root deviation loss coefficient is determined in advance by dynamic simulation fitting of the historical blade root bending moment equivalent fatigue load and the historical deviation angle of the nacelle.
[0037] Based on the nacelle deviation angle and the tower deviation loss coefficient, the initial tower thrust equivalent fatigue load is corrected to obtain the final tower thrust equivalent fatigue load under the corresponding yaw control parameters; the tower deviation loss coefficient is determined in advance by dynamic simulation fitting of the historical tower thrust equivalent fatigue load and the historical deviation angle of the nacelle.
[0038] Alternatively, the final power generation can be determined using the following formula:
[0039]
[0040] Among them, T p For the final power generation, T p,0 f is the initial power generation capacity. p (θ) is the power deviation loss coefficient, ω is the first harmonic, θ is the cabin deviation angle, and a p,0 a p,1a p,2 b p,1 b p,2 All are undetermined coefficients, a p,0 a p,1 a p,2 b p,1 b p,2 Determined through dynamic simulation fitting.
[0041] Optionally, adjusting the nacelle orientation of the wind turbine based on the current deviation angle, the Pareto optimal threshold, and the Pareto optimal delay time specifically includes:
[0042] The deviation angle is filtered to obtain the filtered deviation angle;
[0043] Determine whether the filter bias angle is greater than the Pareto optimal threshold;
[0044] If the filter deviation value is less than or equal to the Pareto optimal threshold, the nacelle orientation of the wind turbine will not be adjusted.
[0045] If the filter deviation angle is greater than the Pareto optimal threshold, then timing begins;
[0046] Determine whether the timing time is greater than the Pareto optimal delay time;
[0047] If the timing time is less than or equal to the Pareto optimal delay time, then the nacelle orientation of the wind turbine will not be adjusted or the adjustment of the nacelle orientation of the wind turbine will be stopped.
[0048] If the timing time is greater than the Pareto optimal delay time, it is determined whether the nacelle orientation is equal to the set orientation value. If the nacelle orientation is equal to the set orientation value, the nacelle orientation of the wind turbine is not adjusted or the adjustment of the nacelle orientation of the wind turbine is stopped. If the nacelle orientation is not equal to the set orientation value, the adjustment of the nacelle orientation of the wind turbine begins or the adjustment of the nacelle orientation of the wind turbine continues.
[0049] To achieve the above objectives, the present invention also provides the following solution:
[0050] A wind turbine yaw control system that takes load calculation into account includes:
[0051] The historical data determination unit is used to determine multiple sets of historical yaw data using a grid search method. Each set of historical yaw data includes yaw control parameters, the equivalent fatigue load of the wind turbine and its power generation during yaw control under the action of the yaw control parameters. The yaw control parameters include a threshold and a delay time. The threshold represents the maximum allowable deviation angle between the nacelle and the wind direction. When the deviation angle between the nacelle and the wind direction exceeds the threshold and continues for a delay time, yaw control of the wind turbine begins.
[0052] The optimization unit, connected to the historical data determination unit, is used to optimize the threshold and delay time based on multiple sets of historical yaw data, with the optimization objectives of minimizing equivalent fatigue load and maximizing power generation, using the Pareto optimization method to obtain the Pareto optimal threshold and Pareto optimal delay time.
[0053] The azimuth acquisition unit is used to acquire the current inflow wind direction and the azimuth angle of the wind turbine nacelle;
[0054] A deviation angle determination unit, connected to the azimuth angle acquisition unit, is used to determine the current deviation angle based on the current inflow wind direction and the azimuth angle of the wind turbine nacelle.
[0055] The adjustment unit is connected to the optimization unit and the deviation angle determination unit respectively, and is used to adjust the nacelle orientation of the wind turbine according to the current deviation angle, the Pareto optimal threshold and the Pareto optimal delay time.
[0056] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: Based on Pareto optimality theory and grid search algorithm, considering the influence of yaw control process on the aerodynamic load of wind turbine, the threshold and delay time in the yaw control strategy of existing wind turbine are optimized in multiple objectives, thereby improving the power generation of wind turbine and reducing the equivalent fatigue load of wind turbine at the same time. Attached Figure Description
[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. 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 effort.
[0058] Figure 1 This is a flowchart of the wind turbine yaw control method that takes load calculation into account according to the present invention;
[0059] Figure 2 This is a schematic diagram of a linearized model of a wind turbine.
[0060] Figure 3 This is a schematic diagram of the wind direction deviation angle;
[0061] Figure 4 A schematic diagram of the Pareto front;
[0062] Figure 5 A schematic diagram of the yaw nacelle control strategy for a wind turbine.
[0063] Figure 6 This is a schematic diagram of a passive yaw control system for a wind turbine.
[0064] Figure 7 This is a schematic diagram of the module structure of the wind turbine yaw control system that takes load calculation into account in this invention.
[0065] Symbol explanation:
[0066] Historical data determination unit-1, optimization unit-2, azimuth angle acquisition unit-3, deviation angle determination unit-4, adjustment unit-5. Detailed Implementation
[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0068] The purpose of this invention is to provide a yaw control method and system for wind turbines that takes load calculation into account. Based on Pareto optimality theory and grid search algorithm, it considers the impact of the yaw control process on the aerodynamic load of the wind turbine and performs multi-objective optimization on the threshold and delay time in the yaw control strategy of existing wind turbines, thereby improving the power generation of the wind turbine and reducing the equivalent fatigue load of the wind turbine at the same time.
[0069] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0070] like Figure 1 As shown, the wind turbine yaw control method of the present invention, which takes load calculation into account, includes:
[0071] S1: Multiple sets of historical yaw data are determined using a grid search method. Each set of historical yaw data includes yaw control parameters, the equivalent fatigue load of the wind turbine during yaw control under the action of the yaw control parameters, and the power generation. The yaw control parameters include a threshold and a delay time. The threshold represents the maximum allowable deviation angle between the wind turbine nacelle and the wind direction. When the deviation angle between the nacelle and the wind direction exceeds the threshold and continues for a certain delay time, yaw control of the wind turbine begins.
[0072] Specifically, a wind turbine includes a nacelle, generator, blades, and tower. The equivalent fatigue load of a wind turbine includes the equivalent fatigue load of generator torque, the equivalent fatigue load of blade root bending moment, and the equivalent fatigue load of tower thrust.
[0073] S2: Based on multiple sets of historical yaw data, with the optimization objectives of minimizing equivalent fatigue load and maximizing power generation, the Pareto optimization method is used to optimize the threshold and delay time, and obtain the Pareto optimal threshold and Pareto optimal delay time.
[0074] S3: Collect the current inflow wind direction and the azimuth angle of the wind turbine nacelle.
[0075] S4: Determine the current deviation angle based on the current inflow wind direction and the nacelle azimuth angle of the wind turbine.
[0076] S5: Adjust the nacelle orientation of the wind turbine based on the current deviation angle, the Pareto optimal threshold, and the Pareto optimal delay time.
[0077] This invention selects a threshold and a delay time as optimized control parameters based on the principles of yaw control systems and commonly used yaw control logic. The yaw deviation threshold refers to the maximum acceptable deviation angle between the nacelle and the wind direction during wind turbine operation. The yaw delay time acts as a filter for the yaw threshold; it can be understood as the yaw actuator initiating wind-following actions only after the yaw deviation angle exceeds the threshold and persists for a certain period.
[0078] This invention is mainly based on Pareto optimality theory to improve the yaw control strategy of existing wind turbines. It takes into account the impact of the yaw control process on the aerodynamic load of the wind turbine and performs multi-objective optimization of the yaw control of the wind turbine to achieve the goal of increasing power and reducing load.
[0079] Furthermore, step S1 specifically includes:
[0080] S11: Multiple pairs of yaw control parameters are determined using a grid search method. Each pair of yaw control parameters includes a threshold and a delay time.
[0081] S12: For any pair of yaw control parameters, under the action of the yaw control parameters, wind speed environment, pitch angle position and rated torque, obtain the initial power generation, generator torque stress time history curve, blade root bending moment stress time history curve, tower top thrust stress time history curve and nacelle deviation angle within a set time period; wind speed environment, pitch angle position and rated torque are pre-applied to the wind turbine.
[0082] S13: Based on the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve, and tower top thrust stress time history curve, the rainflow counting method is used to determine the initial generator torque equivalent fatigue load, initial blade root bending moment equivalent fatigue load, and initial tower thrust equivalent fatigue load within the corresponding time period.
[0083] This invention provides an in-depth analysis of the main aerodynamic loads experienced by wind turbines during yaw control. The load on a wind turbine is the collective term for the forces or moments acting on its various components. Wind turbine loads are complex and can be broadly categorized into five types: aerodynamic loads, gravity loads, inertial loads, mechanical loads, and impact loads. Aerodynamic loads, also known as aero-force loads, primarily refer to the loads exerted on the wind turbine by the inflow wind. Due to the complex and variable nature of the inflow wind environment, the dynamic changes in the turbine's aerodynamic loads are quite complex and exhibit a degree of periodicity. Therefore, the impact of aerodynamic loads on the turbine's fatigue is crucial among all load types. Furthermore, the blades and tower are considered the main components of the wind turbine that bear the aerodynamic loads. During yaw control, the aerodynamic loads of the turbine undergo significant changes due to the turbine's deviation from the wind direction. These changes are primarily reflected in the stress on the blades and tower.
[0084] Therefore, this invention selects the aerodynamic loads of blade root bending moment and tower top thrust, which have a significant impact on the stable operation of the unit, as the targets for threshold and delay time optimization, and calculates the equivalent fatigue load. Regarding the load distribution on the blades, the closer to the blade root, the more drastic the frequency of load changes and the amplitude of vibration. Similarly, a similar trend exists in the tower distribution; the loads closer to the top of the tower exhibit more typical fatigue variation cycles, namely, blade root bending moment and tower top thrust.
[0085] Specifically, step S13 includes:
[0086] S131: Obtain the fatigue curve slope of the generator material, the fatigue curve slope of the blade root material, and the fatigue curve slope of the tower material.
[0087] S132: Using the rainflow counting method, based on the generator torque stress time history curve, the stress cycle period of the generator torque is statistically analyzed to obtain the first statistical result.
[0088] S133: The first statistical results are linearly superimposed based on the linear damage accumulation rule to obtain the generator cumulative rainflow spectrum. Specifically, the stress on the generator within a set time period is divided into multiple ranges. For any range, the load amplitude within that range is determined based on the stress on the generator at each moment within that range. The load amplitudes within each range are cyclically accumulated using the rainflow counting method to obtain the generator cumulative rainflow spectrum.
[0089] The cumulative rainflow spectrum represents the cumulative number of cycles of load cycle amplitude across different intervals. To plot the cumulative rainflow spectrum, the amplitude range of stress cycles needs to be linearly accumulated. First, the amplitude of all cycles is divided into intervals (according to the IEC standard, the number of intervals should be no less than 100 during the evaluation of wind turbine loads), and then the number of cycles occurring in a given amplitude interval is calculated.
[0090] S134: Determine the initial generator torque equivalent fatigue load within the corresponding time period based on the cumulative rainflow spectrum of the generator and the slope of the fatigue curve of the generator material.
[0091] Specifically, a weighted average of the cumulative rainflow spectrum of the generator is calculated, with the weighting index using the slope of the SN fatigue curve of the generator material. The equivalent fatigue load of the initial generator torque is calculated using the following formula:
[0092]
[0093] Among them, R DEL R is the initial generator torque equivalent fatigue load. i Let n be the load amplitude during the i-th stress cycle in the cumulative rainflow spectrum of the generator. i Let be the number of load cycles accumulated in the i-th stress cycle of the cumulative rainflow spectrum of the generator, where i>0, and m be the slope of the fatigue curve of the generator material. Reference values for the slope of the SN curve of commonly used wind turbine materials are shown in Table 1.
[0094] Table 1 Reference Values of Slope of SN Curve for Commonly Used Materials in Wind Turbine Components
[0095] Types of materials Welding steel Ductile iron Glass fiber reinforced plastics SN curve slope 3-5 6-8 10-12
[0096] S135: Using the rainflow counting method, based on the time history curve of the blade root bending moment stress, the stress cycle period of the blade root is statistically analyzed to obtain the second statistical result.
[0097] S136: The second statistical results are linearly superimposed based on the linear damage accumulation rule to obtain the leaf root cumulative rainflow spectrum.
[0098] S137: Determine the initial equivalent fatigue load of the blade root within the corresponding time period based on the cumulative rainflow spectrum of the blade root and the slope of the fatigue curve of the blade root material.
[0099] S138: Using the rainflow counting method, based on the time history curve of the thrust stress at the top of the tower, the stress cycle period of the tower is statistically analyzed to obtain the third statistical result.
[0100] S139: The cumulative rainflow spectrum of the tower is obtained by linearly superimposing the third statistical results based on the linear damage accumulation rule.
[0101] S1310: Determine the initial equivalent fatigue load of the tower within the corresponding time period based on the cumulative rainflow spectrum of the tower and the slope of the fatigue curve of the tower material.
[0102] The definition and calculation method of equivalent fatigue load have become standardized within the industry and enjoy high recognition throughout the wind power generation field. Therefore, this invention selects equivalent fatigue load as the quantitative indicator for the quantitative evaluation of unit load. Equivalent fatigue load is the stress on the material (blade, tower). By using the rules of the rainflow counting method, a histogram of stress cycle amplitude-mean-cycle number is obtained. Ignoring the mean of stress cycles, only the amplitude is linearly accumulated to obtain the cumulative rainflow spectrum.
[0103] The calculation formulas for the initial blade root equivalent fatigue load, the initial tower equivalent fatigue load, and the initial generator torque equivalent fatigue load are similar. The only difference is the value of the load amplitude, the number of load cycles, and the slope of the fatigue curve in each range of the cumulative rainflow spectrum of the blade and the cumulative rainflow spectrum of the tower. The specific formulas will not be elaborated here.
[0104] In this embodiment, a linearized model of the wind turbine is constructed based on relevant parameters of a certain type of wind turbine. The inputs of the model include wind speed, blade pitch angle position, and generator rated torque. The outputs are power generation, blade root bending moment equivalent fatigue load, tower top thrust equivalent fatigue load, and generator torque equivalent fatigue load. Figure 2 As shown.
[0105] S14: Based on the nacelle deviation angle and deviation loss coefficient, the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root moment equivalent fatigue load, and the initial tower thrust equivalent fatigue load are corrected respectively to obtain the final power generation, generator torque equivalent fatigue load, blade root moment equivalent fatigue load, and tower thrust equivalent fatigue load under the corresponding yaw control parameters; the deviation loss coefficient is determined in advance by dynamic simulation fitting.
[0106] In this embodiment, the deviation loss coefficient includes: power deviation loss coefficient, generator torque deviation loss coefficient, blade root deviation loss coefficient, and tower deviation loss coefficient.
[0107] Step S14 specifically includes:
[0108] S141: Based on the nacelle deviation angle and the power deviation loss coefficient, the initial power generation is corrected to obtain the final power generation under the corresponding yaw control parameters. The power deviation loss coefficient is determined in advance by dynamic simulation fitting of historical power generation and historical nacelle deviation angles.
[0109] Specifically, the absolute azimuth is defined as true north, and the zero-degree point is recorded as the absolute direction. Figure 3 illustrates the relationship between the inflow wind direction, the nacelle azimuth angle, and the wind direction deviation angle. Typically, the yaw deviation value θ of the wind turbine is equal to the current absolute wind direction angle θ. W -Current cabin orientation θN When the deviation angle between the nacelle and the current absolute wind direction is 0°, the generator's power output or load value is denoted as T0. Assume there exists a mapping relationship f with respect to the deviation angle θ. x (θ), such that when the nacelle position deviates from the current wind direction by an angle θ, the generator's power output or load value satisfies the following formula:
[0110] T = T0·f x (θ);
[0111] Where x represents the power generation or equivalent fatigue load. f(θ)=a0+a1 cos(ωθ)+ b1 sin(ωθ)+a2cos(2ωθ)+b2 sin(2ωθ), where ω is the first harmonic and 2ω is the second harmonic, and a0, a1, a2, b1, and b2 are all undetermined coefficients.
[0112] Depending on the selected variable x (power generation, equivalent fatigue load of generator torque, equivalent fatigue load of blade root bending moment, equivalent fatigue load of tower top thrust), different mapping relationships (relationship between deviation angle and variable) should be obtained, which are achieved by data fitting through dynamic simulation.
[0113] Specifically, the final power generation capacity is determined using the following formula:
[0114]
[0115] Among them, T p For the final power generation, T p,0 f is the initial power generation capacity. p (θ) is the power deviation loss coefficient, ω is the first harmonic, θ is the cabin deviation angle, and a p,0 a p,1 a p,2 b p,1 b p,2 All are undetermined coefficients, a p,0 a p,1 a p,2 b p,1 b p,2 Determined through dynamic simulation fitting.
[0116] S142: Based on the nacelle deviation angle and the generator deviation loss coefficient, the initial generator torque equivalent fatigue load is corrected to obtain the final generator torque equivalent fatigue load under the corresponding yaw control parameters. The generator torque deviation loss coefficient is determined in advance by dynamic simulation fitting of the historical generator torque equivalent fatigue load and the historical nacelle deviation angle.
[0117] S143: Based on the nacelle deviation angle and the blade root deviation loss coefficient, the initial blade root bending moment equivalent fatigue load is corrected to obtain the final blade root bending moment equivalent fatigue load under the corresponding yaw control parameters. The blade root deviation loss coefficient is determined in advance by dynamic simulation fitting of the historical blade root bending moment equivalent fatigue load and the historical deviation angle of the nacelle.
[0118] S144: Based on the nacelle deviation angle and the tower deviation loss coefficient, the initial tower thrust equivalent fatigue load is corrected to obtain the final tower thrust equivalent fatigue load under the corresponding yaw control parameters. The tower deviation loss coefficient is determined in advance by dynamic simulation fitting of historical tower thrust equivalent fatigue loads and historical nacelle deviation angles.
[0119] In this embodiment, the wind direction was set during the dynamic simulation fitting of the deviation loss coefficient, such as deflecting from +45 degrees to -45 degrees. The time history data of the output variables (power generation, equivalent fatigue load of generator torque, equivalent fatigue load of blade root bending moment, and equivalent fatigue load of tower thrust) were extracted, processed, and the fitting result obtained is the deviation loss coefficient.
[0120] The calculation formulas for the final generator torque equivalent fatigue load, blade root bending moment equivalent fatigue load, and tower thrust equivalent fatigue load are similar to those for calculating power generation, with the only difference being T. p,0 a p,0 a p,1 a p,2 b p,1 b p,2 The specific values for will not be elaborated here.
[0121] This embodiment selects the yaw threshold and yaw delay time of the unit as benchmarks based on on-site unit parameters (such as blade length, tower height, cut-in and cut-out wind speeds, etc.), and uses a grid search method to traverse and optimize the two parameters. The grid search method is an exhaustive search method for specified parameter values, typically applicable when there are multiple parameters to be optimized. By estimating the value range of each parameter, a grid is constructed representing all possible permutations and combinations of each parameter. A turbulent wind environment with a sinusoidal wind direction is set, and all combinations in the grid are then verified and evaluated to find the optimal solution. The equivalent fatigue loads of blade root bending moment and tower top thrust under different yaw control parameters (threshold and delay time) are used as the X and Y axes of a three-dimensional coordinate system, respectively, with the negative of the generated power as the Z axis, to create a three-dimensional scatter plot. This qualitatively confirms the existence of an optimal solution for yaw control optimization targeting equivalent fatigue load and equivalent power generation.
[0122] The core of this invention is based on Pareto optimization theory. Therefore, it is necessary to explain that the objective of this invention is to increase the generating power of the unit while minimizing the load increase and achieving load reduction. A Pareto relationship exists among the multiple objectives of wind turbine yaw control optimization, thus the multi-objective optimization problem can be solved using Pareto optimality theory.
[0123] Pareto optimization theory was proposed in 1906 by economics professor Vilfreto Pareto. Here's a simplified explanation of the Pareto optimal front: Pareto optimal solutions to multi-objective optimization problems generally exist in multiple ways. Using optimization objectives f1 and f2 as two-dimensional coordinate axes in a planar space, in this embodiment, optimization objective f1 is the generated power, and f2 is the equivalent fatigue load (generator torque equivalent fatigue load, blade root bending moment equivalent fatigue load, tower top thrust equivalent fatigue load). All feasible solutions to the corresponding optimization problem constitute the feasible region in this plane. This feasible region has a boundary such that the feasible solutions corresponding to all points on the boundary are superior to the feasible solutions inside the boundary, such as... Figure 4 As shown, the curve containing the solid points is called the Pareto front of the optimization problem. The objective values corresponding to points B and C are both better than those of point D. Point A on the Pareto front is better than B and C, while other points on the Pareto front are no different from point A.
[0124] In this embodiment, the equivalent fatigue loads of the blade root bending moment and the tower top thrust are taken separately, and Pareto optimality is solved with the equivalent power generation (power output) respectively. The relationship between the reference point and the Pareto front is also analyzed.
[0125] To visually observe whether there is an optimization relationship between the two sets of Pareto optimalities, based on the three-dimensional Pareto front, the points included in the Pareto fronts of blade root moment and tower top thrust are marked in another Pareto optimization scatter plot, and a solution is sought that satisfies the three objectives of power-blade root moment-to-tower top thrust as much as possible.
[0126] Taking the equivalent fatigue load of blade root bending moment and the equivalent power generation as two optimization objectives, it is evident that this optimization problem has a Pareto optimal point. Similarly, taking the equivalent fatigue load of tower top thrust and the equivalent power generation as two optimization objectives, it is also evident that this optimization problem has a Pareto optimal point. Therefore, the multi-objective optimization problem of wind turbine yaw control based on Pareto optimization theory has a Pareto optimal solution, and theoretically, it can find an optimal solution that is superior to the benchmark point for multiple objectives compared to the control strategies of known commercial wind turbines.
[0127] This invention aims to optimize a set of yaw control parameters (threshold and delay time) to simultaneously increase power generation and reduce fatigue load. Pareto optimization theory is introduced, and the yaw control parameters of the wind turbine are optimized and solved. Pareto front analysis is then used to verify the control objective of increasing power and reducing load.
[0128] Furthermore, step S5 specifically includes:
[0129] S51: The deviation angle is filtered to obtain a filtered deviation angle. Specifically, the filtering process includes low-pass filtering and high-pass filtering. After low-pass filtering of the deviation angle, the filtered deviation angle Err is obtained. slow .
[0130] S52: Determine whether the filter deviation angle is greater than the Pareto optimal threshold.
[0131] S53: If the filter deviation value is less than or equal to the Pareto optimal threshold, the nacelle orientation of the wind turbine will not be adjusted.
[0132] S54: If the filter deviation angle is greater than the Pareto optimal threshold, then start timing.
[0133] S55: Determine the timing time T start Whether it is greater than the Pareto optimal delay time.
[0134] S56: If the timing time is less than or equal to the Pareto optimal delay time, then the nacelle orientation of the wind turbine will not be adjusted or the adjustment of the nacelle orientation of the wind turbine will be stopped.
[0135] S57: If the timing time is greater than the Pareto optimal delay time, then determine whether the cabin orientation is equal to the set orientation value.
[0136] S58: If the nacelle orientation is equal to the set orientation value, then the nacelle orientation of the wind turbine will not be adjusted or the adjustment of the nacelle orientation of the wind turbine will be stopped.
[0137] S59: If the nacelle orientation is not equal to the set orientation value, start adjusting the nacelle orientation of the wind turbine, or continue adjusting the nacelle orientation of the wind turbine. For example... Figure 5 The diagram shows the specific yaw nacelle control strategy for a wind turbine. In the diagram, Tc is the inertial filter time constant. Tc = 60s indicates that the filter time constant is set to 60s, resulting in the filtered deviation angle sequence.
[0138] This invention starts from the typical aerodynamic load generation mechanism of wind turbines, uses the rainflow counting method to calculate the equivalent fatigue load, and uses it as a quantitative parameter for wind turbine fatigue analysis. It corrects the wind direction deviation for the equivalent fatigue loads of initial power generation, initial generator torque, initial blade root bending moment, and initial tower thrust. The equivalent fatigue load is then introduced into the existing yaw control strategy. Based on the concept of the deviation loss coefficient derived from the linear fitting of dynamic process simulation data, parameter optimization is performed using a grid search method. Pareto front analysis is conducted on the results to optimize the threshold and delay time of the yaw control strategy. By optimizing the threshold and delay time of the yaw control strategy with equivalent power generation and equivalent fatigue load as objectives, the power generation of the wind turbine is improved while simultaneously reducing the equivalent fatigue load.
[0139] like Figure 6 As shown, the passive yaw control system of a wind turbine mainly includes wind measuring devices (anemometer, wind vane), yaw controller, power amplifier, yaw actuator (yaw motor), yaw counter, detection elements, etc. Its working principle can be roughly described as follows: the anemometer and wind vane collect the wind speed and wind direction of the current incoming wind and input them into the yaw controller. After filtering the incoming wind signal, the yaw controller judges the start and stop actions of yaw control according to the yaw control logic to achieve the nacelle's wind control target.
[0140] like Figure 7 As shown, the wind turbine yaw control system of the present invention, which takes load calculation into account, includes: a historical data determination unit 1, an optimization unit 2, an azimuth angle acquisition unit 3, a deviation angle determination unit 4, and an adjustment unit 5.
[0141] The historical data determination unit 1 is used to determine multiple sets of historical yaw data using a grid search method. Each set of historical yaw data includes yaw control parameters, the equivalent fatigue load of the wind turbine during yaw control under the action of the yaw control parameters, and the power generation. The yaw control parameters include a threshold and a delay time. The threshold represents the maximum allowable deviation angle between the nacelle and the wind direction. When the deviation angle between the nacelle and the wind direction exceeds the threshold and continues for a delay time, yaw control of the wind turbine begins.
[0142] The optimization unit 2 is connected to the historical data determination unit 1. The optimization unit 2 is used to optimize the threshold and delay time based on multiple sets of historical yaw data, with the minimum equivalent fatigue load and the maximum power generation as the optimization objectives, and adopts the Pareto optimization method to obtain the Pareto optimal threshold and Pareto optimal delay time.
[0143] Azimuth acquisition unit 3 is used to acquire the current inflow wind direction and the azimuth of the wind turbine nacelle.
[0144] The deviation angle determination unit 4 is connected to the azimuth angle acquisition unit 3. The deviation angle determination unit 4 is used to determine the current deviation angle based on the current inflow wind direction and the azimuth angle of the wind turbine nacelle.
[0145] The adjustment unit 5 is connected to the optimization unit 2 and the deviation angle determination unit 4 respectively. The adjustment unit 5 is used to adjust the nacelle orientation of the wind turbine according to the current deviation angle, the Pareto optimal threshold and the Pareto optimal delay time.
[0146] Furthermore, the historical data determination unit 1 includes: a yaw control parameter determination module, a stress acquisition module, an initial load determination module, and a correction module.
[0147] The yaw control parameter determination module is used to determine multiple pairs of yaw control parameters using a grid search method; each pair of yaw control parameters includes a threshold and a delay time.
[0148] The stress acquisition module is connected to the yaw control parameter determination module. The stress acquisition module is used to acquire, for any pair of yaw control parameters, the initial power generation, generator torque stress time history curve, blade root bending moment stress time history curve, tower top thrust stress time history curve, and nacelle deviation angle within a set time period under the action of the yaw control parameters, wind speed environment, pitch angle position, and rated torque. The wind speed environment, pitch angle position, and rated torque are pre-applied to the wind turbine.
[0149] The initial load determination module is connected to the stress acquisition module. The initial load determination module is used to determine the initial generator torque equivalent fatigue load, initial blade root bending moment equivalent fatigue load and initial tower thrust equivalent fatigue load in the corresponding time period based on the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve and tower top thrust stress time history curve, using the rainflow counting method.
[0150] Specifically, the initial load determination module includes: a fatigue curve acquisition submodule, a first cycle statistics submodule, a first accumulation submodule, an initial generator torque load determination submodule, a second cycle statistics submodule, a second accumulation submodule, an initial blade root load determination submodule, a third cycle statistics submodule, a third accumulation submodule, and an initial tower load determination submodule.
[0151] The fatigue curve acquisition submodule is used to acquire the fatigue curve slopes of generator materials, blade root materials, and tower materials.
[0152] The first cycle statistics submodule is connected to the stress acquisition module. The first cycle statistics submodule is used to use the rainflow counting method to count the stress cycle of the generator torque according to the generator torque stress time history curve, and obtain the first statistical result.
[0153] The first accumulation submodule is connected to the first periodic statistics submodule. The first accumulation submodule is used to linearly superimpose the first statistical results based on the linear damage accumulation rule to obtain the generator cumulative rainflow spectrum.
[0154] The initial generator torque load determination submodule is connected to the fatigue curve acquisition submodule and the first accumulation submodule. The initial generator torque load determination submodule is used to determine the initial generator torque equivalent fatigue load in the corresponding time period based on the generator cumulative rainflow spectrum and the slope of the fatigue curve of the generator material.
[0155] The second period statistics submodule is connected to the stress acquisition module. The second period statistics submodule is used to use the rainflow counting method to count the stress cycle of the blade root according to the time history curve of the blade root bending moment stress, and obtain the second statistical result.
[0156] The second accumulation submodule is connected to the second periodic statistics submodule. The second accumulation submodule is used to linearly superimpose the second statistical results based on the linear damage accumulation rule to obtain the leaf root cumulative rainflow spectrum.
[0157] The initial blade root load determination submodule is connected to the fatigue curve acquisition submodule and the second accumulation submodule. The initial blade root load determination submodule is used to determine the initial blade root equivalent fatigue load in the corresponding time period based on the cumulative rainflow spectrum of the blade root and the slope of the fatigue curve of the blade root material.
[0158] The third cycle statistics submodule is connected to the stress acquisition module. The third cycle statistics submodule is used to use the rainflow counting method to count the stress cycle of the tower according to the time history curve of the thrust stress at the top of the tower, and obtain the third statistical result.
[0159] The third accumulation submodule is connected to the third periodic statistics submodule. The third accumulation submodule is used to linearly superimpose the third statistical results based on the linear damage accumulation rule to obtain the tower cumulative rainflow spectrum.
[0160] The initial tower load determination submodule is connected to the fatigue curve acquisition submodule and the third accumulation submodule. The initial tower load determination submodule is used to determine the initial equivalent fatigue load of the tower in the corresponding time period based on the cumulative rainflow spectrum of the tower and the slope of the fatigue curve of the tower material.
[0161] The correction module is connected to the initial power determination module of the stress acquisition module. The correction module is used to correct the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root moment equivalent fatigue load, and the initial tower thrust equivalent fatigue load based on the nacelle deviation angle and deviation loss coefficient, respectively, to obtain the final power generation, generator torque equivalent fatigue load, blade root moment equivalent fatigue load, and tower thrust equivalent fatigue load under the corresponding yaw control parameters. The deviation loss coefficient is determined in advance through dynamic simulation fitting.
[0162] Specifically, the correction module includes: a power correction submodule, a generator load correction submodule, a blade load correction submodule, and a tower load correction submodule.
[0163] The power correction submodule is connected to the stress acquisition module. It corrects the initial power generation based on the nacelle deviation angle and the power deviation loss coefficient to obtain the final power generation under the corresponding yaw control parameters. The power deviation loss coefficient is determined in advance through dynamic simulation fitting of historical power generation and historical nacelle deviation angles.
[0164] The generator load correction submodule is connected to the stress acquisition module and the initial load determination module. The generator load correction submodule is used to correct the initial generator torque equivalent fatigue load based on the nacelle deviation angle and the generator torque deviation loss coefficient, thereby obtaining the final generator torque equivalent fatigue load under the corresponding yaw control parameters. The generator deviation loss coefficient is determined in advance through dynamic simulation fitting of historical generator torque equivalent fatigue loads and historical nacelle deviation angles.
[0165] The blade root load correction submodule is connected to the stress acquisition module and the initial load determination module. This submodule corrects the initial blade root bending moment equivalent fatigue load based on the nacelle deviation angle and the blade root deviation loss coefficient, obtaining the final blade root bending moment equivalent fatigue load under the corresponding yaw control parameters. The blade deviation loss coefficient is determined in advance through dynamic simulation fitting of historical blade root bending moment equivalent fatigue loads and historical nacelle deviation angles.
[0166] The tower load correction submodule is connected to the stress acquisition module and the initial load determination module. The tower load correction submodule is used to correct the initial tower thrust equivalent fatigue load based on the nacelle deviation angle and the tower deviation loss coefficient, thus obtaining the final tower thrust equivalent fatigue load under the corresponding yaw control parameters. The tower deviation loss coefficient is determined in advance through dynamic simulation fitting of historical tower thrust equivalent fatigue loads and historical nacelle deviation angles.
[0167] Furthermore, the adjustment unit 5 includes: a filtering module, a first judgment module, a holding module, a timing module, a second judgment module, a third judgment module, and an adjustment module.
[0168] The filtering module is connected to the deviation angle determination unit 4. The filtering unit is used to filter the deviation angle to obtain the filtered deviation angle.
[0169] The first judgment module is connected to the filtering module. The first judgment module is used to determine whether the filtering deviation angle is greater than the Pareto optimal threshold.
[0170] The holding module is connected to the first judgment module. The holding module is used to not adjust the nacelle orientation of the wind turbine when the filter deviation value is less than or equal to the Pareto optimal threshold.
[0171] The timing module is connected to the first judgment module. The timing module is used to start timing when the filter deviation angle is greater than the Pareto optimal threshold.
[0172] The second judgment module is connected to the timing module. The second judgment module is used to determine whether the timing time is greater than the Pareto optimal delay time.
[0173] The holding module is also connected to the second judgment module. The holding module is also used to not adjust the nacelle orientation of the wind turbine or to stop adjusting the nacelle orientation of the wind turbine when the timing time is less than or equal to the Pareto optimal delay time.
[0174] The third judgment module is connected to the second judgment module. The third judgment module is used to determine whether the cabin bearing is equal to the set bearing value when the timing time is greater than the Pareto optimal delay time.
[0175] The holding module is also connected to the third judgment module. The holding module is also used to prevent or stop adjusting the nacelle orientation of the wind turbine when the nacelle orientation is equal to the set orientation value.
[0176] The adjustment module is connected to the third judgment module. The adjustment module is used to start adjusting the nacelle orientation of the wind turbine when the nacelle orientation is not equal to the set orientation value, or to continue adjusting the nacelle orientation of the wind turbine.
[0177] Compared with the prior art, the wind turbine yaw control system of the present invention that takes load calculation into account has the same beneficial effects as the wind turbine yaw control method that takes load calculation into account described above, and will not be repeated here.
[0178] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0179] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A wind turbine yaw control method taking load calculation into account, It is characterized in that The wind turbine yaw control method taking load calculation into account includes: A grid search method is used to determine multiple groups of historical yaw data; each group of historical yaw data includes yaw control parameters, equivalent fatigue load and power generation of the wind turbine during yaw control of the wind turbine under the action of the yaw control parameters; the yaw control parameters include a threshold value and a delay time; the threshold value represents the maximum value of the allowable deviation angle between the nacelle and the wind direction of the wind turbine; when the deviation angle between the nacelle and the wind direction exceeds the threshold value and lasts for a delay time, the yaw control of the wind turbine is started; the wind turbine includes a generator, a blade and a tower; the equivalent fatigue load of the wind turbine includes an equivalent fatigue load of generator torque, an equivalent fatigue load of blade root bending moment and an equivalent fatigue load of tower thrust; The grid search method is used to determine multiple groups of historical yaw data, specifically including: A grid search method is used to determine multiple pairs of yaw control parameters; each pair of yaw control parameters includes a threshold and a delay time; For any pair of yaw control parameters, under the action of the yaw control parameters, wind speed environment, pitch angle position and rated torque, the initial power generation within a set period, the generator torque stress time history curve, the blade root bending moment stress time history curve, the tower top thrust stress time history curve and the deviation angle of the nacelle are obtained; the wind speed environment, pitch angle position and rated torque are pre-applied to the wind turbine; According to the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve and tower top thrust stress time history curve, the rain flow counting method is used to determine the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load in the corresponding period; According to the deviation angle and deviation loss coefficient of the nacelle, the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load are respectively corrected to obtain the final power generation, generator torque equivalent fatigue load, blade root bending moment equivalent fatigue load and tower thrust equivalent fatigue load under the corresponding yaw control parameters; the deviation loss coefficient is determined in advance by dynamic simulation fitting; According to multiple groups of historical yaw data, with the minimum equivalent fatigue load and the maximum power generation as the optimization goals, the Pareto optimization method is used to optimize the threshold and delay time, and the Pareto optimal threshold and Pareto optimal delay time are obtained; Collect the current inflow wind direction and the nacelle azimuth of the wind turbine; Determine the current deviation angle according to the current inflow wind direction and the nacelle azimuth of the wind turbine generator set; The nacelle orientation of the wind turbine generator set is adjusted according to the current deviation angle, the Pareto optimal threshold and the Pareto optimal delay time.
2. The yaw control method of a wind turbine taking load calculation into account according to claim 1, It is characterized in that The method of using the rain flow counting method to determine the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load in the corresponding time period according to the generator material, blade material, tower material, generator torque stress time history curve, blade root bending moment stress time history curve and tower top thrust stress time history curve specifically includes: Obtain the fatigue curve slope of the generator material, the fatigue curve slope of the blade root material, and the fatigue curve slope of the tower material; Using a rain flow counting method, according to the generator torque stress time history curve, the stress cycle period of the generator torque is counted to obtain a first statistical result; Based on the linear damage accumulation rule, the first statistical result is linearly superimposed to obtain the generator cumulative rain flow spectrum; Determining the initial generator torque equivalent fatigue load in a corresponding period according to the generator cumulative rain flow spectrum and the fatigue curve slope of the generator material; Using a rain flow counting method, according to the blade root bending moment stress time history curve, the stress cycle period of the blade root is counted to obtain a second statistical result; Based on the linear damage accumulation rule, the second statistical result is linearly superimposed to obtain the blade root cumulative rain flow spectrum; Determining an initial blade root equivalent fatigue load in a corresponding period according to the blade root cumulative rain flow spectrum and the fatigue curve slope of the blade root material; Using the rain flow counting method, according to the tower top thrust stress time history curve, the stress cycle period of the tower is counted to obtain a third statistical result; Based on the linear damage accumulation rule, the third statistical result is linearly superimposed to obtain the tower cumulative rain flow spectrum; The initial tower equivalent fatigue load in the corresponding time period is determined according to the tower cumulative rain flow spectrum and the slope of the fatigue curve of the tower material.
3. The method for yaw control of a wind turbine taking load calculation into account according to claim 2, It is characterized in that The initial generator torque equivalent fatigue load is calculated using the following formula: Among them, R DEL is the equivalent fatigue load of the initial generator torque, R i is the load amplitude in the i-th stress cycle in the generator cumulative rain flow spectrum, n i is the cumulative number of load cycles in the i-th stress cycle period in the generator cumulative rain flow spectrum, i>0, and m is the slope of the fatigue curve of the generator material.
4. The yaw control method of a wind turbine generator system taking load calculation into account according to claim 1, It is characterized in that The deviation loss coefficients include: power deviation loss coefficient, generator torque deviation loss coefficient, blade root deviation loss coefficient and tower deviation loss coefficient; According to the deviation angle and deviation loss coefficient of the nacelle, the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load are respectively corrected to obtain the final power generation, generator torque equivalent fatigue load, blade root bending moment equivalent fatigue load and tower thrust equivalent fatigue load under the corresponding yaw control parameters, specifically including: According to the deviation angle of the nacelle and the power deviation loss coefficient, the initial power generation is corrected to obtain the final power generation under the corresponding yaw control parameters; the power deviation loss coefficient is determined by dynamically simulating and fitting the historical power generation and the historical deviation angle of the nacelle in advance; According to the deviation angle of the nacelle and the generator torque deviation loss coefficient, the initial generator torque equivalent fatigue load is corrected to obtain the final generator torque equivalent fatigue load under the corresponding yaw control parameters; the generator torque deviation loss coefficient is determined by dynamically simulating and fitting the historical generator torque equivalent fatigue load and the historical deviation angle of the nacelle in advance; According to the deviation angle of the nacelle and the blade root deviation loss coefficient, the initial blade root bending moment equivalent fatigue load is corrected to obtain the final blade root bending moment equivalent fatigue load under the corresponding yaw control parameters; the blade root deviation loss coefficient is determined by dynamically simulating and fitting the historical blade root bending moment equivalent fatigue load and the historical deviation angle of the nacelle in advance; According to the deviation angle of the nacelle and the tower deviation loss coefficient, the initial tower thrust equivalent fatigue load is corrected to obtain the final tower thrust equivalent fatigue load under the corresponding yaw control parameters; the tower deviation loss coefficient is determined in advance by dynamically simulating and fitting the historical tower thrust equivalent fatigue load and the historical deviation angle of the nacelle.
5. The yaw control method of a wind turbine generator system taking load calculation into account according to claim 4, It is characterized in that Use the following formula to determine the final power generation: Among them, T p is the final power generation, T p,0 is the initial power generation, f p (θ) is the power deviation loss coefficient, ω is the first harmonic, θ is the deviation angle of the nacelle, a p,0 、a p,1 、a p,2 、b p,1 、b p,2 are all undetermined coefficients, a p,0 、a p,1 、a p,2 、b p,1 、b p,2 Determined by dynamic simulation fitting.
6. The yaw control method of a wind turbine generator system taking load calculation into account according to claim 1, It is characterized in that The adjusting the nacelle orientation of the wind turbine generator set according to the current deviation angle, the Pareto optimal threshold and the Pareto optimal delay time specifically includes: Performing filtering on the current deviation angle to obtain a filtered deviation angle; Determining whether the filtering deviation angle is greater than the Pareto optimal threshold; If the filtering deviation angle is less than or equal to the Pareto optimal threshold, the nacelle orientation of the wind turbine generator set is not adjusted; If the filtering deviation angle is greater than the Pareto optimal threshold, start timing; Determine whether the timing time is greater than the Pareto optimal delay time; If the timing time is less than or equal to the Pareto optimal delay time, the nacelle orientation of the wind turbine generator set is not adjusted or the adjustment of the nacelle orientation of the wind turbine generator set is stopped; If the timing time is greater than the Pareto optimal delay time, it is determined whether the cabin orientation is equal to the set orientation value. If the cabin orientation is equal to the set orientation value, the cabin orientation of the wind turbine generator set is not adjusted or the adjustment of the cabin orientation of the wind turbine generator set is stopped. If the cabin orientation is not equal to the set orientation value, the cabin orientation of the wind turbine generator set is started to be adjusted, or the cabin orientation of the wind turbine generator set is continued to be adjusted.
7. A wind turbine yaw control system taking load calculation into account, It is characterized in that The wind turbine yaw control system taking load calculation into account includes: A historical data determination unit is used to determine multiple groups of historical yaw data using a grid search method; each group of historical yaw data includes a yaw control parameter, an equivalent fatigue load of the wind turbine during yaw control of the wind turbine under the action of the yaw control parameter, and a generated power; the yaw control parameter includes a threshold value and a delay time; the threshold value represents the maximum value of the allowable deviation angle between the nacelle and the wind direction of the wind turbine; when the deviation angle between the nacelle and the wind direction exceeds the threshold value and lasts for a delay time, the yaw control of the wind turbine is started; the wind turbine includes a generator, a blade, and a tower; the equivalent fatigue load of the wind turbine includes an equivalent fatigue load of generator torque, an equivalent fatigue load of blade root bending moment, and an equivalent fatigue load of tower thrust; The historical data determining unit comprises: A yaw control parameter determination module is used to determine multiple pairs of yaw control parameters using a grid search method; each pair of yaw control parameters includes a threshold value and a delay time; A stress acquisition module is connected to the yaw control parameter determination module and is used to obtain the initial power generation, generator torque stress time history curve, blade root bending moment stress time history curve, tower top thrust stress time history curve and the deviation angle of the nacelle within a set period for any pair of yaw control parameters under the action of the yaw control parameters, wind speed environment, pitch angle position and rated torque; the wind speed environment, pitch angle position and rated torque are pre-applied to the wind turbine; an initial load determination module, connected to the stress acquisition module, for determining the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load in the corresponding time period by using a rain flow counting method according to the generator material, the blade material, the tower material, the generator torque stress time history curve, the blade root bending moment stress time history curve and the tower top thrust stress time history curve; a correction module, connected to the stress acquisition module and the initial load determination module respectively, for respectively correcting the initial power generation, the initial generator torque equivalent fatigue load, the initial blade root bending moment equivalent fatigue load and the initial tower thrust equivalent fatigue load according to the deviation angle and the deviation loss coefficient of the nacelle, so as to obtain the final power generation, generator torque equivalent fatigue load, blade root bending moment equivalent fatigue load and tower thrust equivalent fatigue load under the corresponding yaw control parameters; the deviation loss coefficient is determined by dynamic simulation fitting in advance; an optimization unit connected to the historical data determination unit, for optimizing the threshold and the delay time by adopting a Pareto optimization method according to multiple groups of historical yaw data, with the minimum equivalent fatigue load and the maximum power generation as the optimization target, to obtain a Pareto optimal threshold and a Pareto optimal delay time; Azimuth angle acquisition unit, used to collect the current inflow wind direction and the nacelle azimuth of the wind turbine; a deviation angle determination unit, connected to the azimuth angle acquisition unit, and configured to determine a current deviation angle according to the current inflow wind direction and the nacelle azimuth angle of the wind turbine generator set; The adjustment unit is connected to the optimization unit and the deviation angle determination unit respectively, and is used to adjust the nacelle orientation of the wind turbine according to the current deviation angle, the Pareto optimal threshold and the Pareto optimal delay time.
8. The wind turbine yaw control system taking load calculation into account according to claim 7, It is characterized in that The initial load determination module comprises: A fatigue curve acquisition submodule is used to obtain the fatigue curve slope of the generator material, the fatigue curve slope of the blade root material, and the fatigue curve slope of the tower material; A first cycle statistics submodule is connected to the stress acquisition module and is used to use a rain flow counting method to count the stress cycle period of the generator torque according to the generator torque stress time history curve to obtain a first statistical result; A first accumulation submodule, connected to the first period statistics submodule, is used to linearly superimpose the first statistical result based on a linear damage accumulation rule to obtain a generator accumulated rain flow spectrum; An initial generator torque load determination submodule, connected to the fatigue curve acquisition submodule and the first accumulation submodule, for determining the initial generator torque equivalent fatigue load in a corresponding period according to the generator cumulative rain flow spectrum and the fatigue curve slope of the generator material; A second cycle statistics submodule is connected to the stress acquisition module and is used to use a rain flow counting method to count the stress cycle period of the blade root according to the blade root bending moment stress time history curve to obtain a second statistical result; A second accumulation submodule, connected to the second period statistics submodule, is used to linearly superimpose the second statistical results based on a linear damage accumulation rule to obtain a blade root accumulated rain flow spectrum; an initial blade root load determination submodule, connected to the fatigue curve acquisition submodule and the second accumulation submodule, for determining the initial blade root equivalent fatigue load in a corresponding period according to the blade root accumulated rain flow spectrum and the fatigue curve slope of the blade root material; A third cycle statistics submodule is connected to the stress acquisition module and is used to use a rain flow counting method to count the stress cycle period of the tower according to the tower top thrust stress time history curve to obtain a third statistical result; A third accumulation submodule is connected to the third period statistics submodule and is used to linearly superimpose the third statistical results based on a linear damage accumulation rule to obtain a tower accumulated rain flow spectrum; The initial tower load determination submodule is connected to the fatigue curve acquisition submodule and the third accumulation submodule, and is used to determine the initial tower equivalent fatigue load in the corresponding time period according to the tower accumulated rain flow spectrum and the fatigue curve slope of the tower material.
