Fan reverse reconstruction technology based on tower bottom load
By reconstructing the wind turbine model using tower base load data, the problem of low wind turbine design efficiency under commercial confidentiality restrictions was solved, enabling efficient and independent wind turbine design and improving design efficiency and system reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-03
- Publication Date
- 2026-04-03
AI Technical Summary
Due to commercial confidentiality restrictions, the floating structure designers were unable to obtain detailed parameters of the upper wind turbine, which led them to rely on the wind turbine manufacturer for load calculations that took up to a month, severely restricting the design efficiency and iteration speed of the floating wind turbine.
By collecting the time history data of the tower base load and the tower cross-sectional parameters, a functional relationship between the tower base load and the wind speed is established. A continuous cross-sectional function is constructed using linear interpolation. Combining the relationship between the wind turbine's power generation, speed and torque, a wind turbine power curve is established. Through parameter sensitivity analysis, the wind turbine parameters are optimized. Finally, an overall model of the upper part of the floating wind turbine is established, realizing the reverse reconstruction of the wind turbine.
It significantly improves the design efficiency of floating wind turbines, shortens the design cycle, reduces communication and data acquisition costs, enhances the confidentiality and autonomous controllability of the design process, supports multiple rounds of iterative simulation and optimization, and improves system adaptability and reliability.
Smart Images
Figure CN121787044A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of time-domain integrated analysis technology for floating wind turbines, and more specifically, to a wind turbine reverse reconstruction technology based on tower base load. Background Technology
[0002] With the increasing maturity of technology, the development of offshore wind farms towards "deep water" and "offshore" has become a major trend. Floating wind turbines are rigid-flexible coupled systems consisting of a lower floating platform and an upper wind turbine. The design process requires full consideration of the influence of the upper structure on the floating body's response. Due to the commercial confidentiality of both the floating body designer and the wind turbine manufacturer, the floating body designer cannot establish a complete integrated coupled analysis model for accurate time-domain analysis. Typically, the floating body designer provides the wind turbine manufacturer with hydrodynamic information and a simplified mooring model, allowing the wind turbine manufacturer to complete the integrated modeling. The wind turbine manufacturer then conducts extensive design condition calculations and provides the floating body designer with information such as tower base loads. However, each load calculation by the wind turbine manufacturer takes about a month.
[0003] Therefore, a wind turbine reverse reconfiguration technology based on tower base load is provided. Summary of the Invention
[0004] The purpose of this invention is to provide a wind turbine reverse reconfiguration technology based on tower base load, in order to solve the problem mentioned in the background art that floating body designers cannot obtain detailed parameters of the upper wind turbine due to commercial confidentiality restrictions, resulting in reliance on wind turbine manufacturers for load calculations that take up to a month, which seriously restricts the design efficiency and iteration speed of floating wind turbines.
[0005] To achieve the above objectives, the present invention aims to provide a wind turbine reverse reconfiguration technology based on tower bottom load, comprising the following steps: S1. Collect the time history data of the tower base load and the tower section parameters, establish the functional relationship between the tower base load and the wind speed, and extract the wind turbine speed function at each wind speed; S2. Based on the discrete data of the inner and outer diameters of the tower along the height direction, a continuous cross-sectional function is constructed using linear interpolation, and an equivalent model of the tower is established. S3. Combining the relationship between wind turbine power generation, speed and torque, establish the wind turbine power curve using the aforementioned load and speed data, and construct the objective function for reconstructing the wind turbine with power, speed and tower top load as output variables. S4. Select an open-source wind turbine as the initial model, calculate the impeller performance based on blade element momentum theory, and optimize the wind turbine parameters by establishing an influence function through parameter sensitivity analysis. S5. Based on the optimized wind turbine parameters and the equivalent tower model, establish an overall model of the upper part of the floating wind turbine, calculate the load response at the bottom of the tower and compare it with the original data to realize the reverse reconstruction of the wind turbine.
[0006] As a further improvement to this technical solution, in step S1, the functional relationship between the tower base load and wind speed is established, and the wind turbine speed function at each wind speed is extracted, including the following steps: S1.1 Collect load time history data at the bottom of the wind turbine tower, and at the same time, obtain the cross-sectional geometric parameters of the tower along the height direction; S1.2 Perform statistical analysis and spectral decomposition on the collected tower bottom load time history data, and decompose the load components into the main gravity load component and the wind load induced component according to the frequency domain characteristics. S1.3, based on measured wind speed sequence Using the tower base load as the input variable, the relationship between the tower base load and wind speed is fitted to establish a functional relationship between the tower base load and wind speed. S1.4 Perform spectral analysis on the horizontal force component at the bottom of the tower to identify the main characteristic frequency peaks; S1.5 Extract the fan speed function based on the spectral analysis results.
[0007] As a further improvement to this technical solution, in step S1.5, the fan speed function is extracted based on the spectral analysis results, involving the following specific steps: Based on the correspondence between the frequency and rotational speed of the wind turbine blades, the wind turbine rotational speed under various operating conditions is calculated; the load signal is bandpass filtered, and the 1p frequency is obtained by peak detection and time window averaging; the rotational speed data calculated under different wind speeds are fitted, and the least squares method is used to construct the functional relationship between the wind turbine rotational speed and the wind speed.
[0008] As a further improvement to this technical solution, step S2 involves constructing a continuous cross-sectional function using linear interpolation and establishing an equivalent tower model, including the following steps: S2.1 Obtain the set of discrete cross-sectional parameters of the tower along the height direction; S2.2 For the height measurement points in the discrete section parameter set, linear interpolation is used to construct the tower outer diameter function respectively. With inner diameter function ; S2.3, Based on the tower outer diameter function With inner diameter function Calculate the key cross-sectional mechanical parameters of the tower at any height; S2.4. Based on the Euler-Bernoulli beam theory, establish an equivalent dynamic model of the tower.
[0009] As a further improvement to this technical solution, in S2.4, based on the Euler-Bernoulli beam theory, an equivalent dynamic model of the tower is established, involving the following specific steps: The tower is treated as a flexible beam structure with continuously changing geometric and mechanical properties along the height direction. Based on the mechanical parameters of key sections, the Euler-Bernoulli beam theory is used to describe its stress and deformation relationship. Combined with the external load distribution and boundary conditions, the dynamic equilibrium equation of the tower under lateral vibration is established to express the coupling relationship between structural stiffness, mass and external load.
[0010] As a further improvement to this technical solution, in step S3, a wind turbine power curve is established using the aforementioned load and speed data, and a target function for reconstructing the wind turbine is constructed using power, speed, and tower top load as output variables, including the following steps: S3.1 Based on the principle of aerodynamic energy conversion of wind turbines, establish the basic relationship function between power generation and speed and torque; S3.2. Based on the fundamental relational function analysis, analyze the characteristics of torque component variation with wind speed and construct a torque function. ; S3.3, The fan speed function obtained in step S1 With torque function Substitute the power relationship into the formula to calculate the theoretical power generation at different wind speeds; S3.4 Determine the multi-objective output variables of the reconfigurable fan; S3.5. Using power, speed, and tower top load as output targets, establish the objective function for wind turbine reverse reconstruction, and verify and normalize the objective function.
[0011] As a further improvement to this technical solution, in step S3.2, the characteristics of the torque component changing with wind speed are analyzed based on the basic relational function, and a torque function is constructed. The specific steps involved are as follows: Extracting the tower top bending moment. The characteristics of wind speed variation are analyzed, including its amplitude and trend, and the bending moment at the top of the tower is considered. With spindle torque The linear approximation relationship between them is established by introducing a correction coefficient. The tower top bending moment is mapped to the principal axis torque; By applying this linear mapping to the tower top bending moment data under different wind speeds, a sequence of principal shaft torques corresponding to wind speeds is obtained, and an interpolation method is used to construct the torque function. .
[0012] As a further improvement to this technical solution, in step S4, the impeller performance is calculated based on the blade element momentum theory, and the fan parameters are optimized by establishing an influence function through parameter sensitivity analysis, including the following steps: S4.1 Select an open-source reference wind turbine with complete aerodynamic parameters and structural information as the initial model for reverse reconstruction, and extract the set of key design parameters from the initial model; S4.2 Based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculate the aerodynamic load and power distribution at each blade element location; S4.3. Using the BEM model, obtain the thrust-torque-power curves of the wind turbine under various wind speed conditions, and combine them with the rotational speed at each wind speed point in the objective function constructed in step S3. With power Establish the correspondence between the simulated output and the target value; S4.4 Apply small perturbations to key aerodynamic and structural parameters respectively. The change in the target variable is calculated and output; among which, the key aerodynamic and structural parameters include at least blade length, chord length distribution, twist angle distribution, and hub radius; S4.5 Summarize the sensitivity of each parameter to the multi-target output to form an influence function matrix; S4.6. Based on the objective function of reconstructing the wind turbine, the wind turbine parameters are iteratively optimized using an optimization iterative strategy based on gradient descent. S4.7. Determine the stability of the optimized fan parameters.
[0013] As a further improvement to this technical solution, in step S4.2, based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculating the aerodynamic load and power distribution at each blade element location involves the following specific steps: dividing the impeller into several blade element intervals along the radial direction. At each leaf element position, based on the local chord length Twist Number of leaves and free-flowing wind speed Calculate local realism and inflow angle Combined with local thrust coefficient With power coefficient The axial induction factor is solved iteratively. With tangential inducing factor The equations are used to calculate the thrust and power of leaf element micro-elements based on the distribution of induction factors.
[0014] As a further improvement to this technical solution, in step S5, based on the optimized wind turbine parameters and the equivalent tower model, an overall model of the upper part of the floating wind turbine is established, the load response at the bottom of the tower is calculated and compared with the original data, including the following steps: S5.1. Combine the reconstructed wind turbine parameter vector obtained after optimization in S4 with the tower continuous section function established in step S2. Mass distribution Together they serve as input data for constructing the overall model above; S5.2 Establish an upper model that includes impeller-nacelle rigid body dynamics and local aerodynamic bending-torsional coupling of blades; S5.3 Couple the upper model with the tower equivalent model obtained in step S2 through the tower top node; S5.4, at the selected time step and simulation duration Next, time-domain coupled dynamic simulation is performed to calculate the load response at the bottom of the tower and compare it with the original data in order to realize the reverse reconstruction of the wind turbine; S5.5. Compare and extract features from the simulated tower bottom load time history data with the original input tower bottom load time history data to verify and determine the convergence of the reverse reconstruction results.
[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention relates to a wind turbine reverse reconfiguration technology based on tower base load, which significantly improves the design efficiency and economy of floating wind turbine systems. Compared to the traditional design process that relies on wind turbine manufacturers to provide a large amount of calculated operating conditions and load transfer data, this technology only needs to obtain partial tower base load time history data and tower geometric parameters to quickly reconstruct the aerodynamic and structural characteristics of the upper wind turbine, greatly shortening the design cycle and reducing communication and data acquisition costs. Simultaneously, this method effectively avoids frequent exchanges of sensitive design parameters between the floating body designer and the wind turbine manufacturer, reducing the risk of leakage of commercial secrets such as platform structure information, enhancing the confidentiality and independent controllability of the design process, thereby achieving efficient collaborative design while protecting the commercial interests of all parties.
[0016] 2. This invention relates to a wind turbine reverse reconfiguration technology based on tower base load. Through reconfiguration, the power curve, speed characteristics, tower top load response, and key aerodynamic parameters of the wind turbine can be obtained, compensating for the information blind spots caused by the lack of original data. This supports design units in conducting multiple rounds of iterative simulation and optimization under different environmental conditions and operating conditions, significantly improving the adaptability and reliability of the overall system. More importantly, design units can independently complete the modeling and performance evaluation of the upper unit without completely relying on the technical support of the wind turbine manufacturer. This enhances design autonomy and flexibility, reduces technical dependence on external suppliers, and provides strong technical support for the independent and intelligent development of offshore wind power in my country, especially deep-sea floating wind power. Attached Figure Description
[0017] Figure 1 This is a flowchart illustrating the overall method of the present invention; Figure 2 This embodiment reconstructs the tower top loads of the wind turbine and the target wind turbine. Spectral analysis comparison chart; Figure 3 This embodiment reconstructs the tower top loads of the wind turbine and the target wind turbine. Spectral analysis comparison chart; Figure 4 This embodiment reconstructs the tower top loads of the wind turbine and the target wind turbine. Amplitude comparison chart; Figure 5 This embodiment reconstructs the tower top loads of the wind turbine and the target wind turbine. Amplitude comparison chart; Figure 6 This embodiment reconstructs the tower top loads of the wind turbine and the target wind turbine. Time-based comparison chart. Detailed Implementation
[0018] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0019] Example: Please refer to Figure 1-6 As shown, this embodiment provides a wind turbine reverse reconfiguration technology based on tower base load, including the following steps: S1. Collect the time history data of the tower base load and the tower section parameters, establish the functional relationship between the tower base load and the wind speed, and extract the wind turbine speed function at each wind speed; In this embodiment, the functional relationship between the tower base load and wind speed is established, and the wind turbine speed function at each wind speed is extracted, including the following steps: S1.1, Collect load time history data at the bottom of the wind turbine tower, including six degrees of freedom components ( , , , , , Simultaneously, the cross-sectional geometric parameters of the tower along the height direction are obtained, including the inner diameter at each discrete height point. , outer diameter The wall thickness and material density provide input conditions for the subsequent establishment of the tower equivalent model; S1.2 Statistical analysis and spectral decomposition are performed on the collected tower base load time history data, and the load components are decomposed into the main gravity load component and the wind-induced load component according to the frequency domain characteristics; among which, the vertical load component Primarily caused by the self-weight of the wind turbine and tower, it can be used to reverse the total mass of the superstructure; horizontal load component , Mainly affected by wind speed and aerodynamic forces, it can reflect the characteristics of wind load distribution and action height. Through power spectral density analysis, the frequency energy distribution characteristics of different load components can be identified, providing a basis for establishing a coupling model of load and wind speed. S1.3, based on measured wind speed sequence Using the tower base load as the input variable, a functional relationship between the tower base load and wind speed is established by fitting the relationship between the two variables: ,in For load components, The load function corresponding to the wind speed is constructed by least squares fitting or spline interpolation, which yields a continuous expression for the load at the bottom of the tower as a function of wind speed. S1.4, Regarding the horizontal force component at the bottom of the tower ( Spectral analysis is performed to identify the main characteristic frequency peaks, including harmonic components such as 1p, 3p, and 6p. Specifically, the acquired horizontal force time-history signal is preprocessed, including removing DC components and low-frequency structural vibration noise, and bandpass filtering can be used to retain high-frequency components related to blade rotation. Then, the time-domain signal is converted to the frequency domain using Fast Fourier Transform (FFT) or Power Spectral Density (PSD) methods, and the spectral amplitude distribution is calculated. Next, the main peaks in the spectrum are identified, including harmonic components such as 1p (passing frequency of a single blade), 3p, and 6p, and the frequency and amplitude of each peak are extracted using a peak detection algorithm. Finally, the extracted characteristic frequencies are averaged or smoothed using a time window to obtain stable main frequency information that can be used to back-calculate the wind turbine speed, providing a basis for establishing the wind speed-rotation function. Among them, the 1p peak corresponds to the passing frequency of a single blade, which often overlaps with the low-frequency excitation peak. The 3p and 6p peaks reflect the response characteristics of the wind turbine's aerodynamic excitation and rotational imbalance. By removing the low-frequency peaks, the effective high-frequency peak frequencies are extracted. S1.5 Extract the fan speed function based on the spectral analysis results; Among them, the relationship between the 1p peak frequency and the fan speed obtained from the spectral analysis results is as follows: ; In the formula, The peak frequency (Hz) is 1p. The fan speed (rpm); Based on the 1p frequency values corresponding to various wind speeds, a functional relationship between fan speed and wind speed is established: ; The extraction of the fan speed function based on the spectral analysis results involves the following specific steps: Based on the correspondence between the frequency and rotational speed of the fan blades Calculate the fan speed under various operating conditions. To eliminate low-frequency structural vibration and noise interference, the load signal was first bandpass filtered to retain the high-frequency components related to impeller rotation, and a stable 1p frequency was obtained through peak detection and time window averaging. The calculated rotational speed data under different wind speeds were then fitted, and the least squares method was used to construct a functional relationship between the fan speed and wind speed. This yields a continuous rotational speed function describing the wind turbine's variation with wind speed, providing key input parameters for subsequent power curve fitting and aerodynamic performance inversion. Simultaneously, the accuracy of the extracted rotational speed is verified using 3p and 6p frequency peaks: if the 3p frequency... And 6p frequency If the speed extraction is reliable, then the load signal extraction is reliable; otherwise, it is necessary to check whether the load signal is affected by noise.
[0020] S2. Based on the discrete data of the inner and outer diameters of the tower along the height direction, a continuous cross-sectional function is constructed using linear interpolation, and an equivalent model of the tower is established. In this embodiment, a continuous cross-sectional function is constructed using linear interpolation, and an equivalent model of the tower is established, including the following steps: S2.1 Based on the design input or measured data, obtain the discrete cross-sectional parameter set of the tower along the height direction, including the height of each measuring point. Corresponding outer diameter , inner diameter Wall thickness and material density The collected raw data undergoes consistency checks and outlier removal to ensure that the cross-sectional dimensions change monotonically in the height direction and meet the requirements for the continuity of the tower structure. S2.2, Height measurement points for discrete cross-section parameter sets , The outer diameter function of the tower was constructed using linear interpolation. With inner diameter function : ; ; in, Interpolation can be used to obtain the outer and inner diameters of the tower along the height. The continuously varying function allows the tower's geometric parameters to be calculated continuously throughout the entire structural height range; S2.3, Based on the tower outer diameter function With inner diameter function Calculate the key section mechanical parameters of the tower at any height, including: cross-sectional area : ; Moment of inertia : ; Local quality : ; These parameters can be used to describe the tower's local bending stiffness, shear performance, and mass distribution characteristics; S2.4. Based on Euler-Bernoulli beam theory or Timoshenko beam theory, establish an equivalent dynamic model of the tower: ; In the formula, The elastic modulus of the material. For the lateral displacement of the tower, For loads distributed per unit length; Boundary conditions are set according to the wind turbine type: for stationary wind turbines, the tower base is considered the fixed end, i.e. and For floating wind turbines, the connection between the tower base and the platform is considered as an elastic support, and the boundary conditions are as follows: and ,in and The rotational stiffness and translational stiffness of the platform are respectively provided by the floating hydrodynamic model; The establishment of an equivalent dynamic model for the tower, based on Euler-Bernoulli beam theory or Timoshenko beam theory, involves the following specific steps: The tower is treated as a flexible beam structure with continuously varying geometric and mechanical properties along its height. Based on the mechanical parameters of key sections, the Euler-Bernoulli beam theory is used to describe its stress-deformation relationship. Combined with the distribution of external loads and boundary conditions, the dynamic equilibrium equation of the tower under lateral vibration is established to express the coupling relationship between structural stiffness, mass and external load. Finally, by numerically solving this equation, the displacement and stress response of the tower under wind load and superstructure mass can be obtained, realizing the equivalent modeling and dynamic characteristic description of the overall mechanical behavior of the tower.
[0021] S3. Combining the relationship between wind turbine power generation, speed and torque, establish the wind turbine power curve using the aforementioned load and speed data, and construct the objective function for reconstructing the wind turbine with power, speed and tower top load as output variables. In this embodiment, a wind turbine power curve is established using the aforementioned load and speed data, and an objective function for reconstructing the wind turbine is constructed using power, speed, and tower top load as output variables, including the following steps: S3.1. Based on the principle of aerodynamic energy conversion of wind turbines, establish the basic relationship function between power generation and speed and torque: In the formula, For wind turbine power generation, Main spindle torque, Angular velocity ( (Represents rotational speed, in rpm). S3.2. Based on the fundamental relational function analysis, analyze the characteristics of torque component variation with wind speed and construct a torque function. ; Among them, the characteristics of torque component variation with wind speed are analyzed based on the fundamental relational function, and a torque function is constructed. The specific steps involved are as follows: Extracting the tower top bending moment. The characteristics of wind speed variation are analyzed, including its amplitude and trend, and the bending moment at the top of the tower is considered. With spindle torque The linear approximation relationship between them is established by introducing a correction coefficient. Map the tower top bending moment to the principal axis torque: ; By applying this linear mapping to the tower top bending moment data under different wind speeds, a sequence of principal shaft torques corresponding to wind speeds is obtained, and an interpolation method is used to construct the torque function. This reflects the torque variation law of the wind turbine throughout the entire operating wind speed range, providing input for power curve calculation and reverse reconstruction of the objective function; Among them, the correction coefficient The value is determined by the tower structure stiffness, nacelle mass distribution, and transmission system efficiency, specifically obtained through calibration experiments or finite element analysis, and typically ranges from 0.8 to 1.2. (Tower top bending moment) Through the bending moment at the base of the tower and tower height calculate: ,in The horizontal force at the base of the tower; S3.3, The fan speed function obtained in step S1 With torque function Substituting into the power relationship formula, calculate the theoretical power generation at different wind speeds: ; S3.4 Determine the multi-objective output variables of the reconfigurable wind turbine, including power generation. : Reflects wind energy conversion efficiency; rotational speed : Reflects aerodynamic balance characteristics; Tower top load : Reflects the structural transmission response; S3.5. Establish the objective function for the reverse reconfiguration of the wind turbine, with power, speed, and tower top load as the output targets. The objective function is then verified and normalized. Wherein, objective function for: ; In the formula, The design parameter vector for the wind turbine includes blade length, hub radius, chord length distribution, and torsion angle distribution. To use current wind turbine parameters The power generation obtained from the simulation calculation (function value, which can vary with wind speed). This refers to the measured or target power generation value (reference value). To use current wind turbine parameters The simulated fan speed, This refers to the measured or target rotational speed (reference value). To use current wind turbine parameters The simulated tower top load response (such as force or bending moment). This refers to the measured or target tower top load value (reference value). This is the weighting factor for power generation. The weighting coefficient for rotational speed. This is the weighting coefficient for the load at the top of the tower.
[0022] S4. Select an open-source wind turbine as the initial model, calculate the impeller performance based on blade element momentum theory (BEM), and optimize the wind turbine parameters by establishing an influence function through parameter sensitivity analysis; In this embodiment, the impeller performance is calculated based on the blade element momentum theory, and the fan parameters are optimized by establishing an influence function through parameter sensitivity analysis, including the following steps: S4.1 Select an open-source reference wind turbine with complete aerodynamic parameters and structural information as the initial model for reverse reconstruction, such as a standardized wind turbine model like NREL-5MW, DTU10MW, IEA10MW, or IEA15MW, and extract the key design parameter set from the initial model. ; in, : Blade length; Hub radius; : chord length distribution; : Twisted angular distribution; Number of leaves; S4.2 Based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculate the aerodynamic load and power distribution at each blade element location; Furthermore, based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculating the aerodynamic load and power distribution at each blade element location involves the following specific steps: dividing the impeller into several blade element intervals along the radial direction. At each leaf element position, based on the local chord length Twist Number of leaves and free-flowing wind speed Calculate local realism and inflow angle Combined with local thrust coefficient With power coefficient The axial induction factor is solved iteratively. With tangential inducing factor The equations are used to calculate the thrust and power of leaf element micro-element based on the distribution of induction factors; Among them, the micro-element thrust and power at the leaf element are respectively: ; ; In the formula, air density; For free-flowing wind speed; , These are the thrust coefficient and the power coefficient, respectively. Axial induction factor is solved iteratively. With tangential inducing factor The system of equations: ; In the formula, The normal force coefficient of the leaf element. The tangential force coefficient, The local inflow angle refers to the relative angle between the incoming air velocity and the direction of blade movement at the blade element position; it is affected by both the blade rotation speed and the axial wind speed, and is a key parameter for calculating local aerodynamic loads and induction factors. in, For the angle of entry, For local realism; the induction factor distribution of each blade element can be obtained through iterative solution, and then the overall thrust of the impeller can be obtained by integration. Torque and power : ; S4.3. Using the BEM model, obtain the thrust-torque-power curves of the wind turbine under various wind speed conditions, and combine them with the rotational speed at each wind speed point in the objective function constructed in step S3. With power Establish the correspondence between the simulation output and the target value: ; In the formula, The wind turbine output error vector is given by the wind speed. The difference between the simulated output and the target output is used to measure the deviation of the current parameters from the output. The output variables include three categories: power generation, rotational speed, and tower top load. To use the current parameters The multi-objective output vector obtained from the simulation includes power Rotation speed and tower top load , The target output vector includes measured or reference power, rotational speed, and tower top load. To use initial fan parameters wind speed The output vector obtained from the simulation, In order to be in wind speed The target output vector includes measured or reference power generation, rotational speed, and tower top load; S4.4 Apply small perturbations to key aerodynamic and structural parameters (such as blade length, chord length distribution, twist angle distribution, hub radius, etc.). ( (used for numerical calculation of sensitivity), and calculates the change in the output target variable: ; Sensitivity of the derivative form This reflects the degree of influence of each parameter on power, speed, and load; by calculating the sensitivity distribution of all parameters, the response strength of the wind turbine system to different design variables can be obtained. For the output vector For the Parameters The instantaneous rate of change, In the first Each parameter increases a small perturbation. Then, the simulated wind turbine output vector, For the current number The parameters are The output vector at that time; S4.5. Summarize the sensitivity of each parameter to the multi-objective output (power, speed, tower top load) to form an influence function matrix: ,matrix It describes the linear influence of each design parameter on the target variable and is the core mathematical foundation for achieving parameter optimization; S4.6. Based on the objective function of reconstructing the wind turbine, the wind turbine parameters are iteratively optimized using a gradient descent-based optimization iterative strategy: ; in, To learn the step size or convergence coefficient; the wind turbine parameters are corrected through multiple iterations to make the objective function... Minimization, i.e., achieving convergence in the reconfigurable wind turbine performance and load response, For the current iteration step Wind turbine design parameter vector, For the next iteration step Updated wind turbine design parameter vector; S4.7. The stability of the optimized wind turbine parameters is assessed, specifically the rate of change of parameters after iteration. Less than the set threshold, and the output error When the convergence is within the allowable range, the parameter optimization is considered complete. Smoothness constraints are applied to the final parameter set (especially the blade chord length and twist angle distribution) to ensure the continuity of the aerodynamic shape and the manufacturability of the structure.
[0023] S5. Based on the optimized wind turbine parameters and the equivalent model of the tower, establish an overall model of the upper part of the floating wind turbine, calculate the load response at the bottom of the tower and compare it with the original data to realize the reverse reconstruction of the wind turbine. In this embodiment, the wind turbine parameter vector optimized by S4 is... (Including blade length, hub radius, chord length distribution, and twist angle distribution) are used as inputs and integrated with the tower equivalent model; among them, the chord length distribution and twist angle distribution are ensured to be continuous along the blade radius through spline interpolation; Based on the optimized turbine parameters and equivalent tower model, an overall model of the upper part of the floating wind turbine is established, the load response at the tower base is calculated and compared with the original data, including the following steps: S5.1. Combine the reconstructed wind turbine parameter vector obtained after optimization in S4 with the tower continuous section function established in step S2. Mass distribution This data will be used as input for the construction of the overall upper model. At the same time, prepare the parameters of the engine room, main shaft, transmission system and mass inertia (if unknown, substitute reasonable engineering approximations or reference values and record the uncertainty). S5.2. Establish an upper model incorporating the rigid body dynamics of the impeller-nacelle (RNA) and the local aerodynamic bending-torsional coupling of the blades. The aerodynamic forces of the impeller are calculated using BEM (including induced corrections, dynamic stall corrections, etc.) to obtain the force distribution along the radius and accumulated as the forces / torques of the nacelle. The nacelle is considered a rigid body, and its mass and moment of inertia are calculated using reconstructed parameters and standard material density. When higher accuracy is required, the blade bending modes (if known or approximable) are introduced into the model to form a rigid-flexible coupling description. Specifically, the modal reduction method is used to represent the blade bending vibration as modal coordinates. With modal shape function Superposition: ,in Let be the modal order; the dynamic equation is written as: ,in , , These represent the modal mass, damping, and stiffness matrices, respectively. This form represents the projection of aerodynamic loads in modal space. It retains the main vibration characteristics while significantly reducing the system's degrees of freedom, making it easier for numerical simulation and control analysis. S5.3 Couple the upper model with the tower equivalent model obtained in step S2 through the tower top node; For floating platforms: Set the interface mechanical boundary between the tower base and the floating platform (mooring stiffness, 6-DOF hydrodynamic characteristics of the floating body, or a simplified mass / damping model); for reverse reconstruction that only concerns the load at the tower base, the floating body dynamics can be replaced by an equivalent stiffness / damping or time-domain hydrodynamic model; if necessary, introduce a mooring model coupled with the floating body kinematics for calculation. S5.4, at the selected time step and simulation duration Next, time-domain coupled dynamic simulation is performed to calculate the load response at the base of the tower and compare it with the original data to achieve reverse reconstruction of the wind turbine. Specifically, the optimized wind turbine parameters are first coupled with the equivalent model of the tower and the overall upper model (impeller-nacelle rigid dynamics and blade local aerodynamic bending-torsional coupling) to construct the time-domain dynamic equations of the wind turbine-tower system, including rigid body inertia, blade modal mass and damping, aerodynamic load projection, and tower structural response. Then, numerical integration methods (such as Newmark-β, Runge-Kutta, or other explicit / implicit time integration algorithms) are used to iteratively solve for displacement, velocity, acceleration, and corresponding loads, calculating the six-degree-of-freedom load components at the base of the tower at each time step. , , , , , During the simulation, the tower top displacement, impeller speed and power generation output are recorded simultaneously to ensure the integrity of the dynamic response. Finally, the calculated tower bottom load time history is compared step by step with the measured or target tower bottom load, and the average value, amplitude distribution, spectrum characteristics and key peak values are extracted. Through error analysis and convergence judgment, the verification and calibration of the wind turbine reverse reconstruction are completed. S5.5. The simulated tower bottom load time history data is compared with the original input tower bottom load time history data to extract key features, including average value, amplitude distribution, spectral characteristics, and main load components (horizontal force, vertical force, bending moment, etc.). This evaluates the dynamic response accuracy of the reconstructed wind turbine model, providing quantitative basis for reverse reconstruction verification, and is used to verify and determine the convergence of the reverse reconstruction results. In this embodiment, after obtaining the complete tower bottom load response, spectral analysis and amplitude time-domain comparison are performed on the key components of the tower top load, such as... Figure 2-6 As shown.
[0024] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.
Claims
1. A wind turbine reverse reconfiguration technology based on tower base load, characterized in that, Includes the following steps: S1. Collect the time history data of the tower base load and the tower section parameters, establish the functional relationship between the tower base load and the wind speed, and extract the wind turbine speed function at each wind speed; S2. Based on the discrete data of the inner and outer diameters of the tower along the height direction, a continuous cross-sectional function is constructed using linear interpolation, and an equivalent model of the tower is established. S3. Combining the relationship between wind turbine power generation, speed and torque, establish the wind turbine power curve using the aforementioned load and speed data, and construct the objective function for reconstructing the wind turbine with power, speed and tower top load as output variables. S4. Select an open-source wind turbine as the initial model, calculate the impeller performance based on blade element momentum theory, and optimize the wind turbine parameters by establishing an influence function through parameter sensitivity analysis. S5. Based on the optimized wind turbine parameters and the equivalent tower model, establish an overall model of the upper part of the floating wind turbine, calculate the load response at the bottom of the tower and compare it with the original data to realize the reverse reconstruction of the wind turbine.
2. The wind turbine reverse reconfiguration technology based on tower base load according to claim 1, characterized in that: In step S1, the functional relationship between the tower base load and wind speed is established, and the wind turbine speed function at each wind speed is extracted, including the following steps: S1.1 Collect load time history data at the bottom of the wind turbine tower, and at the same time, obtain the cross-sectional geometric parameters of the tower along the height direction; S1.2 Perform statistical analysis and spectral decomposition on the collected tower bottom load time history data, and decompose the load components into the main gravity load component and the wind load induced component according to the frequency domain characteristics. S1.3, based on measured wind speed sequence Using the tower base load as the input variable, the relationship between the tower base load and wind speed is fitted to establish a functional relationship between the tower base load and wind speed. S1.4 Perform spectral analysis on the horizontal force component at the bottom of the tower to identify the main characteristic frequency peaks; S1.5 Extract the fan speed function based on the spectral analysis results.
3. The wind turbine reverse reconfiguration technology based on tower base load according to claim 2, characterized in that: In step S1.5, the wind turbine speed function is extracted based on the spectral analysis results, which involves the following specific steps: Based on the correspondence between the frequency and rotational speed of the wind turbine blades, the wind turbine rotational speed under various operating conditions is calculated; the load signal is bandpass filtered, and the 1p frequency is obtained by peak detection and time window averaging; the rotational speed data calculated under different wind speeds are fitted, and the least squares method is used to construct the functional relationship between the wind turbine rotational speed and the wind speed.
4. The wind turbine reverse reconfiguration technology based on tower base load according to claim 1, characterized in that: In step S2, a continuous cross-sectional function is constructed using linear interpolation, and an equivalent tower model is established, including the following steps: S2.1 Obtain the set of discrete cross-sectional parameters of the tower along the height direction; S2.2 For the height measurement points in the discrete section parameter set, linear interpolation is used to construct the tower outer diameter function respectively. With inner diameter function ; S2.3, Based on the tower outer diameter function With inner diameter function Calculate the key cross-sectional mechanical parameters of the tower at any height; S2.
4. Based on the Euler-Bernoulli beam theory, establish an equivalent dynamic model of the tower.
5. The wind turbine reverse reconfiguration technology based on tower base load according to claim 4, characterized in that: In S2.4, based on the Euler-Bernoulli beam theory, an equivalent dynamic model of the tower is established, involving the following specific steps: The tower is treated as a flexible beam structure with continuously changing geometric and mechanical properties along the height direction. Based on the mechanical parameters of key sections, the Euler-Bernoulli beam theory is used to describe its stress and deformation relationship. Combined with the external load distribution and boundary conditions, the dynamic equilibrium equation of the tower under lateral vibration is established to express the coupling relationship between structural stiffness, mass and external load.
6. The wind turbine reverse reconfiguration technology based on tower base load according to claim 1, characterized in that: In step S3, a wind turbine power curve is established using the aforementioned load and speed data, and a target function for reconstructing the wind turbine is constructed using power, speed, and tower top load as output variables, including the following steps: S3.1 Based on the principle of aerodynamic energy conversion of wind turbines, establish the basic relationship function between power generation and speed and torque; S3.
2. Based on the fundamental relational function analysis, analyze the characteristics of torque component variation with wind speed and construct a torque function. ; S3.3, The fan speed function obtained in step S1 With torque function Substitute the power relationship into the formula to calculate the theoretical power generation at different wind speeds; S3.4 Determine the multi-objective output variables of the reconfigurable fan; S3.
5. Using power, speed, and tower top load as output targets, establish the objective function for the reverse reconstruction of the wind turbine, and verify and normalize the objective function.
7. The wind turbine reverse reconfiguration technology based on tower base load according to claim 6, characterized in that: In step S3.2, the characteristics of torque component variation with wind speed are analyzed based on the basic relational function, and a torque function is constructed. The specific steps involved are as follows: Extracting the tower top bending moment. The characteristics of wind speed variation are analyzed, including its amplitude and trend, and the bending moment at the top of the tower is considered. With spindle torque The linear approximation relationship between them is established by introducing a correction coefficient. The tower top bending moment is mapped to the principal axis torque; By applying this linear mapping to the tower top bending moment data under different wind speeds, a sequence of principal shaft torques corresponding to wind speeds is obtained, and an interpolation method is used to construct the torque function. .
8. The wind turbine reverse reconfiguration technology based on tower base load according to claim 1, characterized in that: In step S4, the impeller performance is calculated based on the blade element momentum theory, and the wind turbine parameters are optimized by establishing an influence function through parameter sensitivity analysis, including the following steps: S4.1 Select an open-source reference wind turbine with complete aerodynamic parameters and structural information as the initial model for reverse reconstruction, and extract the set of key design parameters from the initial model; S4.2 Based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculate the aerodynamic load and power distribution at each blade element location; S4.
3. Using the BEM model, obtain the thrust-torque-power curves of the wind turbine under various wind speed conditions, and combine them with the rotational speed at each wind speed point in the objective function constructed in step S3. With power Establish the correspondence between the simulated output and the target value; S4.4 Apply small perturbations to key aerodynamic and structural parameters respectively. The change in the target variable is calculated and output; among which, the key aerodynamic and structural parameters include at least blade length, chord length distribution, twist angle distribution, and hub radius; S4.5 Summarize the sensitivity of each parameter to the multi-target output to form an influence function matrix; S4.
6. Based on the objective function of reconstructing the wind turbine, the wind turbine parameters are iteratively optimized using an optimization iterative strategy based on gradient descent. S4.
7. Determine the stability of the optimized fan parameters.
9. The wind turbine reverse reconfiguration technology based on tower base load according to claim 8, characterized in that: In S4.2, based on the blade element momentum theory, the impeller is divided into several blade element intervals along the radial direction. Calculating the aerodynamic load and power distribution at each blade element location involves the following specific steps: dividing the impeller into several blade element intervals along the radial direction. At each leaf element position, based on the local chord length Twist Number of leaves and free-flowing wind speed Calculate local realism and inflow angle Combined with local thrust coefficient With power coefficient The axial induction factor is solved iteratively. With tangential inducing factor The equations are used to calculate the thrust and power of leaf element micro-elements based on the distribution of induction factors.
10. The wind turbine reverse reconfiguration technology based on tower base load according to claim 1, characterized in that: In step S5, based on the optimized turbine parameters and the equivalent tower model, an overall model of the upper part of the floating wind turbine is established, the load response at the bottom of the tower is calculated and compared with the original data, including the following steps: S5.
1. Combine the reconstructed wind turbine parameter vector obtained after optimization in S4 with the tower continuous section function established in step S2. Mass distribution Together they serve as input data for constructing the overall model above; S5.2 Establish an upper model that includes impeller-nacelle rigid body dynamics and local aerodynamic bending-torsional coupling of blades; S5.3 Couple the upper model with the tower equivalent model obtained in step S2 through the tower top node; S5.4, at the selected time step and simulation duration Next, time-domain coupled dynamic simulation is performed to calculate the load response at the bottom of the tower and compare it with the original data in order to realize the reverse reconstruction of the wind turbine; S5.
5. Compare and extract features from the simulated tower bottom load time history data with the original input tower bottom load time history data to verify and determine the convergence of the reverse reconstruction results.