Wind turbine blade aerodynamic characteristic optimization and power generation efficiency improvement system

Abstract

The invention discloses a wind turbine blade aerodynamic characteristic optimization and power generation efficiency improvement system, and relates to the technical field of wind turbine operation optimization. A wind speed analysis module establishes wind energy distribution models of different wind speed grades according to historical wind regime data and real-time wind speed changes, analyzes whether wind speed sudden change exists or not based on the wind energy distribution models, and sends the analyzed wind speed sudden change to a power generation module; the load analysis module introduces a power grid end load demand to form a load change model, whether load fluctuation exists or not is analyzed based on the load change model, and the collaborative optimization module takes wind speed abrupt change and the load fluctuation as input variables and collaboratively optimizes a real-time adjustment strategy of a paddle through an adaptive control algorithm. According to the lifting system, a wind speed and load double sensing mechanism is constructed, an intelligent optimization algorithm and a pneumatic real-time monitoring network are introduced, the self-adaptive capacity of a fan in a variable environment is remarkably improved, the real-time performance and the fineness of paddle control are enhanced, and the control efficiency is improved. And the wind energy conversion efficiency, the power generation output stability, the structural safety and the operation intelligence level are obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine operation optimization, and in particular to a system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency. Background Art

[0002] As a green and renewable energy source, wind energy is gaining more and more attention. Compared with traditional fossil energy, wind energy has significant advantages such as zero emissions and strong sustainability. It is an important part of the future energy system. The current mainstream development trend of wind turbines is "large-scale" and "high efficiency". Wind turbine blades are the core components of wind turbines. Their aerodynamic performance directly determines the wind energy capture efficiency and the power generation capacity of the wind turbine. Therefore, improving the aerodynamic characteristics of the blades has become a key link in improving the overall performance of the wind power system.

[0003] The existing technology has the following defects:

[0004] 1. Most traditional wind power control strategies fail to consider the synergy between wind changes and load fluctuations, making it difficult to achieve optimal strategy matching under complex operating conditions. For example, when wind speed increases and load decreases simultaneously, the system often experiences control logic conflicts or fuzzy regulation targets, reducing operational efficiency.

[0005] 2. The lifting system mainly adjusts the blades based on the preset wind speed range, lacks the ability to recognize and predict sudden changes in wind speed in real time, and generally uses wind speed as the dominant variable to adjust the speed and output power. It has weak dynamic perception and feedback response capabilities to changes in grid-side load demand. This not only easily causes local areas of the blades to exceed the critical angle of attack range, resulting in aerodynamic stall or buffeting, but also during peak power consumption or sudden increases in grid load, the wind turbine output cannot cooperate in a timely manner, resulting in delayed response and increased output fluctuations, affecting grid stability.

[0006] Based on this, the present invention proposes a system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency. By constructing a dual perception mechanism of wind speed and load, introducing an intelligent optimization algorithm and an aerodynamic real-time monitoring network, the adaptive ability of the wind turbine in a changing environment is significantly improved, and the real-time and precision of blade control are enhanced, thereby significantly improving the wind energy conversion efficiency, power generation output stability, structural safety and operational intelligence level. Summary of the Invention

[0007] The purpose of the present invention is to provide a system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency, so as to solve the shortcomings of the background technology.

[0008] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency, comprising a wind speed analysis module, a load analysis module, a collaborative optimization module, and a simulation application module;

[0009] Wind speed analysis module: This module establishes wind energy distribution models for different wind speed levels based on historical wind data and real-time wind speed changes, and analyzes whether there is a sudden change in wind speed based on the wind energy distribution model.

[0010] Load analysis module: Introduces grid-side load demand to form a load change model, and analyzes whether there is load fluctuation based on the load change model;

[0011] Collaborative Optimization Module: This module uses wind speed changes and load fluctuations as input variables and collaboratively optimizes the real-time blade adjustment strategy through an adaptive control algorithm.

[0012] Simulation application module: Combining the wind energy distribution model with the load variation model, CFD numerical simulation is used to generate multi-condition aerodynamic optimization solution applications.

[0013] In a preferred embodiment, the collaborative optimization module receives the output information from the wind speed analysis module and the load analysis module, and extracts the wind speed mutation index ΔV t and load fluctuation index Φ t ;

[0014] Input variable set X for the lifting system t :X t ={ΔV t ,Φ t ,β t ,α t ,ω t}, where X t is the variable set at the current time t, ΔV t is the wind speed mutation index at the current time t, Φ t is the load fluctuation index at the current time t, β t is the pitch angle at the current moment t, α t is the angle of attack at the current moment t, ω t is the speed at the current moment t;

[0015] The variable set is used as input variables and the fuzzy control algorithm is used to model the blade adjustment logic, including:

[0016] If the wind speed suddenly increases and the load rises, the pitch angle is increased and the angle of attack is reduced;

[0017] If the wind speed drops suddenly and the load decreases, the pitch angle is reduced and the angle of attack is increased;

[0018] The fuzzy control algorithm combines the actual input and outputs the corresponding control quantity U t ;

[0019] U t ={Δβ t ,Δα t,Δω t ,ΔCl}, where Δβ t is the pitch angle adjustment, Δα t is the angle of attack adjustment, Δω t is the spindle speed adjustment, ΔC l is the correction value of the local lift coefficient of the airfoil.

[0020] In a preferred embodiment, the data collected by the collaborative optimization module through the sensor will be converted into digital form and transmitted to the central controller to form a real-time aerodynamic distribution map F t :

[0021] F t ={σ t (x),p t (x),M t (x)}, where σ t (x) is the stress distribution at each position x on the blade, p t (x) is the instantaneous surface pressure at each monitoring point, M t (x) is the monitoring value of bending moment and torque;

[0022] The current control effect will be evaluated based on the difference between the real-time feedback value and the expected aerodynamic response. If the current blade response error exceeds the error threshold, the fuzzy rule parameters will be adjusted.

[0023] In a preferred embodiment, the collaborative optimization module evaluates the current control effect based on the difference between the real-time feedback value and the expected aerodynamic response, and the expression is: Where, e t is the current blade response error, is the ideal aerodynamic distribution under the current wind speed and load conditions, F t It is the real-time feedback value.

[0024] In a preferred embodiment, the simulation application module first receives data transmitted from the wind speed analysis module and the load analysis module, extracts the working conditions, performs combined mapping, and constructs a combination sample set of multiple working conditions;

[0025] Based on each working condition sample set, a CFD simulation model of the wind turbine blade is constructed, including:

[0026] 3D blade geometry modeling: import actual blade structure or topology optimized structure;

[0027] Flow field boundary condition setting: inlet wind speed boundary condition, take the given wind speed in the wind energy distribution model;

[0028] Outlet pressure boundary; no-slip boundary condition is used on the blade surface;

[0029] Under each operating condition, a CFD solver is used to perform unsteady aerodynamic simulations. The simulation results are used to evaluate the aerodynamic performance of the blade under each operating condition, including the following indicators: wind energy conversion rate per unit area, lift-to-drag ratio, local load gradient, and fatigue stress distribution.

[0030] By integrating the simulation results under multiple working conditions, the optimization output is established, including the working condition-adjustment strategy mapping table, airfoil adjustment suggestions, and working condition priority list.

[0031] In a preferred embodiment, the load analysis module obtains continuous grid load data from a grid monitoring platform or a dispatching center to form a time series;

[0032] After obtaining the original load data, the relative change amplitude of the load within a unit time is calculated to quantify the load fluctuation of the power grid, and the load extreme difference within the time period is statistically analyzed using a sliding window.

[0033] In a preferred embodiment, the load analysis module obtains continuous grid load data from a grid monitoring platform or a dispatching center to form a time series: Where G t Represents the actual grid load at the tth moment, t is the sampling time index, For the complete load sequence;

[0034] The formula for calculating the rate of change is: Where ΔG t Indicates the relative rate of change of the load at time t. If |ΔG t If the rate of change is greater than the preset threshold, it is considered that a sudden load change has occurred at that moment;

[0035] A sliding window is used to calculate the load range within a certain period of time. The calculation formula is as follows: t =max(G t-w+1 ,…,G t )-min(G t-w+1 ,…,G t ), where W t Indicates the load range with the current moment as the end point of the window, w is the width of the sliding window, if W t If the value is greater than the load range change rate threshold, it means that the load fluctuates violently during this period.

[0036] In a preferred embodiment, the wind speed analysis module extracts multi-time scale wind speed data including annual, seasonal and daily data from the wind farm historical database, continuously samples the wind speed and wind direction at the current moment through an anemometer installed on the top or leading edge of the nacelle, and synchronously maps the real-time data with the historical data on the time axis;

[0037] A wind energy distribution model is constructed based on real-time wind speed data and historical wind speed data. According to the wind speed time series data of the wind energy distribution model, the first-order difference, sliding window range analysis or weighted moving average method is used to determine whether there is a significant change in the current wind speed.

[0038] In a preferred embodiment, the wind energy distribution model processing logic is as follows: based on historical data, the operating environment is divided into multiple wind speed levels according to the wind speed, each wind speed level corresponds to a corresponding wind energy density interval, and the occurrence frequency, duration, daily distribution, etc. within each wind speed interval are statistically modeled to form a wind energy density function;

[0039] Combining the wind speed grade classification with the real-time wind speed changes, the current wind energy distribution trend in each wind speed range is dynamically updated to establish a wind energy distribution model.

[0040] In a preferred embodiment, forming the wind energy density function comprises the following steps:

[0041] Based on the collected historical wind speed data set V={v1,v2,...,v n}, where v i Indicates the wind speed value obtained for the i-th time, and divides the wind speed into several level intervals;

[0042] After the wind speed levels are divided, the probability of occurrence of the i-th wind speed level is obtained by dividing the number of samples in the wind speed data that fall into the i-th level by the total number of wind speed data samples;

[0043] For each wind speed level, the duration of the time period in a unit time period is counted, and the average duration is expressed as: Where, T i Wind speed level L i The average duration, t i,j The wind speed on the jth day is at wind speed level L i The total duration of M is the number of observation days;

[0044] The functional expression of wind energy density function is: Where E(v) represents the wind energy density per unit time when the wind speed is v, p is the air density, A is the swept area of the wind turbine, and v represents the wind speed.

[0045] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0046] The present invention establishes a wind energy distribution model for different wind speed levels based on historical wind data and real-time wind speed changes through a wind speed analysis module, analyzes whether there is a sudden change in wind speed based on the wind energy distribution model, introduces the load demand at the power grid end to form a load change model, and analyzes whether there is a load fluctuation based on the load change model. The collaborative optimization module uses the sudden change in wind speed and load fluctuation as input variables, and collaboratively optimizes the real-time adjustment strategy of the blades through an adaptive control algorithm. The simulation application module combines the wind energy distribution model with the load change model, and uses CFD numerical simulation to generate a multi-condition aerodynamic optimization solution application. This lifting system significantly improves the adaptability of the wind turbine in a changing environment by constructing a dual perception mechanism of wind speed and load, introducing an intelligent optimization algorithm and an aerodynamic real-time monitoring network, and strengthens the real-time and precision of blade control, so that the wind energy conversion efficiency, power output stability, structural safety and operation intelligence level are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0048] Figure 1 This is a system architecture diagram of the present invention.

[0049] Figure 2 This is a system timing diagram of the present invention.

[0050] Figure 3 This is the system mind map of the present invention.

[0051] Figure 4 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0053] Example 1: Please refer to Figure 1-Figure 3 As shown, the system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency described in this embodiment includes a wind speed analysis module, a load analysis module, a collaborative optimization module, and a simulation application module;

[0054] Wind speed analysis module: This module establishes wind energy distribution models for different wind speed levels based on historical wind data and real-time wind speed changes. It then analyzes whether there are sudden changes in wind speed based on the wind energy distribution model. The analysis results are sent to the collaborative optimization module, and the wind energy distribution model is sent to the simulation application module.

[0055] Load analysis module: This module introduces grid-side load demand to form a load change model. Based on the load change model, it analyzes whether there is load fluctuation, which includes sudden increases and decreases in load. The analysis results are sent to the collaborative optimization module, and the load change model is sent to the simulation application module.

[0056] Collaborative Optimization Module: This module uses sudden wind speed changes and load fluctuations as input variables to collaboratively optimize the real-time blade adjustment strategy (such as pitch angle, angle of attack, speed, and airfoil fine-tuning) through an adaptive control algorithm (such as a fuzzy control algorithm). It also provides real-time feedback on the aerodynamic load distribution through a distributed sensor network (strain gauges, pressure sensors, and fiber optic monitoring) deployed at key locations on the blades.

[0057] Simulation application module: Combining the wind energy distribution model with the load variation model, CFD numerical simulation is used to generate multi-condition aerodynamic optimization solution applications.

[0058] This application uses a wind speed analysis module to establish a wind energy distribution model for different wind speed levels based on historical wind data and real-time wind speed changes. Based on the wind energy distribution model, it analyzes whether there is a sudden change in wind speed. The load analysis module introduces the load demand at the power grid end to form a load change model. Based on the load change model, it analyzes whether there is a load fluctuation. The collaborative optimization module uses wind speed mutations and load fluctuations as input variables, and collaboratively optimizes the real-time adjustment strategy of the blades through an adaptive control algorithm. The simulation application module combines the wind energy distribution model with the load change model, and uses CFD numerical simulation to generate a multi-condition aerodynamic optimization solution application. This improvement system significantly improves the wind turbine's adaptability in a changing environment by constructing a dual-sensing mechanism for wind speed and load, introducing an intelligent optimization algorithm and an aerodynamic real-time monitoring network, and strengthens the real-time and precision of blade control, thereby significantly improving wind energy conversion efficiency, power output stability, structural safety, and operational intelligence.

[0059] See also Figure 4 As shown in the figure, the specific workflow of the promotion system is as follows:

[0060] The lifting system establishes wind energy distribution models for different wind speed levels based on historical wind condition data and real-time wind speed changes, analyzes whether there is a sudden change in wind speed based on the wind energy distribution model, introduces the load demand on the grid end to form a load change model, and analyzes whether there is a load fluctuation based on the load change model. Load fluctuation includes sudden increase / drop in load. The sudden change in wind speed and load fluctuation are used as input variables, and the real-time adjustment strategy of the blade (such as pitch angle, angle of attack, speed, and airfoil fine-tuning) is collaboratively optimized through an adaptive control algorithm (such as fuzzy control algorithm). The aerodynamic load distribution is fed back in real time through a distributed sensor network (strain gauges, pressure sensors, and optical fiber monitoring) deployed at key positions of the blade. Combined with the wind energy distribution model and the load change model, CFD numerical simulation is used to generate a multi-condition aerodynamic optimization solution application.

[0061] Example 2: The wind speed analysis module establishes wind energy distribution models for different wind speed levels based on historical wind data and real-time wind speed changes, and analyzes whether there is a sudden change in wind speed based on the wind energy distribution model. The analysis results are sent to the collaborative optimization module, and the wind energy distribution model is sent to the simulation application module.

[0062] The wind speed analysis module extracts multi-time-scale wind speed data, including annual, seasonal, and daily data, from the wind farm historical database, covering wind speed information at different heights (such as wind rotor height). The data is cleaned by denoising, removing outliers, and performing interpolation and completion to build a highly reliable data foundation.

[0063] Through high-precision anemometers (such as ultrasonic anemometers, lidar, etc.) installed on the top or leading edge of the cabin, the current wind speed and wind direction are continuously sampled at high frequency (such as 1Hz to 10Hz), and the real-time data and historical data are synchronously mapped on the timeline to ensure that the wind speed change trend can be accurately portrayed.

[0064] The processing logic of the wind energy distribution model is as follows: based on historical data, the operating environment is divided into multiple wind speed levels (such as 0-3m / s, 3-6m / s, 6-9m / s, etc.) according to the wind speed. Each level of wind speed corresponds to a certain wind energy density range, and the frequency of occurrence, duration, daily distribution, etc. in each level of wind speed range are statistically modeled to form a "wind energy density function". Combined with the wind speed level division and real-time wind speed changes, the current wind energy distribution trend in each wind speed range is dynamically updated to establish a wind energy distribution model. This model reflects the core characteristics of the wind conditions in the current period, such as volatility, stability and energy density.

[0065] First, the system collects historical wind speed data sets V={v1,v2,...,v n}, where v iDenote the wind speed value obtained in the $i$-th acquisition, and divide the wind speed into intervals. According to the performance parameters of the wind turbine such as the starting wind speed, rated wind speed, and cut-out wind speed, the wind speed is usually divided into several grade intervals. For example:

[0066] Low wind speed interval: $0 \lt v \leq 3\ m / s$, medium-low wind speed interval: $3 \lt v \leq 6\ m / s$, medium-high wind speed interval: $6 \lt v \leq 9\ m / s$, high wind speed interval: $9 \lt v \leq 12\ m / s$, extremely high wind speed interval: $v \gt 12\ m / s$. Each wind speed interval is defined as a wind speed grade $L$ i , which is used for subsequent statistical analysis.

[0067] After dividing the wind speed grades, count the occurrence frequency of each wind speed interval, that is, the proportion of the number of times the wind speed of this grade appears in the historical data to the total number of observations. Obtain the occurrence probability of the $i$-th wind speed grade by dividing the number of samples falling into the $i$-th grade in the wind speed data by the total number of samples in the wind speed data. This probability value reflects the dominant degree of this wind speed grade in the entire wind energy distribution and has important reference significance for the wind energy utilization potential.

[0068] For each wind speed grade, count the length of the time period during which it continuously appears within a unit time period (such as daily, hourly). For example, count the total time when the wind speed is continuously in the interval of $6 - 9\ m / s$ per day. Its average duration can be expressed as: In the formula, $T$ i is the average duration of the wind speed grade $L$ i , $t$ i,j is the total duration of the wind speed at the wind speed grade $L$ i on the $j$-th day, $M$ is the number of observation days. The average duration of the wind speed grade $L$ i is used to judge the stability and dispatchability of the wind energy resource. The longer the duration, the more suitable the wind speed of this grade is as the main operation interval.

[0069] Consider the time distribution characteristics of the wind speed, especially the distribution law within a 24-hour cycle. Count the wind speed per hour and calculate the distribution frequency of a certain grade of wind speed in different time periods. Construct the following time distribution vector: In the formula, $d$ i,k represents the number of times the $i$-th wind speed grade appears at the $k$-th hour, represents the intraday distribution vector of the wind speed grade $L$ i . The intraday distribution vector reflects the change trend of the wind speed under the circadian rhythm and can be used to simulate the intraday volatility of the wind energy supply.

[0070] The functional expression of the wind energy density function is: In the formula, $E(v)$ represents the wind energy density when the wind speed is $v$ per unit time, $p$ is the air density, approximately $1.225\ kg / m$ 3, A is the fan swept area, and v represents the wind speed.

[0071] Based on wind speed time series data, methods such as first-order difference, sliding window range analysis or weighted moving average are used to determine whether there is a significant change in the current wind speed. For example, when the wind speed change rate per unit time is greater than a set threshold (such as 2m / s / s), or the wind speed exceeds 3 times the standard deviation of the average value for several consecutive seconds, it is determined to be a "wind speed mutation."

[0072] Extract wind speed sequences within a certain time period from field sensors or historical records. The sampling interval can be per second or per minute, depending on the sensitivity requirements for wind speed changes. The first-order difference is the most direct method for detecting the speed of change in the time series, and is used to reveal the increase or decrease in wind speed between adjacent time points. The calculation formula is as follows:

[0073] Δv t =v t -v t-1 , where Δv t Indicates the difference between the current wind speed and the previous wind speed, v t represents the wind speed value at the tth time point, represents the vth t-1 The wind speed value at time point t-1, if |Δv t If the difference exceeds the preset threshold, it is determined that the wind speed has changed significantly at that moment. The first-order difference is more sensitive to high-frequency wind speed disturbances and is suitable for identifying the instantaneous response of sudden increases or decreases in wind speed.

[0074] In order to eliminate local noise and reflect the wind speed fluctuation amplitude in a short period of time, the sliding window range analysis method is introduced to extract the wind speed range in each window: R t =max(v t-w+1 ,...,v t )-min(v t-w+1 ,...,v t ), where R t is the extreme difference in wind speed within the current window, w is the window length (e.g. 5 minutes or 10 seconds), if R t If the wind speed exceeds the set extreme difference threshold, it is considered that there is a severe wind speed fluctuation in this time period. This method is suitable for capturing severe wind speed changes in a short period and is suitable for determining whether local disturbances may affect the aerodynamic load.

[0075] In order to extract the overall trend of wind speed changes, the weighted moving average (WMA) method can be used to smooth the wind speed data series, thereby reducing noise interference and observing long-term changes. The calculation formula is as follows:

[0076] Where, is the weighted moving average wind speed at time t, ω iis the weight corresponding to the i-th time point, usually satisfying ω0>ω1>...>ω w-1 , the common weight sequence is linearly decreasing or exponentially decaying. If multiple moments When the rising or falling slope exceeds a certain threshold, it is judged that a "wind speed sudden change trend" has occurred.

[0077] Key parameters such as the time of occurrence, intensity, and direction (forward or reverse) of a sudden wind speed change are fed as inputs to the collaborative optimization module. This module uses these parameters to determine whether to initiate immediate corrections to the aerodynamic control strategy to maintain turbine operational stability. The constructed and updated wind energy distribution model is then fed to the simulation application module for CFD simulation modeling. The simulation module then establishes realistic and representative wind farm boundary conditions, supporting aerodynamic optimization and blade control strategy simulation analysis under multiple operating conditions.

[0078] The wind speed analysis module's workflow constitutes its complete operating mechanism. It not only improves the efficiency and accuracy of identifying sudden wind speed changes, but also provides precise data support for the intelligent and personalized application of subsequent control strategies, thereby enhancing the entire wind power system's adaptability and response efficiency to dynamic wind conditions.

[0079] The load analysis module introduces the load demand at the grid end to form a load change model. Based on the load change model, it analyzes whether there is load fluctuation, which includes sudden increase / drop in load. The analysis results are sent to the collaborative optimization module, and the load change model is sent to the simulation application module.

[0080] The load analysis module first obtains continuous grid load data from the grid monitoring platform or dispatch center to form the following time series: Where G t Indicates the actual grid load at the tth moment, where t is the sampling time index, usually sampled at the minute or second level. It is a complete load sequence and the basic data set for establishing the load change model, thereby ensuring the integrity and time continuity of the data to support subsequent fluctuation monitoring and trend analysis.

[0081] After obtaining the original load data, the system quantifies the grid load fluctuation by calculating the relative change amplitude of the load per unit time. The change rate calculation formula is as follows: Where ΔG t Indicates the relative rate of change of the load at time t. If |ΔG t If the rate of change is greater than a preset threshold (e.g., 10%), it is considered that a sudden load change may have occurred at that moment. The rate of change is used to identify sudden increases or decreases in load within a short period of time. This allows for rapid response and marking of "sudden change" load fluctuation events, with high real-time performance.

[0082] In order to avoid false alarms of instantaneous abnormal points, a sliding window is used to calculate the load range within a certain period of time. The calculation formula is as follows: t =max(G t-w+1 ,…,G t )-min(G t-w+1 ,…,G t ), where W t Indicates the load range with the current time as the end point of the window, w is the width of the sliding window, for example, 10 minutes, if W t If the load range change rate is greater than the load range change rate threshold, it means that the load fluctuated violently during the period and can be identified as a medium-term fluctuation. The load range is used to identify load disturbance trends in scenarios such as periodic load disturbances, industrial and commercial pull, and centralized charging of electric vehicles.

[0083] The collaborative optimization module takes sudden changes in wind speed and load fluctuations as input variables, and collaboratively optimizes the real-time adjustment strategy of the blades (such as pitch angle, angle of attack, speed, and airfoil fine-tuning) through adaptive control algorithms (such as fuzzy control algorithms), and provides real-time feedback on the aerodynamic load distribution through a distributed sensor network (strain gauges, pressure sensors, and fiber optic monitoring) deployed at key positions of the blades.

[0084] The collaborative optimization module receives the output information from the wind speed analysis module and the load analysis module, and extracts the wind speed mutation index ΔV t , which indicates the wind speed change rate or wind energy distribution difference at the current moment, and the load fluctuation index Φ t , is a Boolean value or a fluctuation intensity value, indicating whether the load is in a high fluctuation state, the current blade state information, including pitch angle, angle of attack, speed, and current wind rotor torque, etc. The above variables are combined into the lifting system input variable set X t :X t ={ΔV t ,Φ t ,β t ,α t ,ω t}, where X t is the variable set at the current time t, ΔV t is the wind speed mutation index at the current time t, Φ t is the load fluctuation index at the current time t, β t is the pitch angle at the current moment t, α t is the angle of attack at the current moment t, ω t is the speed at the current moment t. This variable set will be fed into the subsequent adaptive control algorithm to determine whether the blades need to be adjusted in real time and how to adjust them.

[0085] Taking the variable set as input variables, the fuzzy control algorithm is used to model the blade adjustment logic in the case that both wind speed and load are dynamic disturbances and cannot be accurately predicted:

[0086] Convert input variables into fuzzy linguistic variables, such as sudden increase in wind speed, high load fluctuation, small pitch, etc. to build a fuzzy rule base, for example:

[0087] If the wind speed suddenly increases and the load rises, the pitch angle is increased and the angle of attack is reduced;

[0088] If the wind speed drops suddenly and the load decreases, the pitch angle is reduced and the angle of attack is increased;

[0089] Fuzzy control algorithm (such as Mamdani type) combines actual input and outputs the corresponding control quantity U t :

[0090] U t ={Δβ t ,Δα t ,Δω t ,ΔC l}, where Δβ t is the pitch angle adjustment, Δα t is the angle of attack adjustment, Δω t is the spindle speed adjustment, ΔC l It is the correction value of the local lift coefficient of the airfoil, which is used to fine-tune the local aerodynamic performance of the blade. This control value will directly drive the servo mechanism to adjust the blade attitude and achieve dynamic optimization.

[0091] To ensure a closed-loop control algorithm and real-time adaptive capabilities, the collaborative optimization module deploys a sensor network at key structures such as the blade root, midsection, and tip. These include:

[0092] Strain gauge: monitors the local bending moment and shear force changes of the blade;

[0093] Pressure sensor array: Obtain surface airflow pressure distribution and evaluate lift and drag changes:

[0094] Fiber Bragg sensor (FBG): senses the coupled behavior of multiple physical quantities such as stress and vibration.

[0095] The data collected by the sensor will be converted into digital form and transmitted to the central controller to form a real-time aerodynamic distribution map F t :F t ={σ t (x),p t (x),M t (x)}, where σ t (x) is the stress distribution at each position x on the blade, p t (x) is the instantaneous surface pressure at each monitoring point, M t(x) is the bending moment and torque monitoring value. The real-time aerodynamic distribution diagram is used to verify whether the fuzzy control output is effective. If there is a significant error or nonlinear response distortion, the control algorithm self-calibration or parameter self-adjustment mechanism can be triggered.

[0096] Finally, the collaborative optimization module will evaluate the current control effect based on the difference between the real-time feedback value and the expected aerodynamic response: Where, e t is the current blade response error, is the ideal aerodynamic distribution under the current wind speed and load conditions (which can be preset by CFD simulation), F t is the real-time feedback value. If the current blade response error exceeds the error threshold, the fuzzy rule parameters are adjusted.

[0097] Assume that the wind turbine is currently operating in the following state: the real-time wind speed suddenly increases from 12m / s to 18m / s (a sudden wind speed increase); the grid load suddenly decreases (a load drop event);

[0098] The control system implemented a fuzzy control strategy: Rule R1: If there is a sudden increase in wind speed and a sudden drop in load, then the pitch angle should be slightly increased, the angle of attack should be reduced, and the main shaft speed should be reduced. However, real-time sensor feedback indicated an abnormal lift distribution on the blade surface.

[0099] Stress monitoring data is too high: The system calculates that the current response error is greater than the error threshold, which means that the blade has not reached the target aerodynamic state. One of the original fuzzy rules is as follows:

[0100] IF wind speed change level = "significantly increased" AND load fluctuation level = "significantly decreased";

[0101] THEN pitch angle = "slightly increase" AND angle of attack = "decrease" AND speed = "decrease";

[0102] Among them, "slightly increase", "decrease" and "downgrade" are linguistic variables, corresponding to a certain numerical range.

[0103] Identify the failure rule: The current environment meets the conditions: "significant increase in wind speed" + "significant decrease in load" → corresponding to rule R1. The system goal is to quickly reduce lift to prevent blade vibration due to excessive lift. The current control action is not aggressive enough in adjusting the angle of attack and pitch angle. It is inferred that the problem lies in the overly conservative action of "slightly increasing the pitch angle."

[0104] Dynamically adjust the language variable mapping: adjust the original control output variable "slightly increase" to "significantly increase": previously "slightly increase" corresponded to the actual numerical range [0.5°, 1.5°]; now it is changed to "significantly increase", and the corresponding range is adjusted to [2.0°, 4.0°]; update the fuzzy control rules.

[0105] Original rules:

[0106] IF wind speed = "significantly increased" AND load = "significantly decreased";

[0107] THEN pitch angle = "slightly increase" AND angle of attack = "decrease";

[0108] After update:

[0109] IF wind speed = "significantly increased" AND load = "significantly decreased";

[0110] THEN pitch angle = "significantly increase" AND angle of attack = "decrease";

[0111] You can also fine-tune the "Angle of Attack" output variable synchronously. The original "Decrease" has now been changed to "Fast Decrease", making the control response more sensitive.

[0112] The simulation application module combines the wind energy distribution model with the load change model, and uses CFD numerical simulation to generate multi-condition aerodynamic optimization solution applications.

[0113] The simulation application module first receives data transmitted from the wind speed analysis module and the load analysis module

[0114] Wind energy distribution model: including the probability distribution of wind energy density, daily occurrence frequency, and duration corresponding to different wind speed levels;

[0115] Load change model: includes the trend curve of grid load changes over time, load fluctuation frequency, sudden increase and decrease identification marks, etc.

[0116] Based on the two types of models, typical representative working conditions (such as medium and low wind speed + light load, high wind speed + heavy load, sudden transition state, etc.) are extracted for combined mapping to construct a sample set of various working condition combinations.

[0117] For each of the above working condition samples, a CFD simulation model of the wind turbine blade is constructed, including the following contents:

[0118] 3D blade geometry modeling: import actual blade structure or topology optimized structure;

[0119] Flow field boundary condition setting: inlet wind speed boundary condition, take the given wind speed in the wind energy distribution model; outlet pressure boundary; blade surface adopts no-slip boundary condition;

[0120] Load model data is introduced to convert grid load changes into dynamic inputs such as pitch angle and angle of attack for reverse thrust control. Unstructured grid technology is used to divide the blade surface and surrounding flow field into a fine boundary layer grid to accurately capture boundary layer separation, vortex wake, etc.

[0121] Under each operating condition, a CFD solver (such as Fluent or Open-FOAM) is called to perform unsteady aerodynamic simulation. The core process is as follows: set the end flow model (commonly used such as k-ωSST or LES); input the initial pitch angle, angle of attack and speed settings; execute the solution and record: the pressure distribution on the blade surface; the lift coefficient, drag coefficient, aerodynamic torque, as well as the local streamline distribution and tail vortex structure.

[0122] The simulation results are used to evaluate the aerodynamic performance of the blades under each operating condition, including the following indicators:

[0123] Calculate the wind energy conversion rate per unit area, the expression is: Where η is the wind energy conversion rate, P out is the output power calculated by simulation, ρ is the air density, Q is the swept area, and U is the inflow wind speed;

[0124] The lift-to-drag ratio is obtained by dividing the lift coefficient by the drag coefficient, which represents the aerodynamic efficiency of the blade under the current working conditions. The local load gradient is obtained, and the presence of a sudden pressure drop zone is detected. The fatigue stress distribution is obtained and possible structural damage risk areas are identified.

[0125] The purpose of evaluating the aerodynamic performance of wind turbine blades under each operating condition is to determine whether the blades can efficiently, stably and reliably convert wind energy into mechanical energy under a specific combination of wind speed and load conditions.

[0126] The closer the wind energy conversion rate per unit area is to the Betz limit (the theoretical maximum is 59.3%), the better the aerodynamic design and the higher the lift-to-drag ratio: the blades are more likely to drive the rotor to rotate at the current angle, and the energy efficiency is high; a low lift-to-drag ratio: the blades are too obstructed, indicating that the angle setting is unreasonable and the angle of attack or airfoil needs to be optimized. The local load gradient refers to the rate of change of pressure along the spatial distribution on the blade surface, reflecting whether the airflow is smooth and whether there is boundary layer separation or vortex. Gradient mutation areas are common at locations where the angle of attack is too large or the boundary layer separates; they can be used as "high-risk stress concentration" areas to prompt design adjustments; they help optimize the smooth transition area of the airfoil and suppress airflow separation.

[0127] Fatigue stress distribution results from the accumulation of microcracks in the material caused by periodic fluctuations in aerodynamic loads. In CFD+structural coupling simulation, the corresponding strain field or equivalent stress (such as VonMises stress) distribution can be calculated based on the aerodynamic loading obtained from the simulation, and "stress hotspots" can be alerted: stress concentration locations such as the blade root and blade tip can be identified. For long-term wind power systems, the more dispersed the stress, the better, and the smaller the peak value, the safer it is. This supports fatigue life prediction (such as through SN curve fitting) and structural residual life assessment.

[0128] Combining the simulation results under multiple working conditions, the expression is:

[0129]

[0130] Where S i is the performance score under working condition i, η i is the wind energy conversion rate under working condition i, is the peak pressure gradient under working condition i, max(σ fatigue,i ) is the maximum fatigue stress under working condition i, is the lift-to-drag ratio under working condition i, C L is the lift coefficient, C D is the resistance coefficient, b1, b2, b3, b4 are weight coefficients.

[0131] The larger the performance score under the current working conditions, the better the aerodynamic performance of the fan blades under the current working conditions.

[0132] Creates the following optimized output:

[0133] Working condition-adjustment strategy mapping table: wind speed level × load level → optimal pitch angle + angle of attack + speed combination;

[0134] Airfoil adjustment suggestions: local airfoil fine-tuning area, leading edge / trailing edge curvature correction suggestions;

[0135] Working condition priority list: Based on the wind energy distribution probability density function, extract the main working condition of "high-frequency wind speed + common load", and give priority to physical prototype testing or hardware solidification.

[0136] The final optimization strategy is transmitted back to the collaborative optimization module and the main control system as feedback information, including:

[0137] The control response matrix generated based on CFD data regression, visual simulation reports under various working conditions (such as lift-drag curve, airflow velocity distribution), and automatic update of boundary value recommendations for fuzzy control rules.

[0138] This simulation application module significantly transcends the limitations of traditional modeling based on a single static operating condition, achieving high-precision aerodynamic optimization under multiple wind speed and load combinations. Compared to traditional strategies, this system can improve power generation efficiency by 8% to 15%, significantly enhance operational stability under extreme wind conditions, reduce the risk of overload stress and fatigue cracks, and extend blade life.

[0139] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0140] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A system for optimizing the aerodynamic characteristics of wind turbine blades and improving power generation efficiency, characterized by: Including wind speed analysis module, load analysis module, collaborative optimization module and simulation application module; Wind speed analysis module: This module establishes wind energy distribution models for different wind speed levels based on historical wind data and real-time wind speed changes, and analyzes whether there is a sudden change in wind speed based on the wind energy distribution model. Load analysis module: Introduces grid-side load demand to form a load change model, and analyzes whether there is load fluctuation based on the load change model; Collaborative Optimization Module: This module uses wind speed changes and load fluctuations as input variables and collaboratively optimizes the real-time blade adjustment strategy through an adaptive control algorithm. Simulation application module: Combining the wind energy distribution model with the load variation model, CFD numerical simulation is used to generate multi-condition aerodynamic optimization solution applications.

2. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 1, characterized in that: The collaborative optimization module receives the output information from the wind speed analysis module and the load analysis module and extracts the wind speed mutation index ΔV t and load fluctuation index Φ t ; Input variable set X for the lifting system t :X t ={ΔV t ,Φ t ,β t ,α t ,ω t }, where X t is the variable set at the current time t, ΔV t is the wind speed mutation index at the current time t, Φ t is the load fluctuation index at the current time t, β t is the pitch angle at the current moment t, α t is the angle of attack at the current moment t, ω t is the speed at the current moment t; The variable set is used as input variables and the fuzzy control algorithm is used to model the blade adjustment logic, including: If the wind speed suddenly increases and the load rises, the pitch angle is increased and the angle of attack is reduced; If the wind speed drops suddenly and the load decreases, the pitch angle is reduced and the angle of attack is increased; The fuzzy control algorithm combines the actual input and outputs the corresponding control quantity U t ; U t ={Δβ t ,Δα t ,Δω t ,ΔC l }, where Δβ t is the pitch angle adjustment, Δα t is the angle of attack adjustment, Δω t is the spindle speed adjustment, ΔC l is the correction value of the local lift coefficient of the airfoil.

3. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 2, characterized in that: The data collected by the collaborative optimization module through the sensor will be converted into digital form and transmitted to the central controller to form a real-time aerodynamic distribution map F t : F t ={σ t (x),p t (x),M t (x)}, where σ t (x) is the stress distribution at each position x on the blade, p t (x) is the instantaneous surface pressure at each monitoring point, M t (x) is the monitoring value of bending moment and torque; The current control effect will be evaluated based on the difference between the real-time feedback value and the expected aerodynamic response. If the current blade response error exceeds the error threshold, the fuzzy rule parameters will be adjusted.

4. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 3, characterized in that: The collaborative optimization module will evaluate the current control effect based on the difference between the real-time feedback value and the expected aerodynamic response. The expression is: Where, e t is the current blade response error, is the ideal aerodynamic distribution under the current wind speed and load conditions, F t It is the real-time feedback value.

5. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 4, characterized in that: The simulation application module first receives data transmitted from the wind speed analysis module and the load analysis module, extracts the working conditions for combination mapping, and constructs a combination sample set of multiple working conditions; Based on each working condition sample set, a CFD simulation model of the wind turbine blade is constructed, including: 3D blade geometry modeling: import actual blade structure or topology optimized structure; Flow field boundary condition setting: inlet wind speed boundary condition, take the given wind speed in the wind energy distribution model; Outlet pressure boundary; no-slip boundary condition is used on the blade surface; Under each operating condition, a CFD solver is used to perform unsteady aerodynamic simulations. The simulation results are used to evaluate the aerodynamic performance of the blade under each operating condition, including the following indicators: wind energy conversion rate per unit area, lift-to-drag ratio, local load gradient, and fatigue stress distribution. By integrating the simulation results under multiple working conditions, the optimization output is established, including the working condition-adjustment strategy mapping table, airfoil adjustment suggestions, and working condition priority list.

6. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 5, characterized in that: The load analysis module obtains continuous grid load data from the grid monitoring platform or the dispatching center to form a time series; After obtaining the original load data, the relative change amplitude of the load within a unit time is calculated to quantify the load fluctuation of the power grid, and the load extreme difference within the time period is statistically analyzed using a sliding window.

7. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 6, characterized in that: The load analysis module obtains continuous grid load data from the grid monitoring platform or dispatching center to form a time series: Where G t Represents the actual grid load at the tth moment, t is the sampling time index, For the complete load sequence; The formula for calculating the rate of change is: Where ΔG t Indicates the relative rate of change of the load at time t. If |ΔG t If the rate of change is greater than the preset threshold, it is considered that a sudden load change has occurred at that moment; A sliding window is used to calculate the load range within a certain period of time. The calculation formula is as follows: t =max(G t-w+1 ,…,G t )-min(G t-w+1 ,…,G t ), where W t Indicates the load range with the current moment as the end point of the window, w is the width of the sliding window, if W t If the value is greater than the load range change rate threshold, it means that the load fluctuates violently during this period.

8. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 7, characterized in that: The wind speed analysis module extracts multi-time scale wind speed data including annual, seasonal and daily data from the wind farm historical database, continuously samples the wind speed and wind direction at the current moment through the anemometer installed on the top or leading edge of the nacelle, and synchronously maps the real-time data with the historical data on the time axis; A wind energy distribution model is constructed based on real-time wind speed data and historical wind speed data. According to the wind speed time series data of the wind energy distribution model, the first-order difference, sliding window range analysis or weighted moving average method is used to determine whether there is a significant change in the current wind speed.

9. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 8, characterized in that: The wind energy distribution model processing logic is as follows: based on historical data, the operating environment is divided into multiple wind speed levels according to the wind speed. Each wind speed level corresponds to a corresponding wind energy density range. The frequency, duration, and daily distribution of each wind speed range are statistically modeled to form a wind energy density function. Combining the wind speed grade classification with the real-time wind speed changes, the current wind energy distribution trend in each wind speed range is dynamically updated to establish a wind energy distribution model.

10. The system for optimizing aerodynamic characteristics of wind turbine blades and improving power generation efficiency according to claim 9, characterized in that: Forming the wind energy density function includes the following steps: Based on the collected historical wind speed data set V={v1,v2,...,v n }, where v i Indicates the wind speed value obtained for the i-th time, and divides the wind speed into several level intervals; After the wind speed levels are divided, the probability of occurrence of the i-th wind speed level is obtained by dividing the number of samples in the wind speed data that fall into the i-th level by the total number of wind speed data samples; For each wind speed level, the duration of the time period in a unit time period is counted, and the average duration is expressed as: Where, T i Wind speed level L i The average duration, t i,j The wind speed on the jth day is at wind speed level L i The total duration of M is the number of observation days; The functional expression of wind energy density function is: Where E(v) represents the wind energy density per unit time when the wind speed is v, p is the air density, A is the swept area of the wind turbine, and v represents the wind speed.

Images